Welcome to Stats Drop, an inundation of rugby league numbers.

If you’re receiving this by email, a reminder that the Datawrapper embeds work best on desktop, then next best if you click through on mobile and least best as presented in the email.

“2–0 lead is the most dangerous lead” is a cliche so overused it has its own Wikipedia article:

"2–0 lead is the worst lead" (sometimes phrased as "2–0 is the most dangerous lead") is a cliché used in sporting contests, to describe the situation in which one team is leading by a score of 2–0, causing them to become complacent. The phrase is most common in association football, where it is sometimes applied only to the scoreline at half-time. It is sometimes also encountered in other sports where 2–0 is a moderately large lead, such as ice hockey.

The equivalent in rugby league is the “14-nil dickheading,”1 which posits that a team that leads by the exact score 14-0 is at risk of an embarrassing comeback from the opposition. Whether the impending dickheading is considered more likely because it is specifically a 14-0 scoreline, depends on the interlocutor.

Last Friday, the Melbourne Storm gave up an absolute classic of the genre, going into the sheds up 14-0 over the Brisbane Broncos, only to watch as the defending premiers pile on the points and sap the Victorians of their will to live in the second half. That morning, on the other side of the world, York City Knights raced out to a 14-0 lead after 24 minutes before Wigan ran them down, sealing the result, 23-22, with a field goal in the 69th minute before conceding a late try to make it interesting.

We don’t have to overthink this. The funny part of a 14-0 dickheading is watching a team - preferably not your own - squander a handy lead.

Based on data from the last eleven years of the NRL, a team leading by 14 points at any point in the game wins the game 92% of the time. That means 8% of the time, and ignoring the clock, a 14 point lead can be overhauled. For the record, the biggest comeback in the dataset was 26 points, which has actually happened three times but only once since 2016 (Dolphins, 28, defeated Titans, 26, in 2023).

It is important to understand the nuance of this chart. What it is saying is that in all the games where a team was down by exactly two points for at least three minutes of game time (to skip over games where the margin was briefly a certain number between a try being scored and the conversion taken), the trailing team ended up winning the game 35% of the time.

Given this is over 2,000 games of sample size from 2016 to 2026 round 2, including a bit over 1,000 games where a two point deficit has existed for more than three minutes, we directly infer the frequency at which the event has occurred historically is effectively equal to the actual probability of the event occurring in future.

This is not quite precisely the same thing as “the odds of a team winning down by two are 35%.” The actual win probability from that position is roughly 40 to 45%, and depends on the time remaining in the game. In this analysis, a team that takes the lead with a try with five to go, misses the conversion but doesn’t relinquish the lead is treated the same as another team trailing by a conversion in the first half2, even though the win probabilities for each of those scenarios are quite different.

This chart establishes a baseline of what a 14 point lead represents in our expectations. You may not have known it was exactly an 8% chance of a comeback but if I surveyed enough of my audience, the average response would have landed close to 1-in-10. So what happens with a 14-0 scoreline?

This table needs some explanation. The top row is the leading team’s score. The first column is the trailing team’s score. Remember that this score only has to occur for a few minutes in a game to count as an example and that we infer the probability of an event ocurring based its on historical frequency. From this, you can see what the rates of successful comebacks have been from any combination of even numbered scores up to 20.

To find 14-0, we go across the top to 14 and down one row to 0, to find the chance of a successful 14-0 dickheading is 13%. This is noticeably higher than the baseline 8% we got for any 14 point lead. The latter rate is lower, which reflects a greater number of games where the winning team increases their margin to 14 points near the end of the game, a position that is effectively impossible to overcome, and other games where the margin is 14 for a while, on the way to a bigger beatdown.

As we descend down the 14 column, the margin narrows and the probability of the combeback succeeding improves. This makes sense. Unless, that is, a team finds itself in a situation where it is down 14-2. There is a range of scenarios where a team might take a penalty goal and then concede 14 points, or concede 14 points and then take the shot at goal or some combination of. They all represent a degree of cowardice, lack of self-belief or incompetence. In fact, if the team leading has a double digit score, the lone penalty goal reduces the likelihood of the comeback compared to the situation where the trailing team had scored no points at all.

This is not information on which you could base in-game tactical decisions - the team taking the penalty doesn’t know if it is going to score more points later, and thus move into a different box and improve its probabilities, or if the leading team will make it to double digits - other than acknowledging that two point is, indeed, poo points.3

While the rows and columns form scores, the diagonals show the probabilities of overcoming particular margins. For example, the odds of overcoming a two point deficit range from 49% (4-2) to 32% (20-18). As we move to the bottom right, it is worth remembering that situations in which nearly 40 points have been scored are necessarily much later in the game, certainly in the second half of the match, than a game in which six points have been scored. The clock becomes a more significant factor as to whether the comeback will succeed, as the total score increases.

This chart is the same information as the table but displayed differently. Each curve represents the top row of the table, or the leading team’s score. The x-axis is the same as the first column. Where the x-value intersects the line gives the probability on the y-axis.

Despite the complexity, this helps visualise where probabilities stand out from the trend. I’ve marked 8-4, 16-6 and 14-0 as special scores. A team trailing 14-0 is as likely as a team trailing 14-4 to stage a comeback, acknowledging that 14-4 is likely on the route from 14-0 to victory. If the team leading has 16 points on the board, the best bet for the comeback is to be on six or 12. And being down 8-4 is seemingly better than 8-6, partly because the trailing team is incentivised to score a try instead of playing for a goal and partly because 8-6 is a late game coded scoreline, while 8-4 is a necessary progression to any number of much higher scoring outcomes.

There’s no ideal score to be down. From a win probability perspective, it is always preferable to lead the game or, if trailing, minimise the margin as much as possible. There are more successful pathways to victory than others - going from 14-0 to 14-4 is generally more succesful than going from 14-0 to 14-2 - and these involving scoring tries, rather than kicking goals. All of this is unsurprising in the least because funnily enough:

There is little evidence that 2–0 is the worst lead in practice. In association football, a team leading 2–0 at half-time only goes on to lose the game in about 2% of cases. A 2017 analysis by Sky Sports showed that between August 1992 and July 2017, there had been 2,766 occasions in the English Premier League where a team established two-goal lead. Of these matches, 2,481 resulted in victories for the leading team, 212 ended in draws, and 73 in defeats. In ice hockey, statistics show that if a team builds a two-goal advantage, they go on to win the game in the majority of instances, and that a one-goal lead is far more dangerous. As a result, the cliché is often used in full knowledge that 2–0 is not in fact the worst possible lead.

A 14-0 dickheading is fun and memorable but not necessarily anything special statistically.

Subscribe, share and upgrade

Unfortunately, I must briefly interrupt the flow of charts to use some words to ask you to subscribe to the newsletter.

Word of mouth from you is the easiest way for Stats Drop to find new subscribers and new subscribers keep Stats Drop in print.

Share

Upgrading to a paid subscription is both deeply appreciated and will get you access to what’s below this line, which is discussion of Minutes-Points-Led (MPL) and Wood-Chess-Lathe (WCL).

UPGRADE TO PAID

Thank you for your support.

Read more

Welcome to Stats Drop, an inundation of rugby league numbers.

If you’re receiving this by email, a reminder that the Datawrapper embeds work best on desktop, then next best if you click through on mobile and least best as presented in the email.

Churn

What does the average NRL career look like? We often focus on the stars, who seem to play forever, or the very bad and unqualified, whose time in the NRL is brief and somewhat funny, but there is an entire middle class of players whose careers go unremarked upon. There are as many solid, workmanlike journeymen as there are candles in the wind, brief blips of luminiscence before being swallowed by a vast and uncaring universe.

Let’s look at all the players who made at least one regular season appearance (excluding replacements) between 2000 and 2024, excluding players who made appearances in 1999 or 2025. This approach captures players whose entire NRL careers have been played in the first quarter of the 21st century.1

It is not perfect, as a handful of players may have just happened to skip 1999 or 2025 but this criteria excludes players whose careers predate the millennium or are continuing that would skew the numbers somewhat and, as noted, the data is a little wobbly as is.

In the 16 team era of the NRL (2007 - 2022), an average of around 470 players were used in a given season. Since 2023, with the addition of the Dolphins and other changes to NRL rostering rules, this has an increased to between 500 and 510.

At no point in the period from 2000 to 2024 do the 1450 playing careers of interest make up the entirety of the NRL. They represent 90% or more of the players between 2008 and 2016, with the balance represented by older players whose time precedes 2000 and younger players whose career is still going.

There are 1450 playing careers that meet this criteria. Based on this sample, the mean NRL career in the first quarter of the 21st century is 68.3 regular season games long, played over 5.2 seasons for 13.1 apperances per season.

The median NRL career is only 37 regular season games, spaced over just four seasons for an average appearance rate of 9.3 games per season.

The discrepancy between the median and mean speaks to the lopsided distribution of games played. A handful of players compile lengthy resumes that skew the mean upwards relative to the median.

The modal NRL career is a single game, representing 119 players, a similar number to those who have at least 200 appearances in this data (128).

Looking at the whole population, we see the retention rate of players has been consistent. From 2000 to 2025, on average in any given season:

• 79% of players had played the season before

• 67% of players had played two years before

• 57% played three years before

• 40% played five years earlier

• 26% played seven years earlier

• 12% played 10 years earlier

• <1% played 15 years earlier

These numbers fluctuate from year-to-year but are consistent over the long run. While there is some correlation between the retention rate and time, the slope is effectively zero. You may have expected players having long careers, particularly with the likes of Cam Smith, Ben Hunt and Daly Cherry-Evans, reducing churn but that doesn’t seem to be the case for vast majority of the playing group. Those remain exceptions against the well established trend of exponential decay (λ ≈ -0.2).

But what of quality? Rather than use a volume metric, like WARG, we can use a rate metric, like Z score, to avoid directly correlating quantity on quantity.

From here, we learn two things. While Z score is designed to average out to 100, the actual average is very clearly in the range of 65 to 70. I did not realise (or had forgotten) this about my own handcrafted metric but think it has something to do with the arbitrary cap of +250, which stops occasional games measured in the thousands from skewing averages. I could fix this by inflating the score by 30% across the board but will instead leave it as a monument to my own stupidity.

The second thing we learn is that a lot of long run analytics are bailed out by the very people that analytics fans like to criticise. Good players get long careers. Bad players get weeded out. These decisions are not made by statisticians but by coaches and administrators. Long careers allow for the accumulation of stats and make for good justifications of those metrics in the first place.

There is an upward, weakly correlated trend. Other than the known deficiencies in the principles of production, this may be because Z score lacks resolution (it is meant to be simple) and every career looks largely like everyone else’s at this scale. It is also a function of the fact that there are 500 players churned through first grade every year and at some point, you need to use the bodies you have available. Not all of them are going to be stars but they might still get game time.

Switching to a seasons played, rather than games played, makes this selection effect clearer. The more seasons given, the floor seems to improve while the ceiling is relatively flat. The first season is full of one-hit wonders and outliers but by season four or five, we’re dealing with established professionals and the squibs, the wusses and the dumdums have been rejected earlier in the quality control process.

This line of inquiry originally began as a question about the NRLW. There are players who played Origin or played for the Broncos in the early premiership years whose careers did not last very long. One assumes this is because being a semi-professional rugby league player is a difficult ask at the best of times, and without the infrastructure that is still being put in place to support female players, playing NRLW at a mid-level is incompatible with some career or family paths. Other careers were victims of covid or injury. Today, the NRLW seems dominated by fresh faced 18 to 21 year old girls that don’t have established careers or dependents to consider and can live off $60,000 a year.

About 80 players were used each year in the four team era (2018 - 2020), 140 with six teams (2021 - 2022), 240 with ten (2023 - 2024) and last season’s 12 teams used 307 players, which is a phenomenal rate of growth over just five years. We’ve established the typical retention rates in the NRLM, which we will consider “normal”. Here’s what the NRLW’s churn looks like:

The average retention rates for the NRLW work out to:

• 58% of players had played the season before (78% in the Big M)

• 41% of players had played two years before (67%)

• 30% played three years before (57%)

• 15% played five years earlier (40%)

Retention is much lower in the NRLW than the NRLM, however, retention is improving. For players in the first three categories, the rate is improving on the order of 2 to 3% per season, so perhaps in another six years or so, the NRLW will reach parity with the men’s competition in this aspect.

For players in the five year category, the rate is flat. Perhaps this is because t=3, but I think this speaks to a top echelon of players - your Brigginshaws, Kellys, Sergii, Uptons, et al - who have made something resembling a career out of being star players and they have managed to make it work. The class of player beneath them who do not have this advantage or commitment or opportunity turns over at a noticeably higher rate for reasons already discussed. The NRL have yet to create an environment in which the journeywoman can exist but it is coming.

Subscribe, share and upgrade

Unfortunately, I must briefly interrupt the flow of charts to use some words to ask you to subscribe to the newsletter.

If you like this level of in-depth statistical analysis of rugby league, and don’t want to be treated like a moron-nerd (Fox manages this with alarming regularity), then the only way to get all future Stats Drops in your inbox is to subscribe.

Word of mouth from you is the easiest way for Stats Drop to find new subscribers and new subscribers keep Stats Drop in print.

You can forward this email to anyone you might think is interested or use the button below.

Share

Upgrading to a paid subscription would be deeply appreciated if you can spare the cash in this time of consumer spending crisis. If not, the first part of Stats Drop, which is frankly the most interesting, NRL-related part, will always be free.

UPGRADE TO PAID

If you prefer, you can use Ko-Fi and get a shout-out in the next newsletter.

This is where the paywall would normally be (actually, it would probably be above the bit where we started talking about the NRLW) but the first and last issues of Stats Drop of the year are free.

Also check out The Almanac, in which I am building up a repository of stats information.

Upside

Trying to predict what a team will do next season based on the last season is a crapshoot. There’s a weak but positive relationship between wins this year and wins next year.

During my very lightweight season preview, I noted that the last four years of the Raiders had basically been the luckiest team in the history of the NRL. Of the 460-odd team seasons played of NRL, three of the top 20 outperformances of Pythagorean expectation are the ‘23 (4th, +3.8), ‘24 (18th, +2.8) and ‘25 (2nd +4.1) Raiders. While teams can escape their fate for a while, the arc of history is long but it always bends back towards Pythagoras.

A less poetic way of describing this is “mean regression”. Pythagorean expectation began in baseball as a way of estimating a team’s “true” winning percentage, as calculated by their runs for and against. This principle works equally well in rugby league - hence Pythago NRL - and is an easy way to see which teams’ win-loss record is reflective of their play on the field or is subject to the vagaries of assigning a binary win-loss outcome to games. Substantial discrepancies can be chalked up to good luck, misfortune or whatever similar verbiage is preferred by the writer. Put another way, teams that win consistently by a lot are expected to be better than teams whose wins are more marginal.

While the relationship between Pythagorean expectation, that is points difference, and actual results on the field is expectedly tight, it is not perfectly aligned. If we assign the discrepancies to luck, what happens to the team’s performance when luck turns?

62% of teams experience mean reversion as we would expect: being lucky in one season, leads to a reduced winning percentage in the next (and vice versa). The remainders buck the trend, by having good luck and then getting better or vice versa. Even so, only 18% of the change in winning percentage from season to season is explained by this effect. There are other, much more obvious factors (coaching and player performance and changes, injuries, scheduling, response to late, surprising rule changes and so on) that have greater impacts on year-to-year changes in performance. Nonetheless, the inverse relationship between performance relative to Pythagorean expectation and change in winning percentage is very real.

Luck is not the only emotional ephemera we can quantitatively consider. There is also Disappointment. Based on the pre-season class Elo rating of the team, we calculate an expected number of wins against the league average as a gauge of what we can reasonably expect out of the team in the season to come. This sets the Disappointment Line and the team needs to win more games than this to avoid a disappointing season.

We don’t expect the Disappointment Line to deliver a high R-squared. The intent is to identify which teams have better seasons than a reasonable pre-season expectation (and vice versa). As expected, about 50% of teams outperform their Disappointment Line. A third exceed the Line by more than 1.5 wins and about 15% do so by more than four. Does performance relative to the Disappointment Line have a mean regressive effect on the next season’s winning percentage?

65% of teams experience a mean reversion effect relative to the performance against the Disappointment Line. Better seasons tend to follow disappointing seasons, and vice versa. The effect, while still mild, is actually more impactful than Pythagorean expectation, explaining 28% of the season to season change.

The interesting thing is we have two mostly independent ways of calculating Luck and Disappointment, both of which show weak but real mean reversion effects on the next season’s performance.

Let’s not do anything too complex. What happens if we just add them together?

Introducing upside. Calculating upside is easy. You add the numbers of wins over/under the Disappointment Line to the number of wins over/under that calculated by Pythagorean expectation. From there, we compare that to the delta in winning percentage for the next season.

While the correlation remains weak, upside has a stronger correlation than either disappointment or luck alone. Further, it gives us a trend. All other things being equal, one unit of Upside (that is, one additional win over Pythagorean expectation and/or the Disappointment line) will cost half a win next season.

In 2023, the 14-9-1 Knights had an upside of 5.0 and the inaugural 9-15 Dolphins had an upside of -3.8. In 2024, the Knights won 2.5 games fewer (half of 5.0), finishing 12-12, while the Dolphins went 11-13, two wins (or half of 3.8) better. These are cherry picked, recent examples that happen to land on the trend line, and there are good rationales for the delta in wins that are not a straightforward statistical rectification, but they are useful for demonstrating the point.

For 2026:

Recall that this is based on the ideal that a team has a “true” level. The team can make changes to improve this “true” level. If the 2025 Raiders were a 15 win team but won 19 games, and the 2026 Raiders are a 16 win team because Ricky Stuart finally got his selections sorted out, a host of young players each get a bit better and they didn’t overthink finals scheduling, then the predicted 13.5 wins becomes 14.5 wins.

It bears repeating that this ladder shows a mean regression effect that explains only 30% of the change in winning percentage from season to season. The other 70% also has a casting vote. Deciding what proportion of that 70% residual is applicable to changes in rosters, development in players, alterations in tactics or strategy or other nuances is an exercise for another time.

Projections

Here’s every player who had a decent run in 2025 and what we expect from them in 2026:

Z score is a rate metric and is not like the actual statistical Z score, although that’s where my mind started when trying to overhaul the Stats Drop system of metrics so that’s why it has that name.

Z score, in this context, measures volume of production per game against a benchmark of an average player at that position, given the minutes played. An average player at that position is given a score of 100 (as noted above, the actual average is more like 70). Individual game Z scores are capped at +250 and -250. A minimum number of games is required to qualify for a season Z score.

Where a player has a qualifying season Z score, we can use that to project their performance in the next season using linear regression to identify a trend line. The result is invariably regression towards the mean. Where available, we will use a weighted average of the last three season of qualifying Z scores, with the most recent season given three times the weight of the season two years prior, to get a more accurate projection. A single season Z score is 50% regressed to the mean but the average of three seasons is only regressed 25% to the mean.

Projections have a band of accuracy of 30 units or so, which is fairly wide considering the gap between 100 and 70 is the difference between star and middle of the field, but predicting the performance of individual players in a given year without any information about that year is always going to be something of a dice roll. The gap between projected and actual Z score is measured in Baxes, which over the course of a season and looking across the performance of a whole squad, I ascribe to coaching.

Nonetheless, star players regularly outperform their projection because they are not the ones who regress to mean. For the vast majority of players, seasons well above or below their typical performance are a rare aberration.

Welcome to Stats Drop, an inundation of rugby league numbers.

If you’re receiving this by email, a reminder that the Datawrapper embeds work best on desktop, then next best if you click through on mobile and least best as presented in the email.

As this is a long one, so you may want to open this in a browser anyway.

One of the basic premises of being a sports fan is that you want to watch your team play but fans don’t necessarily have the capacity to attend every single game. We all, after all, have got other stuff on.

In lieu of being there, it is much more convenient and much more cost effective to watch a broadcast of the game on television. In the world of the NRL, it is considered so preferable that many hundreds of thousands choose this as their way of engaging with the sport, compared to the few tens of thousands that attend games in person.

The attention of those hundreds of thousands, if not millions, of people is valuable. The broadcaster can judiciously insert advertisements during the game that will grab the audience’s attention and help sell the advertisers’ product. The broadcasters use the promise of that revenue to buy the exclusive rights to show the game from the NRL, locking in a loyal, attentive audience for the sponsors.

This revenue stream is extremely important to the NRL. To take the example of the Brisbane Broncos, News reported that the Broncos clear $11.9 million on gameday revenue. It costs the Broncos nearly $10 million to make that revenue, without accounting for any of the footballing expenses. Game day makes some, but not much, money and it is far less for teams that aren’t averaging 41,000 at Suncorp every home game.

By comparison, the NRL pays the Broncos, and every other team, a central distribution of approximately $15 million per year out of the broadcast deal. That distribution pays for all of the players and most of the coaching staff, the basic human constituents of what people consider “their team”.

This simple logic gets lost when discussing how the NRL is structured. People seem to lose track of the direct link between fans and the clubs when the broadcasters mediate that connection. It becomes subsumed by the notion of money and greed. Really, the broadcast deal is the medium of exchange from fan interest to player salaries with the networks clipping the ticket along the way.

The idea that the NRL should be built exclusively around the needs of 15,000 people at Brookvale or 12,000 at Shark Park seems quaint when upwards of half a million people are watching at home. Viewers’ needs are rarely entered into the discussion, even though their economic contribution greatly outweighs that of the ticketholders.

If you are the kind of person who reads this newsletter, then it is likely that you would like to be an informed fan. Understanding the mechanisms of the broadcast deal helps clarify administrative decisions that would otherwise be unintelligible. While it is not strictly necessary that you care about the broadcast deal or TV ratings, if you want to know what’s up, then having that understanding will help greatly.

I personally find this stuff fascinating. Back in 2020, as the pandemic completely cleared out my schedule, I wrote Everything you ever wanted to know about NRL TV ratings but were afraid to ask, in which I delved deeply into TV ratings data. It ended up being one of the most popular pieces on PythagoNRL.com and I probably should have charged the NRL tens of thousands to do it as a consultancy, instead of publishing it for free on the internet where anyone could read it.

I had always wanted to go back and revisit and retest some of the conclusions reached with more data. It’s been five and a half years, so like a chump, here we are yet again, publishing valuable research for free.

Save me from myself and thank you for reading Stats Drop

If this seems like the kind of thing you would like to hear about next year, then I can only recommend that you subscribe to get future posts delivered straight to your inbox:

Subscribe now

If you’re already subscribed, then you could do me a huge favour and share this post:

Share

If you find yourself committed to the ideology of The Maroon Observer, then you can support the publication financially through one of two ways.

Firstly, it is the offseason and there’s not much coming until 2026, so I’ve knocked 20% off annual subscriptions until New Year’s Day as incentive to sign up. You’ll get access behind the paywall for 12 months and the ability to comment. If you thought a paid subscription was too pricey, now’s the time to re-examine your priors:

Offseason special

Otherwise, if you’re not in the market for a recurring subscription, you can make a one-off payment at Ko-fi, where I have been publishing deleted scenes from this year’s newsletters on the basis that they were too crazy to send to actual subscribers. For just a single fiver, you too can plumb the depths of my 80%-complete essays about why Latrell Mitchell sucks, people overreacting to players resting and at least two more that are scheduled whose contents I have not reviewed in detail.

Before we return to the grossly underpriced consulting work, Jason at Footy/Sports Industry AU has kindly furnished me with NRL ratings data. He runs a website that I follow to keep up with media data and a Bluesky account. I hadn’t even planned on doing this until he sent me the numbers. If you think he’s a shill for a given code, you’re a dingus and should instead be thanking him.

Sports on TV isn’t dead, it’s just resting

In 2020, I wrote that “its timely to do this analysis now before that becomes a really significant, and completely opaque, component of how fans engage with the sport.” If you’d asked me then, I would’ve guessed that Nine and Fox would’ve gone into bankruptcy long before 2026 and certainly wouldn’t be preparing tenders for the 2028 to 2032 cycle. Instead of streaming obliterating the past, good old fashioned television has shown a great deal of resilience.

While the methodologies that underpin TV ratings numbers have their flaws, there are no perfect statistics. There are also no better ways of measuring the size of a club’s fanbase. You could consult Roy Morgan’s maligned fan survey, you could look at attendances or memberships, but for mine, nothing beats TV ratings. Everyone has access to a TV and can make the choice to turn on the game. Where the ratings go, so do the dollars that allow the sport to exist.1

This post is going to look primarily at free-to-air ratings numbers2. While FTA has been resilient, it hasn’t been impervious to changes in technology. The difference is that FTA rating numbers have been reported publicly, for a long time, using an agreed-upon methodology to allow the networks to sell ad space. While that methodology changed significantly in 2024, in a way that conveniently paints a surprising picture of robustness of demand for free-to-air televised content, it is intuitive what one viewer represents and extrapolate from there.

Pay TV numbers stopped being made publicly available at the end of 2023 and while it was advantageous for the previous analysis to have a number assigned to every single match, with nine years of data to work with, the stickiness of subscriptions and smaller size of the audience for cable outweighs the need to benchmark every game.

Before you ask, Kayo is impenetrable. 400,000 people subscribed to Kayo in 2020 and that figure is closer to 1.5 million now. Unsurprisingly, Foxtel has not made much effort to report viewing statistics, not least because Hubbl cannot seem to get a decent ad platform for Kayo together, but press releases indicate the Kayo audience was about 80% on top of the traditional cable rating. Take this with a serving suggestion of salt. After the sale to DAZN earlier in the year, it all ceases to be their problem. I don’t expect transparency to improve. We also have no idea what’s going on in New Zealand.

This chart paints a slightly different picture to the more narrowly scoped one published in 2020. While the Broncos and Storm remain atop the standings, the Panthers have risen and the Bulldogs managed to pull themselves off the canvas, while the Cowboys and Sharks have slid backwards into the midfield.

Notable is the Broncos’ ongoing dominance of this metric, despite enduring the worst period of the club’s history. One common complaint, especially in the weeks after the draw’s release, is how many primetime slots the Broncos are given. Of course, their ratings are the highest because they’re always playing Friday nights. That is certainly valid to an extent.

Perhaps the biggest broadcast change arising from the introduction of the Dolphins is the reduction of Broncos games on free-to-air television. Pre-Redcliffe, the Broncos were allowed up to 16 regular season appearances on FTA, while all other clubs were capped at 12. With the Dolphins acting as the Aldi Broncos, Brisbane’s cap has been brought down in line with the other teams. As you can see from the data, these are not hard and fast caps, so much as beginning points of negotiation.

In the 20s, the NRL has tried to spread more games around more teams. This Broncos primetime fixation is clearly a bugbear of executives and commentariat alike. As a result, Brisbane will play their first ever Friday 6pm game in 2026 against the Raiders.

The simple averages don’t convey a good sense of what these ratings would be all others things being equal.

What gets people to watch?

The reasons that people decide to watch a NRL game fall into three main buckets:

• They want to watch their favourite team. I assume this is the primary motivator for a plurality of the audience. We will estimate a relative figure for this later.

• They want to watch rugby league football and the NRL is the pre-eminent competition in this sport. We call this the “sicko demographic”, who will watch whatever is on when it is available.

• There are a series of smaller factors that influence marginal decision making as to whether to turn the game on. This is the casual factor.

As you will know from your own experience, decision making isn’t a single thread alone but is weaved from a fabric whose threads change with time and mood. Even the biggest sycophant Knights fan will miss some games because they are beyond hope and also they have to go to a wedding in Dubbo. The worst pervert you know won’t be able to watch eight NRL games live every weekend for over six months.

Let’s unpack some of the elements of the casual factor.

Other than the extremely popular season kick-off in Vegas, Thursday and Friday nights are by far the most widely viewed timeslots in the aggregate. Even within the pay TV data, we see that night time games get better ratings than day time ones.

Fans complain about timeslots assigned their teams or how the schedules games in general. This is called an expressed preference. Fans will watch games in the primetime slots on Thursday and Friday in far greater numbers than the weekend day slots, giving an approximately 20% premium to games played in weekday primetime. This is called a revealed preference. Revealed preferences are far more important than expressed preferences, because they represent what people actually do, rather than what people will say they think they will do.

It is little surprise that people tune out as the season progresses. The novelty of a new season and hopes for change are, for a non-trivial part of the NRL, smashed on the rocks of reality every single year. Most rational people don’t need to see out a 8-16 season in every excruciating detail. It is enough to know that the team is bad, it is not getting better and to find something else to do until those hopes rise again next February.

We can use form Elo ratings as a proxy for the perceived quality of the match. The gap between the two teams’ pre-match form Elo rating gives a good indication of how well matched two teams are in a given fixture. Unsurprisingly, games with better matched teams rate (left hand side) rate better than games with less well matched teams.

Two poor teams may be evenly matched but that does not necessarily make for a compelling contest. Instead, let’s consider the average form Elo rating of the two teams. Two in-form teams will have higher ratings (right hand side). Conversely, two lower rated teams should be actively repellant to the average fan. When we group games of similar quality by this metric, we see a very clear linear relationship between the average form rating of the match and the TV rating. Good teams draw more viewers.

How big are these clubs?

Or, is this just an exercise in Broncos supremacy? And the answer is “mostly, yes”.3

If we account for these four casual factors, as well as the structural changes in how people watch TV (i.e. lower ratings in general now than in 2017), then what’s left should be reflective of the size of the fanbase, as measured by people who watch that team on TV, and by extension, the contribution made to the overall broadcast deal.

For example, the Sea Eagles hosted the Roosters in round 19 of the 2018 season on Sunday afternoon. The Roosters won 56-24, going into the half up 34-6. That game rated 515,000 people on Nine, which is

• 15.4% less than the average game in 2018

• 7.5% more than the average Sunday afternoon game

• 8.6% less than the average round 19 game

• 3.8% less than the average for two teams of that combined form Elo rating

• 5.2% less than the average for two teams of that gap

This averages out to 5.1% less than the average game, all other things being equal. While you could weight these factors, derived from multivariable regression, and you would arrive at marginally different conclusions, the main findings are inescapable.

Obviously, the Broncos having nearly 20 games more than the next team skews this somewhat, so we’ll adjust that by looking at a per game average.

In the previous analysis, I noted that the Warriors rated OK for a team that had three (3) free-to-air games in three years because Nine wants to show as little of the New Zealand team as possible. I thought this might underestimate the size of the Kiwi ex-pat community’s attachment to the club but with the larger dataset, the Warriors have the lowest average FTA audience and, on a per game basis, they have the worst average value-add. Nine were right but the Warriors’ presence greatly improves a New Zealand TV deal that brings in more than their distribution means that all of this is irrelevant to their prospects.

It also means that the Warriors are the ideal replacement level team to benchmark per game value-add. As a team without an especially long or illustrious history and no specific connection to Australia, the Wahs’ Australian audience overwhelmingly consists of people watching the other team or watching NRL for its own sake. On that basis, I’ve shown the above average value-add relative to that of the Warriors (-15.3% unadjusted)4.

The league breaks itself relatively neatly into tiers:

1. Broncos

2. Storm

3. “Big” Sydney clubs, Queensland clubs

4. Small markets

5. Newcastle

This may not be a revelatory result but it is worth confirming.

Are the Tigers a sleeping giant?

That Wests could be a commercial powerhouse in waiting was one of the lesser conclusions reached in 2020. Anecdotally, there certainly seems to be a lot of Tigers fans on the internet. Until this year, the 2005 grand final was the highest rating NRL match of all time. Wests seemed to outperform their expected ratings, given their generally low quality team and less-than-optimal scheduling.

However, the Tigers have been crap for at least the last 15 years. Not being able to make finals in a decade and a half in a competition that sends half its teams to the post-season is about as damning a sporting indictment as possible to imagine. You might then conclude that the Tigers’ relatively low value-add is a result of that.

While there is a relationship between winning and better TV ratings, it is not a strong one. The Broncos were basically 12-12 over this entire period and still rated far higher than the rest of the league. The Tigers, and the Titans, are about where we would expect them to be given their performance.

There is still a trend though. The Storm’s value-add has improved markedly relative to the rest of the league through the 2020s. This may be partly due to changes in calculation methods, both for ratings and in this analysis, but sustained competitiveness would also be a factor in their rise to the commercial apex of the league.

At the resolution of single seasons, the trend is practically random across the league and within individual clubs. One reason is related to the quality of data but another is the vagaries of TV ratings. The Panthers’ prime years of regular season success (2020-22) came while NRL-wide ratings were at their lowest ebb in the dataset. Linearly regressing the Panthers’ season ratings against their winning percentage yields a negative gradient, which implies the more the Panthers win, the fewer people watch. That doesn’t make a lot of sense given what we know.

Similarly, there’s never really been a good Gold Coast team. We have simply no way of reliably predicting what would happen to their TV ratings if the Titans were sustainably and regularly good. A few consecutive seasons in the top six might see ratings not budge after decades of ineptitude has wasted any sense of goodwill in the nation’s sixth largest city. Conversely, had that success come earlier, we might see a different result. Alternatively, the Gold Coast might be the nation’s pre-eminent bandwagon sports city. We just don’t know.

Can the Tigers become a commercial powerhouse? Sure, anything is possible.

But probably not.

The big games

While the regular season is the bread and butter inventory for the NRL, where the real money comes in from the influx of marginal fans who tune in for the big games: finals and State of Origin.

Even before we got to the grand final, 2025’s finals series was astonishingly well received. A rational person would likely look at changes to ratings calculation methods in 2024 and then layer on top of it the biggest club in the sport putting together three successive incredibly unlikely, unbelievably thrilling comebacks and draw the conclusion that was probably a one-off. The administration is likely to conclude that Sunday afternoons are a good time for finals footy and do more of that. Thanks Ricky, you’re only going to disappoint V’Landys.

I was unable to find a suitable method for teasing apart what impact the participating teams had on the final rating. The rating seems to reflect the teams involved, another of general cultural fugazi as to how people felt about the NRL that year, as well as our old friends - the slow and sad decline of free-to-air television and monoculture.

The inexplicable plateau seems to be largely from a mass influx of regional viewers. Quite why the combination of teams in 2012 through 2014 inspired such devotion from our country friends is unclear. The long run average share of audience is about one-third but regional viewers seem to have spiked to 43% in 2013.

While I was inclined to chalk this up to data or measurement issues, there does seem to have been a boomlet in the early 2010s. League wide attendances dropped from a high water mark of 16,423 in 2012 to 15,940 in 2013 and then again from 15,905 in 2014 to 15,074 in 2015, before a long glide into covid had people wondering about the future of the sport. Fortunately, the Dolphins arrested that slide and now the NRL has had back-to-back years over 20,000.

Whether that boomlet was caused by the first part of Queensland’s 15-in-20 dynasty, I’ll leave to the reader to judge.

Code wars

Doing an apples-for-apples comparison of TV ratings between sports is challenging. The games and seasons are different lengths. The NRL has a slate of high rating representative matches that the AFL does not. Sponsors like the AFL’s solid national footprint. It is then hard to render the numbers into something sensible.

The best we can do over the long run is compare grand final TV ratings as our yardstick events that stop the nation. Both the AFL and NRL have had their grand finals lodged in the top ten, if not the top five, rating TV events every year for at least the last decade and a half.

Interestingly, the ratings recorded over the long run by enthusiasts have typically been the 5 city metro. I can’t help but notice the AFL trumpeting its total audience this year with no mention of its metro component. Is it because 2025 is the year that the NRL finally outrated those bums? Maybe, although this doesn’t do a good job of hiding it, seeing as the NRL’s total viewership still blew the AFL out of the water with the highest rated grand final of all time.5 This year’s total AFL rating was similar to last year’s, so it would follow that the 5 city metro rating is similar.

The second game of rugby union’s 2025 Bledisloe Cup test rated 893,000 viewers on Nine. That’s the best result for a union test since they switched to Nine during covid and follows on from a Lions tour, which should give the sport some momentum. It is also less than any game of the last two women’s State of Origin series or this year’s women’s grand final, which were all around the million mark.

Last year’s men’s Pacific Championships final between Australia and Tonga, in the sport where “no one” cares about the international game, rated 630,000, which was fractionally higher than this year’s cracking final between New Zealand and Samoa (615,000) in the absence of Australia. Anyone who tries to tell you that international rugby union is still culturally significant in this country even though the club game is a total irrelevance, you now know that’s cope because the best they can do barely out-rated the most boring match I’ve ever attended (possible exception for Knights-Dolphins).

I don’t have a huge dataset of men’s internationals ratings. Of the highlights, the ANZAC Test in 2017 rated 1.3 million, as did the opening and final matches of the World Cup, while an end of season 2019 trans-Tasman test brought in 888,000.

Much of the more recent data is relatively pedestrian. The international game is a long way from threatening the sport’s sporting and commercial peaks.

These are far from astronomical numbers and largely wouldn’t be competitive with most of the NRLM season. Comparing to the pre-covid ratings, it is demonstrable that this facet of the sport has taken a step back in the public’s imagination.

Potentially, the international game’s issues are less a lack of interest - the Kangaroos, and increasingly the Jillaroos, can do numbers when given the space - but a lack of commitment to a consistent structure and a lack of drawcard nations outside of Australia. When there’s at best half a dozen nations that might pull a decent audience, replacing a full round of eight NRL games for a representative weekend becomes a complete commercial non-starter for the broadcasters. The NRL is pushing on with the Pacific Championships, with some help from the federal government, which I think will pay dividends down the track.

If you made it this far…

…and didn’t decide to take me up on my very generous offers to deliver future posts like this for free to your inbox, there’s still time to subscribe:

Subscribe now

If you think your friends, colleagues and family will enjoy this, you can forward the email or use the conveniently provided share button below:

Share

And if you really enjoyed this piece, consider getting a paid subscription to support The Maroon Observer. Remember its 20% off annual subscriptions until the end of the year:

Offseason special

The next deal

Presuming that something resembling these numbers carry over - it is unlikely that the 4.4 million for the grand final will be broken without the Broncos but a suitably compelling matchup could still deliver more than four - then Peter V’Landys finds himself in an excellent position to commence negotiations for the next broadcast cycle.

The NRL TV deal will expire at the end of the 2027 season. You’ll recognise this from various thought balloons that have been sent up via the media: shorter club season, wild cards, questions as to how to fit in both men’s and women’s expansion.

V’Landys clearly wants a big number. A big number will secure his position in the short term by satiating the sharks that briefly raised their noses when it became apparent just how badly he handled the covid negotiations. A big number will also secure his legacy into the long term as a visionary administrator of the sport (i.e. delivers lots of cash to the clubs so they shut up).

The current deal is something like $2 billion for five years, although the exact terms and conditions vary from report to report and is not ever going to be revealed transparently. The expectation is that the next deal will be a headline of $3 billion. A 50% uplift, in a market where rugby union got 40% earlier in the year and given the wellspring of inflation since the deal was last negotiated, should be the minimum expectation.

If the NRL can show that its inventory, headlined by a 4 million viewer grand final, three 3 million men’s Origin games and a slew of 1 million viewer games, including women’s Origin, men’s finals, women’s grand final, some of the bigger derby games and potentially internationals, backed by 200-ish men’s regular season games plus whatever they assemble for the women6 with a newly achieved national footprint and international status, is sustained and sustainable, would $3.5 billion be out of the question? $4 billion?

Quite what that would do to the brains of the geniuses that run the NRL clubs is unclear. They already seem drunk on their own importance off a central distribution of $15 million. Bumping that up to $25+ million, with a corresponding salary cap, might either break brains or force in a new layer of talent that previously considered the NRL beneath their pay grade.

If the TV deal does double, and the salary cap with it, it kind of puts into perspective how unlikely the NRL is to be threatened by any other rugby competition in the short term. The top paid NRL player might be making $3 million per annum in just three years.

Notes

• Our friends at have turned out two of the best pieces of analysis of the year since I finished the regular newsletter. I highly recommend a read of Who will take home the 2025 Dally M Medal? and A brief stats nerds’ tour of the Broncos’ 2025 season and subscribe.

• The lowest rating game for which we have both FTA and PTV numbers was the Cowboys defeating the Titans, 32-6, in round 13 of the 2022 season. Only 351k tuned in for that. Strangely, Johnathan Thurston’s last game, also a Titans-Cowboys clash, was the highest rated PTV-only game, in round 25 of 2018, with an even 400,000 viewers.

• The lowest rating pay TV game whose number I trust was the Warriors defeating the Tigers, 30-22, in round 24 of the 2023 season. This is the only club game in the dataset with fewer than 100,000 viewers.

• The highest rated regular season game in the dataset was the Eels defeating the Broncos, 34-6, on resumption of play in 2020, with over 1.2 million viewers on FTA and PTV. The NRL claims that the at least one of the free-to-air Vegas games and at least one of the Conflicts on Caxton has out-rated that subsequently, although we don’t have the specific data for that because it includes a heavy component of streaming.

Welcome to Stats Drop, an inundation of rugby league numbers.

If you’re receiving this by email, a reminder that the Datawrapper embeds work best on desktop, then next best if you click through on mobile and least best as presented in the email.

Regular readers will know that I find assessing coaching to be a particular point of fascination. This is a mix of trying to understand club decision-making and good old fashioned spite about self-aggrandising coaches justifying their place and pay packet.1

I set the line for this season’s firings at 2.5 and took the over. While the year is not out, it seems likely we will only get the scalps of Adam O’Brien and Des Hasler. This is a less bloody year, after four firings in 2022 (line: 4.5), three in 2023 and another three in 2024 (line: 1.5). Three coaches lose their job every year on average, which allows a healthy turnover of deadwood that have run out of ideas and allows ambitious assistants opportunity to rise.

What constitutes a “sacking” is a little arbitrary, as is the timeline used for the above. I didn’t include interim coaches finishing their term in office and events like Bennett moving on from Newcastle at the end of 2014 or the Seibold-Bennett swap at the end of 2018 do not count. The reason this chart does not go back further is that I would have to start looking up what happened to individuals to assess whether the coach left of their free own will or were forced out. Still, the clubs that top the list are those that you would expect to see: the most chaotic and the least competent.

Were these justified sackings or the result of unrealistic expectations by management? There are two ways of looking at successful coaching:

1. Is the club in a better place now than before?

2. Is the coach getting more out of the players than expected?

We measure the former by the change in class Elo rating (one rating point change is equivalent to one Thompson), as class Elo ratings go up with wins and go down with losses. The latter compares the pre-season projected Z score of the players to their actual performances (one unit of Z score of outperformance per game is equivalent to one Bax).

Here are the 2025 NRLM coaches’ results:

The tenure of the king of the loser quadrant, Adam O’Brien, is finally finished. It is both funny and galling that O’Brien has delivered some of the best results the Knights have ever had while also outstaying his welcome by two years. While Des Hasler technically got more out of his players than the projections suggested, this was below league average and, worse, the Titans went backwards from a position that was already miles behind the competition. Even for Parkwood, that’s unacceptable.

Wayne Bennett continues to coast on his resume. While Souths were decimated with injury, the old man’s talent is meant to be getting more out of his players than you’d expect. With replacement level guys drafted in to replace the stars suffering a slew of soft tissue injuries, Souths still did not outperform their pre-season projections. This is not the first time this has happened to Bennett and seems to increasing in frequency, a fact that will go by the wayside until it is convenient to a club to start pointing it out. Bennett’s numbers are not terribly dissimilar to Demetriou’s performance in 2024, even if Demetriou managed to cram all of that in before Magic Round.

People, idiots mostly, wanted to wrap Ivan Cleary’s efforts in 2025, as if that man hasn’t accumulated enough plaudits during the dynasty. The argument went that he coached a team from last place mid-season to a preliminary final. I ask, pray tell, whomst put them into last place after 11 rounds? Mayhaps we should assess the entire body of work, including the awful start, instead of applying the Young Principle? While his presence in the hot seat quadrant is more mean regression than anything else, we still don’t have to reward it.

Benji Marshall is an interesting case. His players underproduced what was projected for them but the club won more games. This reverses the polarity of the 2024 season, where the club went backwards slightly but the roster outperformed. Coming into his third year, it’s really make or break time. He landed in one of the worst situations imaginable - a rookie trying to learn the trade under the has-been Sheens, working under incompetents who have been turfed, various other board turmoils, trying to build a roster out of a club with three consecutive spoons - but those excuses, as voluminous and extensive as they are, will only last so long.

Anthony Seibold continues to deliver a replacement level coaching experience at the most mediocre club in the NRL. Cameron Ciraldo remains underwhelming and one wonders how long that Gouldian edifice will last. Ricky Stuart is the new Wayne Bennett, but without the titles, superlative ability or the recruiting attraction of living in a real city. We’ll wait and see if Ryles has what it takes next year. There will be a lot of scrutiny and pressure to deliver something.

For the sickos, here’s the Cup coaching performances in case your NRL club needed a new assistant:

It’s funny to see Dave Penna, someone I had on the hotseat in pre-season, put up pretty similar numbers to Matt Church, who I assumed would have a job for life after breaking the Tigers’ drought in 2023. I feel like the former was due and the latter was hard done by but it seems that the results matter in QCup, irrespective of your resume or the roster to work with.

We might also conclude that Burleigh’s Luke Burt had one of the better rosters with which to work, so his premiership victory is perhaps not as reflective of the coaching situation as the results of Rohan Smith, Paul Aiton, Terry Campese or Mat Head. I still have Burt down as coach of the year because it is hard to go past the minor premiers dominantly winning the grand final and completing a reload at Pizzey Park. Whether Ingebrigtsen can continue to take the Jets forward will be one of the compelling questions of 2026.

I don’t have a lot of insight to offer on the NSW Cup coaching carousel, other than the obvious that Daniel Tangata-Toa got a lot out of what is normally a competitive Warriors operation and was duly rewarded with a premiership, and to note Aaron Payne returning to North Queensland from Western Suburbs, after being given the bum’s rush out of the Blackhawks as part of the fallout of the Kyle Laybutt scandal. I expect he will pop up somewhere in the Cowboys system.

Thanks for reading Stats Drop

If this your first time, you might like to subscribe to get future free versions of these newsletters below.

If you enjoyed it, consider sharing with the people you think it might interest.

Share

Oatesian Tweeners

Before the NRLW season started, I spotted the Roosters’ starting second row for round 1 were a pair of wingers. Over the course of the season, the Broncos also moved Lauren Dam, career centre/winger, to the second row, and Bree Spreadborough, mostly a back, to lock.

Is that enough to call it a trend? Did Corey Oates, who claimed he wanted to play second row but played his entire career on the wing, start a movement? This transition doesn’t follow the usual rugby league spectrum of wing to centre to edge forward to middle forward, changing size and speed in inverse proportion along the line.2 We see moves to adjacent positions regularly but moving from wing to edge misses a station.

That’s weird and intriguing but it also makes sense. There’s a dearth of second rowers and it is an increasingly important role. The prototypical winger now has a big body, not unlike an edge, and is expected to make metres. If that body can also hit a hole or run a decoy or make a tackle, then maybe the leap isn’t as significant as thought. If wingers are in oversupply and second rowers are in undersupply, then perhaps the halves-to-hooker machine that was in vogue a few seasons ago can be retooled to turn wingers into edges.

When I looked into it, it turned out that this was not so much a trend, as just something the Roosters tried for round 1 (and then Mia Wood spent the season oscillating between wing and the bench) and the two Broncos, and even then, Spreadborough was playing lock, not second row.

Of the players who started at least two NRLM regular season games in the second row and wing over the course of 2022 to 2025, only Jayden Okunbor (seven wing starts, five edge) and Asu Kepaoa (5 and 4) qualified and neither of them played NRL in 2025. The qualifiers in the other competitions were:

• NRLW: Jasmin Strange, China Polata, Shenai Lendill, Lauren Dam

• QCup: Ioane Seiuli, Kane Bradley, Tom Farr, Kurtis Farr

• QDub: Madeline Burton, Lauren Dam (again), Mattice McLeod

• NSW Cup: Nicholas O’Meley, Harry Hayes, Asu Kepaoa (again), Jed Stuart, Brandon Mansfield

The only other player who I thought might qualify was Bob Tenza, who only played one game this season and spent his entire QCup career on the wing, and it turns out I had conflated Tenza with Zev John, who has only ever played a forward role, by virtue of them both being Papua New Guineans who play for the Capras with Anglo first names in their name.3

If the confusion of two players with somewhat similar vital statistics is required to generate examples, then there can’t possibly be that many in the first place. Granted, this approach might miss players who have an edge-like role, but come off the bench for the bulk of their minutes or players who are listed in a way that is inconsistent with their on-field role or players who have transitioned between competitions.

Nonetheless, there is definitely an Oatesian “tweener” player type: one that can play in the three-quarter line at centre or winger, as well as in the pack. Who are they and are they any good at it?

To assess whether players are any good at switch hitting between the backs and forwards, I’ve plotted their Z score while starting in the forwards against their Z score in the centre or wing (three-quarters) positions. The further to the up and right, the more productive the player was in those roles, with over 100 representing production over the average expected for that position.

There are more female than male examples and that there are more Cup players than NRL players that meet the criteria. This tells us, much in the same way that completion rates can be used as a proxy for professionalism, we can use this chart as a proxy for specialisation. The NRLM has only one representative that met the criteria of:

• Played at least one regular season game in the competition in 2025 at wing, centre, prop, lock, 2nd row or from the bench.

• Has played at least two regular season games in the period of 2023 to 2025 at wing/centre and another two at prop/lock/2nd row.

• Has not played 100% of their games at centre and 2nd row.

• Has played at least 25% of their games in the backs and at least 25% of their games in the pack.

Which is a complex way of introducing Jack Bird, the utility of utilities. Bird, a former Bronco, Shark, Dragon and current Tiger, can do in a pinch at the centre position but still has some juice when in the middle of the field, even if that assessment is more reflective of his work in 2023 than 2025.

We have five players from the NRLW but 11 from the QDub. There are six from QCup and eight from NSW Cup compared to the lone Bird from the NRLM. Possibly there would be no qualifying players from the NRLM in 2026. As one heads down the pyramid, specialisation declines in the face of the need to get bodies into positions. If a player is professional, they have time to train and hone the craft of centre, wing, edge and moving them becomes more risky. If a player is semi-professional, they are more likely to be used to plug holes as their skillset is less refined, the quality of play lower and the margins less narrow.

Shuffling players along the spectrum also seems to be more common among bad teams than good ones and, with some exceptions, for bad players rather than good ones. The number of Capras that make the list, from both the men’s and women’s competitions, I think is signal, not noise.

Nonetheless, there are interesting notes to be made at the boundary of the chart. Jordan Samrani and Nick Tsougranis, two NSW Cup players, appear to be quite good at playing both roles but definitely have a strong suit (Samrani in the forwards, Tsougranis in the backs). Both made their debuts in the NRL this year, for the Eels and Dragons respectively. Something to keep an eye on to see how they progress and if they’re offered the same positional opportunities if they hold their places in the majors.

I have noted some “pure examples”, which are the players (they’re all women) who have followed the transition hypothesis posed in the first paragraphs. The first pure example was China Polata, current Cowboys and one-time Maroon, whose career has been stricken with injury over the last two years. Far be it from me to blame what I believe to be a Ben Jeffries-initiated move from the backs to the forwards for the repeated injuries - it could just be bad luck - but I can’t say I’ve seen what that move was meant to bring, compared to how she tore the QDub apart and got herself a Maroons call up on the wing in 2023. We’ll see what 2026 delivers if Polata is fully fit.

Of the other pure examples, Jasmin Strange’s move is probably a success: perhaps her old man does indeed know ball. The move to the forwards has also greatly benefitted Bree Spreadborough, who had very little to offer as a wing or centre at the Broncos, despite her history for Central Queensland, which is plain to the eye test.

Portia Bourke is a more curious case. She transferred mid-season from the Norths Devils to the Wests Tigers. At the Devils, she was much stronger in the pack. Of the four games she’s played in the NRLW, she killed it in one game as a centre (a rare W for the Tigers) and has been less impressive at second row in three other games.

Conversely, Lauren Dam might be more suited to staying in the backs but if the next step down on the depth chart for edge forwards is well below replacement level, then the Z score of 70 that Dam will give you is more than enough, especially if you can find someone like Kerri Johnston who is going to give you 142 on the wing in lieu. The team benefits overall, even if the individual player is not wholly optimised.

The other two players I want people to talk about are Kalolo Saitaua and Emily Bella. I am a relatively big fan of both, to the extent that anyone is a fan of players at their level without being related to them. Saitaua deserved a train and trial or a pre-season or something with a NRL club but will be 30 next year and so the moment has probably passed him by. Bella was the U19 Maroons utility that swung the 2023 game in Queensland’s favour and still has her future ahead of her, although I’m worried the Cowboys machine is going to fail her.

Based on their play in the statewide competitions, both can play anywhere the coach throws them. To me, the stats pervert, that seems like a huge advantage. To others, the alleged ball knowers, this may mean they’re not good enough at anything in particular to lock down a position. I don’t know what to tell you if you’re not going to watch a lot of Wynnum men’s or Mackay women’s footy this year but these seem to be truly rare talents: more versatile than even the Oatestian tweeners.

Accolades

Now to announce 2025’s list of The Maroon Observer Players of the Year, which is judged on a mix of eye test, memory and stats by a jury of me alone.

The previous winners are listed here, along with the predictions for this year’s winners. Of the NRL players, the only prediction that got close was Jayden Campbell, maybe Max Plath, which is funny that I could have easily fixed it to make myself look smarter and no one would have noticed or cared but that’s honesty.

This year, I split out separate awards for NRLM and NRLW coaches of Queensland clubs (Karyn Murphy is retrospectively 2023’s winner and Scott Prince, 2024). I also dropped the Dude of the Year from the QCup clubs on the basis that there were very few Dudes of any calibre this year. The #1 players are the top rated in that competition by Wins Above Reserve Grade, on which I will publish more another time.

NRLM

• Broncos Player of the Year: Payne Haas

• Cowboys Player of the Year: Tom Dearden

• Dolphins Player of the Year: Isaiya Katoa (2nd)

• Titans Player of the Year: Phillip Sami

• Dude #1, Player of Players: James Tedesco, Roosters (4th)

• Q4 Coach of the Year: Michael Maguire, Broncos

NRLW

• Broncos Player of the Year: Tamika Upton

• Cowboys Player of the Year: Jakiya Whitfeld

• Titans Player of the Year: Jessika Elliston

• Dude #1, Player of Players: Tamika Upton, Broncos

• Q3 Coach of the Year: Ricky Henry, Cowboys

QCup

• Bears Guy of the Year: Guy Hamilton (2nd)

• Blackhawks Guy of the Year: Dudley Dotoi

• Capras Guy of the Year: Tevita Naufahu

• Clydesdales Guy of the Year: Corey Fenning

• Cutters Guy of the Year: Raydan Burns

• Devils Guy of the Year: James Flack

• Dolphins Guy of the Year: Aublix Tawha

• Falcons Guy of the Year: Zion Johnson

• Hunters Guy of the Year: Sanny Wabo

• Jets Guy of the Year: Oliver Pascoe

• Magpies Guy of the Year: Anthony Milford

• Pride Guy of the Year: Dane Aukafolau

• Tigers Guy of the Year: Asher O’Donnell

• Tweed Guy of the Year: Joseph Vuna

• Wynnum Guy of the Year: Ben Farr

• Guy #1, Player of Players: Takitau Mapapalangi, Bears

• QCup Coach of the Year: Luke Burt, Bears

• Proof of Concept Coach of the Year: Tye Ingebrigtsen, Jets

QDub

• Player of the Year: Georgia Sim, Clydesdales

• Guy #1, Player of Players: Joeannah Naime, Magpies

The Broncos are back where they’re meant to be: Titletown, Premiership City, Queensland

We’re out of the void!

Here’s the last paywall for the year. Behind, we have some minor business to conclude regarding dashboards and link to the final update of The Dataset™ for 2025.

If you’ve read this far down, past the Broncos gronking, you might want to consider upgrading from an unpaid to a paid subscription. Your financial support will keep the lights on at The Maroon Observer and comes with some neat benefits, including the ability to leave a comment, access to The Dataset™, extras in the dedicated club newsletters and my gratitude.

But it’s the offseason and there’s not much coming until 2026, so I’ve knocked 20% off annual subscriptions until New Year’s Day. If you thought a paid subscription was too pricey, now’s the time to rexamine your priors.

Offseason Special

Read more

Welcome to Stats Drop, an inundation of rugby league numbers.

If you’re receiving this by email, a reminder that the Datawrapper embeds work best on desktop, then next best if you click through on mobile and least best as presented in the email.

What is the completion rate? A betting website that topped Google says:

The completion rate is the percentage by which a team 'completes' their set of six tackles, without an error on any of the plays.

Probably the most important NRL stat other than tries, the completion rate indicates everything from how many attacking opportunities a team presents to the opposition, to discipline, and how well coached the team is.

I have added the bold because that is a bold claim. Even without using any graphs, I think you’d be hard pressed to argue that completions are the second most important stat. Perhaps it is the one given the second most emphasis.

There is considerable disdain from the Amateur Rugby League Analytics Fraternity (or ARLAF1) towards completion rates. To take a quote out of context from the Rugby League Eye Test:

This reinforces the theory I have that completion rate is a junk statistic…

This disdain arises from the gap between the importance collectively assigned to the completion rate versus its usefulness. Personally, I’ve occupied different perspectives on completion rates: from not thinking about it at all, to dismissing it as useless information for the uninformed, to coming back around to finding some value, as I try to process a dozen games a week for the newsletter.

Today, we will explore the junkiness of completion rates. We’ll be looking at the NRLM above the fold. Below, after the paywall - UPGRADE TO A PAID SUBSCRIPTION TODAY - we’ll look at the other competitions. The Dataset, a link to which is also after the paywall, only has completion rate data for the NRLM from 2021 to round 19, 20252 because I haven’t gone back and gotten the rest of the data yet. That’s a (very tedious) off-season project.

Over this period, rates have fluctuated little. I think, but will have to prove after I’ve recollected the data, the long run average is something like 75% from 2013, which is as far back as NRL dot com goes. It would not be surprising to discover the predominance of the Panthers’ style of play is a sustained deviation from long term historical trends but within the window we have available, rates are stable.

There isn’t a great deal of variation within the season either. Average rates start a little below average as players lose their off-season rustiness, rise through the middle of the season and then drop away again as teams stop caring. While this is a convenient narrative, most of this change could be within the bounds of statistical insignificance.

Given completions rates are tied to errors, and errors occur randomly at a given rate in accordance with a skewed Poisson distribution [citation needed], then it is less surprising that there’s no meaningful change over a limited timeframe (either of a season or several years) and all of the potential sources of variance, and explanatory power, wash out over a large enough volume of n.

Further to this point, we have to be careful with this chart. NRL dot com doesn’t note the weather or ground condition for every game and presumably this comes down to how the ground manager/stats recorder felt on the day. It’s the kind of thing that fell by the wayside during covid.

With that in mind, we acknowledge that the partly cloudy set was one (1) game of 408 games that we have recorded weather for of the 986 games that were played. 25 games were listed as ‘rain’, 20 as ‘showers’, three as ‘light rain’ and three as ‘cloudy’. The balance of the 408 were listed as ‘fine’.

To reinforce an old point, you will never know who deals with the wet the best, irrespective of how well search engine optimised the post might be. I am hesitant to draw too many conclusions about wet weather footy but it’s an interesting anecdote nonetheless.

There’s a plainly obvious correlation between completion rates and winning football games. The argument about whether this is a junk stat ends there. In two-thirds of games, the team with the superior completion rate won the game. You complete more sets, you win more games. The simplicity is why it is memetically effective.

If you’ve read this far, it’s likely that you’re smart enough to work out the underlying mechanism. If a team has the ball more, uses it and doesn’t turn it over to errors or penalties, then they’re going to score more points while denying the opposition the same opportunity. It’s not complex - even a LLM can figure it out - but the completion rate isn’t the driver - possession, errors and efficiency are the drivers to the outcome - but completion rates are narratively concise.

The correlation between a team’s completion rate and the margin has a low R-squared of just 0.13 - certainly well below expectations for “second most important stat” status - but putting the completion rate into buckets of 5% and comparing to winning percentage of those teams has an R-squared of 0.98, which is line with what we would expect to see for tries scored and line breaks.

The correlation between net completion rate (that is, the home team’s minus the away team’s completion rate) and margin improves the R-squared to 0.23. There is a useful relationship between playing better than your opponent and how badly you beat them by but it is not an especially strong correlation.

This provides important insight into how we should talk about completion rates. The exact percentage is barely relevant. As running for 1405 metres or 1415 metres won’t make a meaningful difference to the outcome of the game, whether the completion rate is 74% or 76% is similarly unimportant.

We should probably just look at the first number - is it a 5, 6, 7, 8 or 9? - and that contains all of the worthwhile information we’re going to get. If both teams have the same significant digit, then both teams are playing at roughly the same level.

To complete Rugby League Eye Test’s out of context quote:

…in a vacuum. All things equal, promoting completion rate above all else would see you value completing 24/27 sets (88%) above completing 33/38 (87%). Again, it’s not the rate that you need to focus on, it’s the relation of it to your opponent.

Completion rates, even in a vacuum still correlate with winning. However, the completion rate lacks precision, in a way that counting errors, line breaks, penalties and tries do not. As discussed under the last Stats Drop, there are better ways of measuring what completion rates attempt to achieve but it is worth remembering that even scoring more tries doesn’t guarantee victory: just ask Zac Lomax or the 2023 Burleigh Bears.

Further, most advanced metrics do not have perfect explanatory power. The team with the superior SCWP wins about 77% of games and the team with the superior FTy wins 85%. Taylor-based player ratings are around 60% successful at predicting the match winner.

Instead, let’s consider what is being communicated. The point of statistics is to take the chaos of a football match and reduce it down to a few comprehensible numbers. Who are our potential users of completion rates?

• Coaches should be operating in a more sophisticated analytical environment than completion rates offer, capable of digesting up to several numbers simultaneously to understand what’s going on.

• Commentators find completion rates useful. Big number better. Closer to 100% means closer to perfection. Easy stuff. Makes it sound like you know what you’re talking about. Great for getting across in halftime spot between ads for McDonalds.

• Players will be a little of column A and a little of column B.

• You, the reader, members of ARLAF and associated sickos are handsome and sophisticated enjoyers of rugby league, so completion rates are worse than useless: they are a de-humanising insult to the collective intelligence of the community.

• Most normal people don’t think about this at all3.

If you gave up on completion rate, could you explain your expected points model or that actually in this case six errors isn’t that many because they’ve had so much ball cogently and quickly enough to make a point that the general audience will actually understand? Even allowing for the low bar set by the mainstream coverage of the sport, this would be unlikely to improve clarity.

Completion rates can also serve a purpose for describing playing style. Famously, Trent Robinson’s teams eschew completions in favour of whatever it is Sam Walker happens to be delivering. Ivan Cleary’s sides play with a rigid discipline and commitment, reflected in the highest completion rate in the league. You might then count the rings over the last four seasons and draw a conclusion, even noting that Robinson’s rate is a whole 5% lower than Cleary’s but as we can see, it’s a much more mixed bag than that.

The comparison between Robinson and Cleary is about as extreme a comparison as could be reliably made4 to make something like a valid point but the rest of the league is there or thereabouts and it’s much of a muchness. Maybe the normal people are onto something.

Thanks for reading Stats Drop

If this your first time, you might like to subscribe to get future free versions of these newsletters below.

If you enjoyed it, consider sharing with the people you think it might interest.

Share

If you really enjoyed it, considering upgrading to a paid subscription to access what's beyond this paywall or chuck us a few bucks on Ko-Fi.

Upgrade to paid

After the paywall, we're looking at completions in the women’s and minor leagues, competition Futures, Mandates of Heaven, updated and historical Dashboards and a fresh link to The Dataset™.

Read more

Welcome to Stats Drop, an inundation of rugby league numbers.

If you’re receiving this by email, a reminder that the Datawrapper embeds work best on desktop, then next best if you click through on mobile and least best as presented in the email.

Long time and close readers will recall that there are a few counting stats that can be ascribed to individual players that neatly and reliably correlate with their teams winning. As explained at the start of the year, these have been litigated often enough that it can stand as axiomatic. These stats are tries scored, running metres (including kick return and post-contact), line breaks, tackle busts, assists for tries and line breaks, errors and missed tackles (the latter two having a negative correlation). Sometimes kick metres matter, sometimes not.

There are other stats that correlate and duplicate some element of the above - hit ups, runs, etc - and stats in other datasets that aren’t included in what NRL.com serves up and on this matter, we defer to the Rugby Leauge Eye Test’s recent piece, Follow this one simple trick to win more NRL games.

Over a long enough time frame, the nature of these stats and their relative values changes. Sometimes this is in response to the changing nature of the game and its athletes or administrators. Sometimes this is in response to the Brownian motion of society’s random walk. Let’s have a look at what’s been happening.

Running metres are a critical stat. Rugby league is a simple game and getting closer to the try line increases the odds of scoring points. Therefore, making metres is a fundamental part of the sport. Cross-checking a scoreline against the yardage battle will give you as good an understanding of the game as any four numbers will. It’s simplicity and elegance remain underrated by the stats connoisseur.

The arc of history bends towards justice, as well as towards greater inflation of running metres. The introduction of Vlandoball I increased average running metres by 100 metres per team per game overnight. That effect dissipated over time but has returned with a vengeance in Vlandoball II, the last three years representing the greatest amount of yards being made in the history of the NRL.

Interestingly, while the generation of metres is at its highest rate, the average gap between the two teams is quite low. Vlandoball I blew that out to historic highs, so whatever re-balancing that Vlandoball II involves has kept the raw production without the lopsidedness, or at least as viewed as a season-long mean. There may be considerable variance within those datasets that we are not exploring today.

The expectation about who is generating those metres has changed materially over time. The share of metres generated by the lock position seems to have crescendoed with peak Taumalolo in 2017 but returned to a more historically consistent ball playing lock mentality in the 20s, more closely reflecting the utilisation of the 90s. Wingers’ and centres’ shares have only risen over time, without much of a drop from the fullback, which explains where the extra metres are coming from to make up the shortfall that had traditionally come from up front.

Less remarked upon is the utter collapse in the front row. The average prop ran for 120 metres in 2000 but now can only manage 113 metres in 2025. Taking into account the overall inflation of running metres, the share of metres made by forwards has reduced by 25% since 2000 and by hookers has fallen by 35%. Not unlike the demise of the Australian-made car in the national vehicle fleet, this recession reflects the changing roles of players: props becoming sharper rams, rather than dull bludgeons, and hookers focussing almost exclusively on playmaking duties out of dummy half, relying on others to advance the ball.

After the fold, we’ll look at a few other metrics, including missed tackles, line breaks and the exchange rate of production.

Thanks for reading Stats Drop

If this your first time, you might like to subscribe to get future free versions of these newsletters below.

If you enjoyed it, consider sharing with the people you think it might interest.

Share

If you really enjoyed it, considering upgrading to a paid subscription to access what's beyond this paywall or chuck us a few bucks on Ko-Fi.

Upgrade to paid

After the paywall, we've got more charts, trivia, Futures, Mandates of Heaven, updated and historical Dashboards for NRLM, NRLW, Queensland and NSW Cups and The Dataset™.

Read more

Welcome to Stats Drop, an inundation of rugby league numbers.

If you’re receiving this by email, a reminder that the Datawrapper embeds work best on desktop, then next best if you click through on mobile and least best as presented in the email.

As the old saying goes, “the forwards decide who wins, the backs decide by how much.”

During halftime of the semi-final in 2021* World Cup, as the Kangaroos trailed the Kiwis, 14-10 and approximately 200 running metres, I tweeted something to the effect of “Nathan Cleary has never won a big game without either James Maloney or the pack winning yardage by at least 300 metres.”

That wasn’t hyperbole by the way. I verified that very quickly against his grand final and Origin appearances. Australia won that game, 16-14, closing the gap to New Zealand to a deficit of just 25 metres for a career first for the bechinned one.

That may seem like an unfair accusation, and it is, but I also do not care about Nathan Cleary’s feelings about a now-deleted social media post. What is relevant for this edition of Stats Drop is that I have no idea what the rate at which playmakers win games with the yardage battle stacked against them. It may prove that Cleary’s conditions for winning a big game are typical, even good?

Let’s find out what the relationship is between the differences in running metres over a game (as a proxy for the fabled ‘platform’ built by the pack) and the performance of individual halfbacks and five-eighths.

This chart plots Z score, as the measure of production of individual halves (minimum 20 starts in those positions since 2020), against the average surplus/deficit of their team’s running metres.

In and of itself, this chart doesn’t prove much about the players themselves. There is a mild correlation, with an R-squared of 0.18 (zero is no correlation or random, one is perfect correlation). That is lower than I would have guessed, both unseen and having eyeballed the chart, but establishes a league average mark.

Perhaps the best takeaway is if you’re regularly getting superior yards but also getting inferior production, we can safely rule the player out of the elite tier of halves. The best case scenario is that they can serve as a foil for a better partner or are only included because they have had plenty of opportunities to fill-in for an Origin-calibre half carrying the team over the last five years.

It doesn’t tell us a lot about the connection between an individual player’s performance and how it correlates to the team’s performance within their specific dataset. On average, Nathan Cleary is basically always getting a lift but that average could cover a considerable variation of outcomes.

If we look within each player’s dataset, we can correlate game Z scores with yardage differences and calculate an R-squared (coefficient of correlation) for each player. We then infer that a higher correlation between yardage surplus and Z score means that player is more reliant on having a well built platform for their own performance.

To be clear, “inferior performance” is a classification label of the player’s production and not meant to imply that players get worse as their reliance increases. As established in the first graph, players generally get better as the yardage battle is more convincingly won.

Assuming that we all agree that a player being able to consistently produce, irrespective of his teammates’ performance, is considered a desirable trait, this graph meets the eye test and the gut feel. The top left quadrant represents players who have been historically considered “good”, current form notwithstanding, and players in the bottom right have not. I’ve drawn a line at Z score = 100, which is to indicate the notional “average” production for the purposes of breaking the graph into quadrants but I think there’s a zone below 80 (the highlighted region), on which Ethan Strange and Jamal Fogarty sit on the cusp and Sean O’Sullivan is just below, where players’ production becomes untenable for first grade, doubly so for a leading half. There are exceptions - Isaiya Katoa, whose first grade career still features a high proportion of nothing games for the Dolphins that I would expect will make up a diminishing part of his resume as time goes by (Ezra Mam’s career has had a similar trajectory) - but there’s recognisable players, including one player I refuse to mention by name, but no stars in that higlighted region.

Nathan Cleary is not labelled but he is the dot next to Jahrome Hughes in the lower reliance / superior performance quadrant. As these are regular club games and not “big games”, using whatever definition of that term best suits the argument I’m making for the purposes of getting faves on a dead social media platform, it doesn’t invalidate my point but gives you a better context. We’ll have more on these two after the paywall.

The results are still not as clear cut as it seems. There is some overlap between player Z score (a component of which is running metres) and team running metres, but you would expect that to have a relatively low impact on the conclusions being drawn here because of how little bulk metres matters to being a half.

Without going into the hundred-odd individual datasets, we don’t know exactly much of this is determined by weird outliers or how diverse each dataset is. For example, Cleary and Hughes may still have datasets comprising almost entirely of yardage wins, which will look like low correlation because they are playing under almost identical conditions from week-to-week, and won’t tell us much about how they might perform in yardage losses. The league average standard deviation for yardage difference within each player’s dataset was 361, with a minimum of 242 (George Williams) and a maximum of 500 (Brandon Wakeham), suggests this effect is common enough across all players. Separating it out may prove difficult and it may be easier to live with it.

Still, I think taking both graphs together paints a picture about the production, performance and effectiveness of specific halves whether their team is on top or not.

Thanks for reading Stats Drop

If this your first time, you might like to subscribe to get future free versions of these newsletters below.

If you enjoyed it, consider sharing with the people you think it might interest.

Share

If you really enjoyed it, considering upgrading to a paid subscription to access what's beyond this paywall or chuck us a few bucks on Ko-Fi.

Upgrade to paid

After the paywall, we've got more about Hughes and Cleary in particular, Futures, Mandates of Heaven, The Dataset™, updated and historical Dashboards for NRLM, Queensland and NSW Cups.

Read more

Welcome to Stats Drop, an inundation of rugby league numbers.

If you’re receiving this by email, a reminder that the Datawrapper embeds work best on desktop, then next best if you click through on mobile and least best as presented in the email.

Where did our chances go?

Winning a premiership is hard. Everything can go right until it doesn’t and then that’s a year of work wasted with nothing to show for it. Just ask Melbourne, Brisbane, Parramatta, South Sydney, Penrith, Canberra, Melbourne (again), North Queensland and Melbourne and only the women at the Gold Coast Titans (the men’s team has no bloody idea).

A fair while ago now, I looked at expected premierships: how many premierships should a club have won in the NRL era? At the time, I ripped off Matt Cowgill’s methodology but variations have also been deployed by Neil Paine for US sports. It feels high time to gaze once more into the abyss where premiership hopes and dreams die in inky blackness and absolute silence.

The classic method to estimate expected premierships is to use Elo ratings (form ratings from my particular toolkit) as teams head into the post-season and run Monte Carlo simulations on the outcomes of the finals series. This will give each finalist a probability of winning the title that, over time, add up to a number of expected premierships that team should have won. The gap between the expectation and the actual grand finals won is where existential crises live.

Having had seven years to think about abysses of varying types, I don’t think this is quite right. The Titans have only made finals four times in their existence which means they would only be credited with four parts of an expected premiership (2009, 2010, 2016 and 2021). In theory, Gold Coast should have nearly 20 parts of an expected premiership because they have a shot at the title every season, until their incompetence rules them out of it. It might not add up to much either way (paying subscribers will get to see exactly what the difference is) but deeming their expected premierships prospects as exactly zero in almost every year doesn’t feel right. There should be some expectations in a closed, tightly salary capped league.

For this post, I ran full season Monte Carlo simulations (5k times) from 1998 to 2025 based on pre-season class Elo ratings.

Class Elo ratings tend to lag reality too much to be useful for predictions, which is feature and not a bug. However, class ratings are a good marker of sentiment, which makes them perfect for this exercise. While the alignment of exactly when an expected premiership is accrued is not quite right (e.g. did the Broncos ever have a 5% chance of winning the 2020 premiership?), over time that effect washes out as the class ratings catch up.1

It is completely unsurprising to see the Panthers well ahead of their expectations. Even with their record-setting ratings, it’s a coin flip of Penrith versus the field, so that comes out as half an expected premiership. If a club wins four coin flips, they’re going to get two whole premierships above the abyss.

Another way of looking at this is that the Panthers and the Roosters won - stole, perhaps, or worse, misappropriated - premierships that should have gone to New Zealand, Parramatta, Canberra but didn’t due to the random walk that is professional sport. We’ll probably be adding Canterbury to that bonfire in due course.

Panthers and Roosters aside, the rest of the league can be broken into rough groups:

• An untouchable caste of just the Titans.

• Perennial losers that can’t get it done, including the Warriors, Raiders and Eels.

• Middle class, above average to below average teams coasting off a single title, comprising the Dragons, Cowboys, Knights, Sharks, Rabbitohs and Tigers.

• The Sea Eagles, as lower upper class types.

• Temporarily embarrassed millionaires fallen on hard times, namely the Broncos and Storm (depending on your perpsective the Storm are either in a huge hole or have dipped their toes in the abyss for the first time since 1999).

Where the Dolphins end up will only be revealed with time but I wouldn’t be surprised to see the untouchable caste come to consume all of the Southeast, including SEQ4 starting 2032.

Thanks for reading Stats Drop

If this your first time, you might like to subscribe to get future free versions of these newsletters below.

If you enjoyed it, consider sharing with the people you think it might interest.

Share

If you really enjoyed it, considering upgrading to a paid subscription to access what's beyond this paywall or chuck us a few bucks on Ko-Fi.

Upgrade to paid

After the paywall, we're going to look at some absolute choke jobs, the unpredictability of tipping, have a glimpse into the future and I’ve finally got the dashboards sorted out.

Read more

Welcome to Stats Drop, an inundation of rugby league numbers.

If you’re receiving this by email, a reminder that the Datawrapper embeds work best on desktop, then next best if you click through on mobile and least best as presented in the email.

One of the early problems I have encountered in 2025 is that while I have erected a paywall, I have not actuall…

Read more

Welcome to Stats Drop, an inundation of rugby league numbers.

For a change, I’m not talking about Queensland rugby league in this newsletter. The Broncos, Dolphins, Cowboys and Titans have already received longer, standalone season previews featuring the full suite of charts below. This is for the rest of the league. This is part 2 and you can read part 1 here.

Some detailed explainers on the stats: Players, Teams and PVRTS/Sims. I didn’t include any coaching stats in this post because it is dense enough as is, but you can refer to this earlier table.

If you’d like to continue to support The Maroon Observer and Stats Drop, you can become a free subscriber:

Or if you’d like unfettered access to all future Stats Drops, you can upgrade to a paid subscription. Note that the introductory offer ends when Vegas kicks off.

Upgrade to paid

Reminder: Datawrapper embeds look and work best on desktop, next best if you tap through on mobile and least best in email. There are so many graphs and charts, I have selected the most relevant for each team and then linked out to the others.

Dragons

Long time readers will know I have a specific antipathy for the St George Illawarra Dragons. It is not born of anything on the field, it is that they may well be the stupidest club in a league that features both Blake Solly and Bryan Fletcher on a regular basis as important people.

Among other things, there was the covid party, the character reference for an alleged pedophile on club letterhead and the underlying assumption that they are an important part of this competition because of something that happened six decades ago. In reality, the NRL would probably be better off handing their junior system over to the Sharks, merging St George into Easts and bringing the Steelers back as an independent entity.

Still, we should still look for a bright spot in this otherwise thoroughly mediocre franchise.

And there it is, right at the bottom, was an above average playmaking platoon. The backs, the pack and the overall team was below average to terrible as measured by Wins Above Reserve Grade but the playmakers were sixth best in the league by this metric.

Obviously, Ben Hunt is gone and he represented one leg of this troika, only missing two games in 2024. Lachlan Ilias is his replacement and the nicest thing I can say is that we will see what happens there. Kyle Flanagan appears to be reprising his role as five-eighth - can’t imagine why - after playing 20 games at #6 last season. Damien Cook, even on the wrong side of the age curve, is a huge upgrade on Jacob Liddle. If he stays on the field, he may even be able to make up the production difference between Ilias and Hunt.

The rest of the team doesn’t inspire a lot of confidence, and even numerically, a lot of that is being propped up by Valentine Holmes, which is best described as a quirk of the system, but maybe Dragons fans can look at their dynamic trio and hope.

Further analysis: Players by Z / Players by WARG / History / Sims

Eels

Last year, I wrote about Parramatta announcing the return of James Hardie1 as their jersey sponsor. During the off season, the Eels terminated their deal with the Northern Territory that saw them take a game to Darwin to get dacked in front of 11,000 people.

One of those commercial arrangements had some benefit to the wider sport, in addition to the financial benefit accrued by the Eels at the cost of the Territorian taxpayer, even if all came at the expense of what was left of the Eels’ collective ego.

The other is unbelievably odious. Even with everything that’s happened in the world over the last quarter century, I still can’t quite wrap my head around James Hardie being an acceptable entity to associate with but it pushes the nostalgia buttons in the most credulous Eels fans’ brains, so that apparently makes it all OK.

Anyway, I’m calling Jason Ryles as a flop right here and now. Eels will bounceback this year, largely on experience, flatline next, largely on age, Dylan Brown will finally exercise one of his contractual options and Ryles won’t see Magic Round 2027. No, there are no stats for this. I am allowed to do hot takes too.

Fine, here’s some numbers:

Further analysis: 2024 Metrics / Players by Z / History / Sims

Panthers

We’ve blown well past the ‘well, this is the year it finally falls apart’ level of analysis that I deployed for the first couple of years of the dynasty. There were no more worlds left to conquer; (Greg) Alexander wept.2

The manner in which they’ve gone about their wins, characterised by unrelenting calm, confidence and stability in applying unavoidable pressure, is as good evidence as any that things won’t fall apart. I am not convinced that Jarome Luai was the linchpin holding this endeavour together and that the collective effort of the forward pack and the dude in the coaching box both get (relatively) little in the way of praise. Those factors may erode away over time but collapse, immediate and catastrophic, is extremely unlikely.

They could win it again. Who is going to stop them? Unless PVL adds the Super League to the NRL, we’ve already thrown our best and brightest at the problem and come up with nothing.

Further analysis: 2024 Metrics / Players by Z / Players by WARG / Sims (save yourself a click and don’t look)

Raiders

This is a good get for mine. Ricky has had enough time.

I never really know how to get a pre-season read on the Raiders. Were such an analysis possible, I think we’d agree they may be the most vibes-reliant team. Some years, Ricky has it right, others not so much. Some seasons they fluke a finals appearance, other they’re a non-entity by round 15.

Is the Raiders Zone real or does the notion represent a Heisenbergian uncertainty about The Milk and the only way to reach a deterministic outcome vis-a-vis whether The Milk is off is to force a ‘waveform collapse’ by playing the season? Perhaps but this is as confusing a metaphor as trying to tease out the Canberra Raiders’ prospects.

Further analysis: 2024 Metrics / Players by Z / Players by WARG / History

Sharks

Cronulla project as the second team, the Panthers’ antagonist, for this season. The Bunnies, the Eels, the Broncos and the Storm have all had a crack at being second best and de-throning the kings, and all came up empty handed.

Other than Will Kennedy, the lowest projected player in the starting 17 has a Z score of 85. That’s just 15% below average for the second worst player on this team, as judged by this metric. That gives us an insight into the depth and width of production on the Sharks’ 17 this year.

Provided Cronulla can hang on to Tom Hazelton, this is a formidably deep team. Layer on last year’s success and the previous successive stepping stones that Fitzgibbon has lead this team over since taking the reins form John Morris, all signs are pointing up.

The last question is whether they can actually get it done.

Further analysis: 2024 Metrics / Players by WARG / History / Sims

Warriors

People are talking about the Warriors as potential spoon contenders. I don’t personally subscribe to the belief that the Warriors are special in such a way that they will never get the spoon. The only team in the league that could make that claim got the spoon in 2020, proving anyone can fall apart to a catastrophic degree but only if they really put their minds to it.

I do have New Zealand slotted in as one of the outside-looking-in, below average teams that is their natural hunting ground.

If everything goes really well - the young guys prove themselves as first graders and Andrew Webster shows that 2023 wasn’t a fluke and a few rivals shit the bed - then a low end finals spot is within reach for this team.

But it is as likely as not that one of those things won’t come off - I am yet to be convinced that Webster wasn’t a flash in the pan, the rivals could just as easily strike it rich as the Warriors, the young guys are still young guys - and the Warriors are an also-ran in the 2025 season.

Further analysis: 2024 Metrics / Players by Z / Players by WARG / History

Welcome to Stats Drop, an inundation of rugby league numbers.

For a change, I’m not talking about Queensland rugby league in this newsletter. The Broncos, Dolphins, Cowboys and Titans have already received longer, standalone season previews featuring the full suite of charts below. This is for the rest of the league. This is part 1 and you can read part 2 here.

Some detailed explainers on the stats: Players, Teams and PVRTS/Sims. I didn’t include any coaching stats in this post because it is dense enough as is, but you can refer to this earlier table. We’ll come back to this later when the rookies have some runs on the board. I am setting this year’s line for coach firings at 2.5 and I’m taking the over.

If you’d like to continue to support The Maroon Observer and Stats Drop, you can become a free subscriber:

Or if you’d like unfettered access to all future Stats Drops, you can upgrade to a paid subscription. Note that the introductory offer ends when Vegas kicks off.

Upgrade to paid

Reminder: Datawrapper embeds look and work best on desktop, next best if you tap through on mobile and least best in email. There are so many graphs and charts, I have selected the most relevant for each team and then linked out to the others.

Bulldogs

Would you think of the Bulldogs as one of the league’s three best defensive teams?

The Bulldogs were the only team to top a metric without having won a premiership in the last decade. Two in particular - SCWP against and Defence Workrate - indicate their opposition had to generate a lot of metres, breaks and other production relative to the number of tries actually scored. More simply, the Bulldogs conceded the second fewest points per game in the league, behind Penrith and equal with Cronulla.

The Sims, by comparison, are way down on the Doggies but I think this is due to a relatively low starting point (refer History) that has not yet reflected the improvement this team has shown. It’s worth remembering that this improvement has been coming for years and its delivery would be marked as late by most mail carriers. Still, defensive prowess is usually replicable and provides a solid foundation for the Bulldogs.

Further analysis: Players by Z / Players by WARG / History / Sims

Knights

Woof, this was not a team that was good at scoring points.

Weirdly, Newcastle were midfield in terms of net sets but were down the bottom of the league in net time of possession. That means they had an average number of chances but managed to blow them so catastrophically, they held the ball for substantially less time than any other team in the league.

Even accounting for that, the Knights still didn’t score a lot of points. They were 16th in points for per game and in a similar position for efficiency (points per set), conversion (ratio of ARM+LB to points scored) and workrate (ratio of FTy to tries scored).

This is not a difficult puzzle to work out: Newcastle’s playmakers are not good. In fact, the Knights’ platoon ranks 17th of 17 teams in the league by Wins Above Reserve Grade. With an average projected Z score of 63 for 2025, it doesn’t look like that’s improving any time soon.

Further analysis: Players by Z / Players by WARG / History / Sims

Rabbitohs

A few weeks ago, I said something to the effect that “despite the numbers rendering them the worst starting 17, the Rabbitohs obviously aren’t the worst line up on paper.” Then this happened and this happened and this is a space for an inevitable third thing to happen between time of writing and time of reading.

When you take all of those guys out, there’s not a lot of meat on those bones. It is so dire, I am not prepared to spend even a couple of minutes putting together the 1-17 tables for Souths. If, as rumoured, Keaon Koloamatangi has a shoulder issue, he can’t even fill the void created by Cameron Murray’s Achilles rupture and that leaves two chasms to backfill with guys I’ve never heard of. Wayne will presumably work his magic and make a decent soup out of it, but it will be thin gruel, bordering on consomme.

Further analysis: 2024 Metrics / History / Sims

Roosters

While their nearest and most hated rivals struggle with an injury crisis, the Sydney City Roosters got ahead of the game by having an injury crisis much earlier.

As a result, Easts find themselves in a similar situation to Souths. Newcastle might be a better comparison as both have some nice pieces but you wouldn’t bank on the listed 6+7+9 as producing much in the way of results. Not being particularly familiar with the Roosters’ depth chart, I popped in Tyler Moriarty, the only other player on the Roosters website listed as a hooker, at utility but frankly, it doesn’t really matter who it is.

Considering that the Roosters were one of the finest attacking teams in 2024, the firm of Townsend, Smith and Watson isn’t going to be the basis for a successful litigation but may be able to hold the fort until some of the senior partners, Smith and Walker, return (we are more likely to see the former than the latter in 2025).

Further analysis: 2024 Metrics / Players by WARG / History / Sims

Sea Eagles

What would really restore my faith in rugby league right now, would be a complete and utter collapse from the Manly-Warringah Sea Eagles. I don’t have a particular animosity towards Manly, other than a low medium level simmering resentment toward the Sydney establishment in general and a few Celsius added by Manly’s small mindedness (both outwardly to the gays and inwardly to the insular peninsula), and am actually quite fond of Cherry-Evans these days.

However, I really do not care for Anthony Seibold. I have to think the success of last year’s Broncos season review was largely down to me referring to him as a “little piggy blowhard” because I expect the French Revolution motif mostly turned people off.

So I’d like to see the biggest bloviator outside of Phil Gould fail, again. I have no evidence that this will happen or even any particular reason to think it might, other than just wanting it. You miss 100% of the shots you don’t take, so here’s hoping some bad juju sinks the entire enterprise into the sea, like Marine Parade rapidly coming to terms with the effects of climate change in, let’s say, a decade or two.

Anyway, here’s what the fifth best team by Wins Above Reserve Grade and the second best back five by the same metric looked like:

Further analysis: 2024 Metrics / Players by Z / History / Sims

Storm

Just ink them in for another preliminary final now and flip a coin to see if they’re going to make the grand final.

Further analysis: 2024 Metrics / Players by Z / Players by WARG / History

Tigers

‘It can’t get any worse’ is an ancient proverb whose truth is disproven by the existence of the Wests Tigers. People want to bang on about the Panthers’ four-in-a-row but no one seems to be highlighting the historic ineptitude required for three spoons in a row. This puts the joint venture in company with such luminous rugby league outposts as Liverpool (Stanley and City, combined from 1948 to 1954), University (Sydney, 1934-37), Gold Coast (1991-93) and Newcastle (2015-17).

Much as Luai is not the linchpin of the Panthers’ endeavour, I doubt his arrival is what turns this team from perennial losers into serial winners. However, it probably does get them off the floor. No, this isn’t like when I said that about Koroisau but there are another half dozen imports to help out. It may be better to deploy another ancient proverb: ‘she’ll be right’.

Further analysis: 2024 Metrics / Players by Z / Players by WARG / Sims (it’s hard to overstate just how far off the pace the Tigers have been but this comes close)

Welcome to Stats Drop, an inundation of rugby league numbers.

This is part 2 of an introduction to Stats Drop metrics. You can read part 1 here.

If you’re receiving this by email, a reminder that the Datawrapper embeds work best on desktop, then next best if you click through on mobile and least best as presented in the email.

One of the nice things about quitting short form posting and posting exclusively in a longer format, other than not having tedious pedants in my replies and giving myself time to confirm whether this thought is worth sharing with the world, is the mitigation of context collapse:

Context collapse "generally occurs when a surfeit of different audiences occupy the same space, and a piece of information intended for one audience finds its way to another" with that new audience's reaction being uncharitable and highly negative for failing to understand the original context.

I would post a table or a chart to Twitter. The real freaks who followed my work would probably understand what it meant and sometimes hit the fav button or discuss the implications like a reasonable person might. Other times, you’d get an interloper, possibly a follower of a follower or someone who liked a joke I made three years earlier and had not paid any attention in the meantime, upset that Shaun Johnson is too highly rated or Nathan Cleary is too low rated. Against what? Who knows. It’s like they hadn’t read a single thing I’d written!

Rather than considering the possibility that there are nuances here, and that I’ve thought about this longer than they have, they would fire off a negative reply and then cease thinking about it, scrolling on or tabbing over to pornography or whatever it is people do to keep their id satisfied on the internet. I’d get annoyed and cause myself psychic damage for no gain.

Now, you’re all my prisoners. If you want the chart, you have to wade through 500 words of explanation that may or may not be all that clear to get to it.

This is another newsletter in that vein.

Elo

Elo ratings were developed by Arpad Elo to rank chess players and are now used in FIFA's official world rankings, among other applications.

I've been using Elo ratings since 2017 to estimate the quality of rugby league teams. Despite the lengthy introduction in part 1, the systems of Elo ratings haven’t changed much since then. The spreadsheet I use to calculate ratings is now efficient enough for Google Sheets to host but you won’t notice that.

The average rating is 1500 and a higher rating reflects a better team. We can use the difference in Elo ratings to estimate the probability of a given team winning against a given opponent. I maintain two systems for each competition:

• Form ratings are designed to reflect short term performance and move quickly to reflect recent results (about six to eight weeks in the NRLM). The system variables are optimised to maximise head-to-head tipping success. When two teams match up, an expected margin is calculated between the two teams based on their respective ratings. If a team beats the line, even if they lose the match, their rating goes up by exactly the same amount the other team's rating goes down. Form ratings only track regular season performance.

• Class ratings are slower moving than form ratings and take multiple seasons to change significantly. Class ratings reflect team's innate quality and act as a handbrake from looking too closely at the last couple of matches. Unlike form ratings, class ratings go up only with a win, the quantity determined by the difference in ratings. Winning finals games are weighted more heavily and grand finals more heavily still.

One change I did make this off-season was to calculate NRL Elo ratings back to 19881. While I will normally limit my analysis to the NRL era, it means that teams will start that period of analysis with a class rating that roughly reflects their actual class, instead of the standard initial rating of 1500.

The more clever readers will realise that this is a zero sum game. They may observe that the last three premiership winning clubs have all peaked in the last five years and then reason that might lead to some lopsidedness across the league. While class Elo is only a measure of relative quality, not absolute quality, that is a reasonable conclusion to reach.

Thank you for reading Stats Drop

Coaches

It's all well and good to rate players but what about the big dogs telling them what they’re supposed to be doing? I think there’s two ways to think about a coach’s impact on a team:

1. Are things better when they leave than when they started?

2. Are players out- or under-performing expectations?

We had previously attempted to measure these ideas with:

1. Changes in class Elo rating and this was the basis of The coaches that fucked up your club back in 2020.

2. Comparisons of pre-season projections to post-season metrics in what was called Coach Factor.

But now we replace them with what I am hoping are the slightly more comprehensible units of Thompson(s) and Bax(es).

Thompsons

We’ve talked a bit about Duncan Thompson previously. Thompsons are units to measure the change in class Elo rating of a team. Each point of change is one Thompson. A regular season game will be worth about five to seven Thompsons, depending on the teams’ ratings, positive if won and negative if lost. As we discovered in 2020, a loss of 50 Thompsons from the starting position is almost certainly a death sentence, although this is not the only way expectations can fail to be met.

Hopefully, we can all be adults about this and acknowledge the shortfalls in this approach without completely discarding it. The situation in which a new coach arrives has considerable impact on what is achievable. Wayne Bennett’s first stint at the Broncos is the lowest rated tenure, joined by Steve Folkes and Bob Fulton at the bottom of this table. They all had very good teams that could not be kept at that level over the length of their tenure. Bennett won three premierships, Folkes one and Fulton had overseen Manly’s three consecutive pre-NRL grand final appearances, so these aren’t exactly bums. There’s a lesson here about the impact of arbitrarily slicing and dicing a coach’s body of work, instead of letting the complete resume speak for itself.

More commonly, coaches come into a bad situation where the class rating is very low. Unless they actively sabotage their team through their incompetence, most teams will rise from a sub-1450 rating back to mean through the magic of mean regression, inflating the record of otherwise mediocre coaches.

To mitigate but not completely eliminate this effect, we’ll create an expected Thompsons (xTh) metric. Using linear regression, we’ll compare the final rating against the initial rating to get a sense of what mean regression will provide (R squared = 0.35) and try to separate that from what the coach provides. The difference between the actual Thompsons generated in the tenure/career and the expected Thompsons is Thompsons over expectation (Thox).

The effect of the mean regression is relatively mild but Thox has a better correlation with grand final wins (0.42) than straight change in Thompsons (0.27), although not as good as the correlation with the sheer number of appearances (0.61)2.

The problem looking at this metric over a career is that Thox is only calculated once for each tenure, based on the initial rating, and is blind to how long that tenure is. After a enough time, ups and downs start to cancel each other out but changing clubs every couple of years allows a coach to get a fresh shot at piling on some easy Thox.

This highlights the crudeness of these metrics and that we will have to accept one number isn’t going to do anyone’s work justice - player or coach - not least because it is difficult to separate the impact of the coach on their players and vice versa. But, again, we’re all adults and we can deal with some crudeness.

Baxes

We’ve also talked a bit about Bob Bax previously. Baxes are the difference between a player’s pre-season projected Z score3 and the player’s actual Z score of a match. A player with a projected Z of 150 that puts up 120 in a game has created -30 Baxes for his or her coach. So many Baxes are generated, we will frequently have cause to use kiloBax (kBx) for 1000 Baxes and hectoBax (hBx) for 100 Baxes.

In theory, good coaches help their players achieve their best so will consistently generate more Baxes than other coaches.

There may be many reasons for a discrepancy between a player’s projection and actual performance - injury, development, aging, career year syndrome, favouritism (or lack of) with the coach, familiarity or suitability (or not) with the coach’s scheme, form of teammates - and it may seem unwise to assign all of the potential effects to the coach4, when they have a staff and a roster to account for. My assertion is that these effects will wash out with quantity and time but there’s always room for more nuance.

There’s some scholarly work to be done to trace the performance of coaches between competitions using these metrics - put that in the pile with doing the same for player performance - but also whether coaches are subject to the same career year syndrome as players.

My instinct is that they are and that when things click into place for player or coach, the career year is very much a number of tail end probabilities all coming up as a royal flush. The distribution of these events, whether the same or different for players and coaches would be interesting to track.

Thank you for reading Stats Drop

Dashboards

So far we’ve talked exclusively about the NRLM, largely because its the most popular and familiar competition and has the deepest dataset to wallow around in. We also have four other competitions’ worth of data to work with and in processing it, have achieved one of my longer term goals of producing a nice looking dashboard with way too much information in it so you can make of the statistical situation what you will.

Here’s the history of the 70 minute era of NRLW:

Don’t even try to look at this on anything less than a 55” TV or preferably on a Minority Report-style hologram. Do not think that the email embed is doing this bad boy justice. I will probably just link to dashboards in future so you will be forced to click through and get the full experience.

There are explanations in the footnotes of the dashboard but here are some extended explanations of more familiar metrics that I’ve written about previously:

• Pythagorean expectation: There is a relationship between for and against and winning percentage, which is expressed by the Pythagorean expectation formula. This formula is used to estimate the number of wins a team should have based on their points scored and conceded. Generally, if a team outperforms their Pythagorean expectation (that is, wins more games than predicted by the formula) they will win fewer games in the following season and vice versa. Outperforming Pythag is not necessarily a good thing. Pythagorean expectation gets increasingly accurate at estimating win percentage over a longer time period.

• SCWP (Should-a Would-a Could-a Points): Using the principle of Pythagorean expectation but, in lieu of using actual points scored and conceded, utilises metres gained and conceded and line breaks gained and conceded to estimate the number of points that the team should have scored and conceded. The results are called "second-order wins", disctinct from "first-order wins" calculated by standard Pythagorean expectation and actual wins (sometimes "zeroth-order").

• Disappointment line (DSP): The number of wins we expect for each team based on their pre-season class rating, as a proxy of general expectations of what will be possible in the coming season. An average team will be expected to win half of their games, an above average team will be expected to win more than half and vice versa. Failing to reach the number of wins indicates a disappointing season and the greater the miss, the more disappointing the season has been.

Once I’ve completed similar dashboards for the QRLW and men’s competitions, we’ll be able to get a better idea of what metrics matter, which metrics converge to a single point of information (e.g. this team is bad at defence) and what metrics are sustainable year-to-year or subject to mean reversion or some other statistical impact.

I have a couple of other ideas for metrics (a panic index that tracks wooden spoon probabilities via Monte Carlo sims, a point-minute lead metric to replace WCL5) that may or may not get added as I get time to calculate them and automate its implementation in the dashboard.

Programming notes

Stats Drop will be making a regular return in 2025. I’m aiming to send two things out most weeks, which seems to be a broadly sustainable workrate. One will be the regular Wednesday newsletter, trimmed down a little, and the other will be from one of the other sections (Stats Drop, Pony Picayune, Bovine Bulletin) or something standalone.

That means Stats Drop will be roughly monthly-ish, maybe twice monthly if I have enough good ideas, time and energy. The next Stats Drop should be a return of the Deep Dive into the NRL season, which I haven’t done in three years but I have missed fussing over the numbers.

The Dataset™ will be made available in due course but it will only be available to paid subscribers. Paid subscriptions are not yet turned on, as I haven’t quite figured out what will be in front or behind the paywall. Subscriptions and other financial support will mostly subsidise a laptop with a functioning battery and more RAM.

I am using that due course to fill in some gaps - mostly event data for the NRL and QCup - that were delayed in the interests of first completing these posts, albeit I am already a bit behind when I wanted to publish (see lack of RAM) and we haven’t even gotten close to season kick-off. I am looking forward to a productive 2025.

Welcome to Stats Drop, an inundation of rugby league numbers.

I haven’t done a Stats Drop in a minute. There are reasons for that, which we’ll get to shortly, but you might be wondering why this looks the way it does. With the success of the Pony Picayune experiment1, I’m spinning Stats Drop off into its own section.

This means, if you were so inclined, you could just subscribe to this part of the newsletter, which is going to be more generally NRL-oriented than the other parts of the newsletter. This might be more to your taste than me regularly castigating Phillip Street for their the latest act of anti-Queensland perfidy, or you could dodge the numbers and stick to the perfidy. It’s up to you. Personally, I like the pontification and the quantification.

If you’re receiving this by email, a reminder that the Datawrapper embeds work best on desktop, then next best if you click through on mobile and least best as presented in the email.

You can read Part 2 of this introduction to Stats Drop metrics here.

Back in 2023, I sensed a problem. Greg Marzhew was rated really highly and Payne Haas was not. This might seem like an inherently anti-Knights/pro-Broncos bias to have, which I do, but Marzhew had accumulated 1.9 wins above reserve grade (WARG) to that point while Haas had 0.8.

Even accounting for Marzhew’s career year and a (hypothetical only) lull in Haas’, has Marzhew ever really been twice the player that Haas is? Over the course of most of a season? Irrespective of how we might choose to measure “twice”, I don’t think Marzhew was on pace for a $1.5 million season.

On closer inspection, it seemed the best middles, of which Haas was one of the most highly rated, and edges were producing at about the same total rate as the bench players, which really doesn’t make sense and merited some investigation.

It turns out if one of the key elements of a player rating system is running metres, and the denominator to assess whether players are out- or under-performing is based on the five previous seasons’ data but not what is happening right now, and the current season is being played at a significantly different rate (e.g. a change of 20% over just four seasons), then the stats are not going to track the eye test well.

The solution was not that complicated. We can use the current season’s data as the denominator to capture league-wide tactical shifts while accepting that the early season numbers will be wonky due to small sample size. I thought I’d also take on a few other issues on the periphery while doing this refurbishment but that took about six months of idle brain power to arrive at the broad strokes of what I wanted it to look like.

By then, I had fallen a long way behind on data collection, which then needed to be collected. The NRL has now posted player stats back to 1998, so obviously I had to go and get that as well. I then figured I may as well go back to first principles, recollect some data I already had (match and coaching details from Rugby League Project, mostly) to make sure it was correct and consistent and rebuild it all from scratch. Once collected, the data had to be processed, regressed, analysed and formatted. I effectively did this last bit twice. All of that took a long time, not least because my laptop kept running out of memory.

This post, and the next one, is what came out of that.

Thank you for reading Stats Drop

Abstraction

To restate the aim, we are trying to build a model of rugby league that uses statistics to provide interesting insights that can complement, confirm or correct the eye test. We are not going to be right but we are, to borrow a hat tip from friend of the newsletter Rugby League Eye Test, going to be less wrong.

If we’re going back to first principles, let’s really get back to first principles:

1. Waves ripple through the quantum fields that make up our universe.

2. These quantum interactions present as a stadium, field, players, referee and ball at a macroscopic level, which will then play out as a game of rugby league in accordance with both the laws of the game and of physics at a rate of one second per second.

3. Photons will bounce off various surfaces and enter an observer’s retina, creating electrical impulses which are in turn interpreted by their brain on a slight delay, causing all sorts of cognitive responses (e.g. disgust if watching the Tigers, horror if watching the Panthers, etc).

4. A different set of photons will enter a camera lens, which will undertake an in-principle similar electronic processing of those photons but present them as digital bits of information which will be transmitted, stored and displayed on various computers or TV monitors.

5. Later, someone will pull up that information, play it as a video and put it through some software and produce some numbers that represent elements of play.

6. Those numbers will be posted on the internet.

7. An intrepid newsletter writer will copy and paste those numbers into a spreadsheet over an off-season and build a statistical model of how rugby league works.

Modelling rugby league at a quantum level is conceptually possible but practically not. I’m not going to estimate how many atoms and electrons are involved in a game but will just assume it’s a big number and it only gets exponentially more complicated from there. If you’d like to know more, see Sean Carroll’s Quanta and Fields, but that’s how a perfect mirror of reality would be built.

Fortunately, no one needs this level of accuracy (less Heisenberg’s uncertainty), although some lunatics might insist that if you’re not going to have this level of accuracy, then it’s a waste of time. Those people are quantum weenies. We can still glean insight, provided we understand what we’re working with.

The process of abstraction removes information - the colour of the home team’s jersey is recorded on video and in visual memory, but not in statistical data - in exchange for the potential for comprehensibility. Abstraction omits data and adds artefacts to the data as it proceeds through the levels. An observer might not be able to see what’s happening in the opposite corner at level 3. The frame rate and resolution of the camera might not be sufficient to see exactly when the ball touched a specific blade of grass at level 4. Video compression and pixelation or a coding error might turn nine metres into ten at level 5.

There are a number of issues introduced at level 6, which you would think might be relatively straightforward but are illustrative. The 1998 stats are borderline non-existent and where they are, they are untrustworthy. The most egregious example was the Paul Carige game having Carige listed as making one (1) error. The player stats for pre-2013 grand finals, which looks like someone tried to take some care with, are in worse shape than the rest of the season’s data. Any game involving the Gold Coast Chargers or Northern Eagles does not show the away team’s stats but it only affects those two teams. Because of the changes to the field goal scoring, field goals are now recorded as 1 point or 2 point, except no one went back to fix the old games, which were tagged as just ‘field goal’, so these don’t show up in the scoring summaries. Sometimes the score shown on draw page is not the same as the score shown on the match page. I don’t think its right to say that the Eels played their home games in 2001 at Bankwest Stadium or that the Northern Eagles played at Lottoland.

Some of this might make clear why supporting Rugby League Project is so important but a lot of it looks like the typical sort of care and attention to detail that we’ve come to expect from the PVL NRL (see thesis #7). I naturally tend to ‘if you’re not going to do it properly, don’t do it at all’ dad logic but I also don’t want someone to do something silly like deciding to following through and taking it all away. Most of the data is fine but a few things need fixing up.

With all of that in mind, we can still build a statistical model that does a passable job of representing reality. The compilation of various NRL, QRL and NSWRL stats with data from Rugby League Project and a light sprinkling of Wikipedia comprises The Dataset™. The Dataset™ is currently a 66.8 MB Excel spreadsheet and is, somehow, still not complete but covers five southern hemisphere competitions: NRLM, NRLW, Queensland Cup, QRLW premiership and NSW Cup.2 This sounds impressive but is only useful if we understand its limitations.

We’re going use The Dataset™ to build some useful metrics to analyse the game. Today, we’re looking at the players. Next time, we’ll look at coaches and teams.

Thank you for reading Stats Drop

Production

We have something like 185,000 lines of player statistics (to one degree or another) over the history of the NRLM. Add another 50k for each of the state cups and 5k for each of the women’s competitions and that’s a near overwhelming amount of data to turn into something useful.

Some of those statistics correlate with winning. The canonical selection from the NRL.com suite are tries, running metres, line breaks, line break assists, try assists, tackle busts, kicking metres and, inversely correlated, errors and missed tackles.

Having done this for so long, I don’t particularly want to relitigate why these statistics matter other than to note this is a slimmed down set from predecessor systems, which included statistics like hit ups and dummy half run metres that also correlate but are kind of double-counting running metres.

Accumulation of the statistics that correlate with winning is called production.

Not every player contributes in the same way on the field but they all function as parts of the team and they all affect the result of the match. Few props score tries but many wingers do. Does that make wingers better players or more valuable than props? We need a reference unit to which we can convert different statistics to allow for comparison of production. We call that unit the Taylor (Ty for short).3

There are, unfortunately, many ways to convert one number into another. Here are two that I will use.

FTy

FTy are Taylors created by playing footy (that's the F). If a player does good (bad) stuff, he or she gets credit (docked) for it. That's what playing footy is all about.

To establish the correlation between the statistics and the outcome, we put games into buckets. Each bucket represents a performance in a given category in a given season. For example, there were 56 instances of teams scoring exactly five tries in a game in the 2005 season. Of those, 40 won their matches. That bucket has a winning percentage of .714. Repeat for ten buckets (zero, one try, two tries, etc) in each statistical category (tries, assists, breaks+busts, etc), with enough games to meet the minimum, for each season and do a linear regression4 between the quantities of the statistic and the winning percentage of the buckets. The variables that make those statistics have more or less impact in a specific game situation will even out over a long enough time period5, although no player rating system is immune to chronic stat padding.

The linear regression gives a gradient of slope, which we will use to convert the categories into Taylors, and a coefficient of correlation, which we will use to weight each of the categories against each other, with a heavier weighting given to higher coefficients, when adding up the categories. This is the same technique that has been used since 2019.

A similar chart could be produced for the other categories but it wouldn't look as neat. In some seasons, categories like kicking metres and errors, have low to very low correlation to winning and so lots of kicking metres or errors are required to have a measurable impact. In most seasons, these things matter.

The advantage of this approach is that it is relatively simple, which makes it robust. Production is calculated independently of listed position or time on field, which means if this data isn’t available, we can still do a reasonable job of calculating something like FTy.

We only have minutes played for 2001 onwards, which I assume is related to changes to the interchange rules from the Super League era to what we are more familiar with today. Some categories are not available until 2003. But we can still pull together something for this dark age of the NRL's statistical history, should someone be interested in knowing the third most productive second rower in season 2000 (it was Gorden Tallis, 24.2 FTy).

Summing each player’s pre-game average FTy per appearance in that season will yield a head-to-head tipping percentage of 60.0%6 in NRLM regular season games (excluding round 1, n=4642) and using a career average will bump that up to 61.1%. State cup rates of return are a similar, about a percentage point either way, but closer to 70% in women’s competitions.

However, we are not that concerned with this specific kind of prediction, although I will dabble in it. Geniuinely good players will get more game time and more opportunities through the coach’s tactics and so naturally generate more production. It takes on the characteristic of a circular argument. We cannot measure the efficiency or the panache or the effectiveness with which the production was generated - you have to watch the games for that - but over time, accumulating more tries, assists, breaks, metres and minimising errors and missed tackles are some of the hallmarks of what we consider good players.

PTy

In the land of 100 metre forwards, the 120 metre forward is king. FTy measures production in a brute force way, the accumulation of raw numbers trumps all and 120 metres is 20% better than 100 metres. PTy attempts to identify that extra 20 metres as the important bit and minimises the first 100 metres as unimportant. If every player can run for 100 metres, then the game will be decided in the yardage above that threshold. If every player can score a try, then the result of the game will rest with the player that can score a second try.

Mathematically, PTy is somewhat simpler to derive than FTy. We calculate the margin on the scoreboard but also for the margin of tries, running metres (broken into categories of pre-contact, return and post-contact where available), errors, kicking metres, missed tackles, assists (line break and try treated separately), line breaks and tackle busts. Within each season, we regress the margin on the scoreboard to the margin in each statistical category to get a gradient of slope and coefficient of correlation, and we then build up the exchange rates for each season in a similar fashion to FTy.

The difference is the player inputs. Instead of working with the raw quantity, we calculate the average try scoring rate per minute played at that position and subtract that from the try scoring rate per minute played by each player in each game (minimum 25% of average position game time) and multiply that by the exchange rate. In this way, we highlight the contributions above average7 per time played at that position, which is where the outcome of the game can really swing.

A player that does nothing in the categories that correlate with winning will rate 0 FTy but approximately will rate -2 PTy for the same performance. Playing to the average level rates 0 PTy but roughly +1 FTy. FTy and PTy have a coefficient of correlation of 0.5 for the NRLM. This is not surprising given the input data is largely the same, but it does show that somewhat different things are being measured.

Z score

In and of themselves, PTy and FTy are not that easy to understand, which makes them less useful for communicating to an audience. It'd be nicer to repackage that information in a single number that is more intuitive. If we set the average production in a game (in both F and P) at that position as 100, then express production as a ratio of that average, capped at -250 and +250, then average the F and P components, then that will do the job. A score of 70 is 70% of average. 120 is 20% better than average. Negative numbers bad. Big number is better number.

We will call this the Z score (or just Z, I forget why I picked Z). Z is a rate statistic of production per game relative to a positional average, in contrast to Taylors which measure volume of production with differing yardsticks.

Z is mostly useful for assessing performance in individual games or production across a season. It is similar in many respects to TPR, the previous player rating for rate of production, but instead of looking like a batting average, it will look more like weighted runs created.

Z scores also form the basis of player projections, which set a mean regressed expectation of what production we might expect out of a player based on their previous one, two or three seasons of play. While there’s still a bit of work to be done here, it’s largely the same as it was in 2019, two player rating systems ago. The mean absolute error between projections and actual performance is in the ball park of 20 points on the Z score (on average, a player projected at 90 will land somewhere between 110 and 70). The dream is to be able cross-project Z scores between competitions but I haven’t gotten around to that in five years, so I’m not sure when I’m going to get to it.

Here’s every NRL player for which we have data (i.e. no 1998, no away teams for games involving the Northern Eagles), ranked by the number of games they had with a Z score of +250 (the maximum allowed). Of the 165,843 Z scores currently calculated, only 5.8% of them are +250, so it’s something like getting a 10/10 by production. The absolute best players by this metric are getting that in every four appearances, which is astonishing, even if this way of looking at things is not particularly meaningful.

Replacement level

Replacement level is a concept borrowed from baseball sabermetrics. For our purposes, a replacement level player is the cheapest player that is readily available that can play regular minutes in first grade. This is not quite the same as being the best player in state cup or being the worst player in the NRL. There are below replacement level players in the NRL and above replacement level players in QCup because the market is not frictionless.

An average player is more valuable than that tag would imply, and there are far more below average players than average or above average players. If a team’s star goes down with an ACL injury, they are unlikely to be replaced by an average player. It is far more likely that a replacement level player will be filling in. That perspective changes both assessment of the match probabilities and how one goes about pricing production. See also: Moneyball.

If it helps, Richard Swain, Casey McGuire and Rory Kostjasyn are three men’s players with over 100 NRL appearances and close to zero wins above reserve grade (replacement). I don’t remember Casey McGuire’s career all that clearly, other than he was one of those Broncos, but Kyle Flanagan, Drew Hutchison and Jaeman Salmon are more recent examples in the same range with fewer appearances. Danny Levi used to be the poster boy for replacement level but he’s put together a few tenths over the last few seasons and has moved away from zero.

The previous version of wins above reserve grade axiomatically stated that a team of replacement level players would win two games on a 24 game first grade schedule. The number of NRL/ARL/NSWRL teams since 1988 that fell below this threshold is roughly the same as the proportion of teams that fell below replacement level (.320) in the MLB since 1901. We will maintain this but use it in a more cosmetic, and less fundamental, way.

Previously, we would use linear regression to estimate what kind of production a 2-22 team would generate and use that to set replacement level. This would work out to around .055 to .070 TPR in the old money. Now, we take the number of teams in the league multiply by 17 and the player at that position in terms of FTy/game and PTy/game (for a minimum number of games played) is deemed the replacement level cutoff. Everyone above him or her is deemed above replacement level. E.g. in a 16 team NRL, the replacement level is set at the 272nd best player as rated by FTy/game and PTy/game with a minimum of 10 games played.

While this is still mostly a measure of relative quality, and less a measure of absolute quality, and this might seem like complete gibberish, what the graph says is: there are relatively a lot of replacement level guys lying around. We can vary the number of teams in the NRL and the threshhold for what we consider regular minutes and the value that comes in for replacement level is fairly consistent over time8.

On the other hand, what constitutes the most productive in a given year fluctuates quite a lot, depending on individual player form, the spread of competence within his or her team and across the league and the nature of the sport in that season. As we know from experience, these factors can vary significantly.

When one is assessing the quality of any sporting enterprise, it pays to think not just about the best but also what the yardstick is. If the yardstick is pretty well constant, then when club bosses like to complain about dilution of existing talent and widening the gap between have and have nots, we can conclude they’re really complaining that there aren’t enough 2021 Tom Trbojevicii to go around to keep their jobs secure.

This perspective ignores that the rarity of the star player is what makes the star valuable. Aligning the stars is part, maybe the most important part, of the job of a club boss if everyone has a star, no one does. There is only one 2021 Tom Trbojevic because even Tom Trbojevic isn’t 2021 Tom Trbojevic. They can’t all have him and someone has to lose but filling in the bottom half dozen spots on the roster should be a piece of cake.

Wins Above Reserve Grade (WARG)

In the previous system, the number of WARG calculated in a given season would vary dramatically, despite the fact that there are basically the same number of wins above reserve grade available every year (number of regular seaosn games played minus replacement level team wins). If production inflates, as it did in 2021, and the bar to clear for an average performance moves from 1400 metres per game to 1800, then it doesn’t necessarily mean that the players in 2021 were historically special, generating twice as many WARG than 2013’s cohort. While this is one way of looking at it, I think it is better to think of it as those metres being diluted in value because they’re chasing the same number of wins.

Accordingly, I changed the way WARG were allocated. Each player’s production above replacement level, as a proportion of the whole league’s production above replacement level, is multiplied by the number of available wins above reserve grade. For example, in 2024, the NRLM had 202 available wins, the NRLW had 44.3, QCup 148.3. A very productive player might generate 0.5% of the league production over replacement and so would be entitled to 0.5% of those wins. It’s that simple.

WARG is a volume statistic, useful for considering production over a season or a career. A career with 6 WARG is good enough to be in the top 10% of all players with a NRLM appearance (now familiar Dataset™ disclaimers apply) and more than 19 is required to be in the top 0.5%. That is, in no particular order, Darren Lockyer, Matthew Bowen, Paul Gallen, Cameron Smith, Johnathan Thurston, Billy Slater, Robbie Farah, Benji Marshall, Cooper Cronk, Greg Inglis, Jarryd Hayne, Daly Cherry-Evans and James Tedesco at the end 2024.

We can account for variations in number of appearances by looking at Single Season Pace (WARG divided by games played times regular season games). The equivalent top tier of 0.5% with at least 50 appearances and a SSP of at least 2.3 WARG per 24 games, are Darren Lockyer, Andrew Johns, Brad Fittler, Jarryd Hayne, Tom Trbojevic and Ryan Papenhuyzen.

I give the PTy generated over replacement three times the weighting of FTy production in allocating WARG but this is a relatively arbitrary number that I suspect does not matter much but I have not spent time fine tuning as yet. If the chart below changes minorly, that is because I have altered some variables.

A related issue I had sensed in 2023, but did not recount in the introduction, was compiling a list of highest rated players by WARG over the previous decade that featured Anthony Milford in the top ten. I could see top twenty or thirty, maybe, as he was playing well until covid, especially in the years of covering for declining Darius, but top ten? He’s now top 35 over that decade, which feels more appropriate.

Incidentally, Haas finished 2023 with 1.5 new WARG and Marzhew 2.2, which feels closer to reality, if still a strange legacy of that year’s super-yardage outside backs.

A very simple Stats Drop this week: what are the longest streaks - hoodoos, if you will -in the NRL?

I removed the Dolphins and defunct clubs from consideration for fairly obvious reasons. Still, congratulations to the Adelaide Rams, who haven’t been beaten by Penrith since round 4 of the 1998 season.

The only team to not beat the Broncos since covid are the Bulldogs, which probably says a lot about that program. The Tigers are in the same spot with respect to the Titans. The longest Cowboys-based hoodoo is the Raiders, who haven’t beaten North Queensland since round 12, 2020. It’s probably worth remembering that the Broncos, Titans and Cowboys have all been pretty bad at points in the last four seasons.

Of the 25 longest droughts, the Tigers have five and the Warriors and Titans have four each. The Broncos have one; no prizes for guessing whomst. The Cowboys have somehow beaten everyone - the Sharks in last year’s qualifying final being the last monkey to be removed from the back - since losing to the Broncos towards the end of last year. Except, that is, for the Dolphins.

Eight teams, nearly half the league (and nine if you include the Dolphins), haven’t beaten Penrith since their rise to the top of the Vlandoball era, seven of which are in the top 25 longest droughts. The Raiders, Rabbitohs, Roosters and Storm have three appearances each in that list.

Here’s the Queensland Cup equivalent list

It’s not surprising to see Redcliffe, Burleigh and Norths make up a large proportion of the right hand column or Ipswich in the left hand column (refer JETS WATCH).

Shout out to the Capras, who have the longest hoodoo against the Dolphins - it’s been a decade now - but also somehow haven’t lost to the Pride since 2019 (this is because 2020 was cancelled, they didn’t play in 2021 and then the Capras got good).

Trivia

• JETS WATCH: Still winless. There are so few games left that Elo is no longer forecasting any wins.

• Following their loss to the Warriors in round 14, the Dolphins’ hoodoo against Auckland dates back to the 1976 AMCO Cup.

• I think the Hoodoo Gurus’ “What’s My Team?” promotional campaign might be the only one I actually have any recollection of seeing. Too young for Tina, too old for pretty much anything after the Gurus. It doesn’t help that I no longer have a lot of clear memories of being a kid. There’s too much trivia in the way.

• The Tigers lost 74-0 to the Cowboys in round 18 and somehow the Bulldogs’ points difference is still worse (-213 and -261). If you were able to replace that loss with a nil-all draw, that would bring the Tigers’ points difference in line with the next worst in the NRL, the Dragons’ -139.

• The Bulldogs are currently losing by an average of 13.1 points per game. If continued to the end of the season, this would only be the 14th worst mark in NRL history. Last year’s Tigers lost by 13.5 points and the year before’s Bulldogs lost by 15.4. So while the 2023 versions are very bad, they have actually been worse in recent memory but not necessarily at the same time.

• The Panthers are winning at an average of 13.5 points per game, which is actually lower than both 2021 and 2020. It would be the ninth best rate, if continued through to round 27, in NRL history. Only above them in the list are Brisbane (1998), Parramatta (2001), Sydney (2004), Melbourne (2007, 2019, 2021) and Penrith. The gap to the next best team, the Broncos is 5.9 points per game. Comparable gaps between the best and next best happened in 2000 (BNE 11.8, NEW 5.9), 2001 (PAR 16.7, BNE 7.1), 2007 (MEL 14.6, MAN 9.2) and 2017 (MEL 12.4, BNE 6.8).

• The heat death of the universe sounds considerably less existentially terrifying than false vacuum decay.

NRL Stats Drop

Scroll to the end for the explanatory notes or visit How It All Works.

Queensland Cup Stats Drop

Explanatory notes

Elo Ratings

Elo ratings are a way of quantitatively assessing teams, developing predictions for the outcomes of games and then re-rating teams based on their performance, home ground advantage and the strength of their opposition. Form Elo ratings are optimised for head-to-head tipping and tend to reflect the relative strengths of each team at that particular point in time, although there are many factors that affect a team’s rating.

Pythagorean Expectation and Winning Percentage Comparison

The black dots on the winning percentage comparison are each team’s actual win-loss record to this point in the season. The coloured dots represent what the stats say about the team’s underlying performance, i.e. how many games they should be winning. Wins and losses are binary and can be prone to good and bad luck in a way that other stats that correlate to wins are not, so we have other metrics to help see through the noise to good teams, rather than just good results. While each metric has strengths and weaknesses, it helps paint a more complete picture of team performance.

Pythagorean expectation (gold) relies on points scored and conceded by estimating a team’s number of wins based on their for and against. Where there is a deviation between a team’s actual record and their Pythagorean expectation, we can ascribe that to good fortune, when a team wins more than they are expected to, or bad fortune, when a team wins less than they are expected to. 2nd order wins (silver) relies on metres and breaks gained and conceded. Elo ratings (maroon) rely on the margin of victory and strength of opponent.

Dots should tend to gravitate towards each other. If a team’s dots are close together, that means their actual results are closely in line with their underlying metrics and represents a “true” or “fair” depiction of how good the team is. If a team’s coloured dots are clustered away from their actual record, then we should expect the actual and the coloured cluster to move towards each other over time.

If the black dot is well above gold, that team is suffering from good fortune and may mean regress to more typical luck in the future (vice versa also holds). The silver dots will tend to hover around .500, so if gold is between silver and .500, the team could have an efficiency issue. On the balance, I would expect more often the actual percentage will move towards the cluster but the opposite is also possible.

Player Leaderboards

Production the amount of valuable work done by a team as measured in counting statistics that correlate with winning. These statistics are converted to a single unit called Taylors. Taylor Player Ratings (TPR) are a rate metric that compares an individual player’s production, time-adjusted, to that of the average player at their position, with a rating of .100 being average (minimum 5 games played). Wins Above Reserve Grade (WARG) is a volume metric that converts player’s production over a nominal replacement level into an equivalent number of wins they contribute to their team.

Team Strengths

A chart that tabulates each team’s WARG, both as a team total and broken down into platoons by listed position, to identify each team’s relative strengths and weaknesses.

The WARG is the sum produced at the position, not by specific players, e.g. if Kalyn Ponga produces a total of 1.5 WARG comprising 1.0 WARG while named at fullback and 0.5 WARG named at five-eighth, then the WARG produced while named at fullback will be added to the 1-5 platoon and the WARG created while named in the halves will be added to the 6+7+9 platoon.

The top team in each category is rated 100, the bottom team is rated 0 and every team in between is scaled accordingly.

Team Efficiencies

This table compares the SCWP produced and conceded by each team (a product of their metres and breaks gained and conceded) against the actual points the team scores and concedes to measure which teams are most efficiently taking advantage of their opportunities. A lack of efficiency here could be the result of bad luck and poor execution - sometimes you have to watch the games.

Thanks for reading The Maroon Observer. If you haven’t already, you can subscribe below to receive all the latest about Queensland rugby league.

Subscribe now

If you really enjoyed this, please forward the email on to someone who might also enjoy it.

Share

Clint Gutherson missed a game for his club1 so he could play Origin for three minutes2. Eels fans are naturally outraged about the injustice of losing one of their best players for an inconsequential3 cameo last Wednesday night.

One of my favourite topics at this time of year is looking at Origin caps by club. I’m going to pick on Mary Konstantopoulos of Ladies Who League this week, not out of spite as would usually be the case but because it’s convenient for the point I want to make:

The Penrith Panthers still sit on top of the NRL ladder despite missing so many players during the State of Origin period and having players out through injury.

But the Eels and the South Sydney Rabbitohs have faced challenges during this period and it may mean they miss out on a top-four spot. We know how important that top-four spot is if you have aspirations of winning a premiership so why should this period have such a big impact?

-Time to pause during Origin and reduce length of season as interstate contest leaves NRL in the shade

Have the Eels, the Panthers and the Rabbitohs been unfairly burdened by the duties of state representation?

Remember that the red numbers represent a genuine concern that seems to be widely held that Origin could possibly affect the premiership aspirations of those clubs. Ignore the green numbers of clubs also competing for a top four spot (and the Cowboys). Definitely don’t look at the gold numbers and draw any conclusions.

Origin has a randomising effect on the competition as a whole and tends to bring the pack closer together. More specifically, Origin has consistenly and negatively affected the Broncos, Cowboys and Roosters, as reflected by their reduced winning percentage from pre-Origin to the inter-Origin period. These three are rarely cited as clubs for whom we should be concerned about Origin’s impact on their premiership aspirations. I can’t imagine why.

This might be my Broncos fan privilege but I’ve always seen Origin selections as a fact of life. Sometimes there’s lots of selections, which usually happens when the team is going well, and sometimes there’s not many. The results will fall where they may and life will move on. Your team is better off with the Origin-calibre players than without.

Perhaps it’s that there’s a perception that Queenslanders would prefer an Origin series win to a premiership. I don’t think that’s true but, in any case, it’s a false dichotomy, as demonstrated in 1998, 2006 and 2015. Maybe the Broncos or Cowboys or Storm or Roosters would have won more titles without Origin’s influence, but it’s also extremely likely that rugby league would have gone bust without the rivers of gold the interstate rivalry generates. People forget that rugby union was ahead of league in the early 2000s and State of Origin has been a huge factor in league overhauling and then systematically dismantling the lesser rugby code in Australia. You can’t win a title in a competition that doesn’t exist.

It is never going to change. The NRL has experimented with non-Wednesday Origins (including an end of season series in 2020) and because mid-season Wednesday night maximises ratings, that’s what Nine wants and so the NRL complies. There isn’t going to be a world in which Origin doesn’t affect the club season because, while I too would like a shorter NRL season to open up space on the calendar for other high value games in lieu of the mid-season sludge we’re served, the NRL is not going to cut inventory for fear of losing broadcast revenue. You may as well accept it for what it is.

Trivia

• The last Origin dead rubber played in Sydney in 2016 attracted a crowd of just 61,000 to Stadium Australia. The dead rubber in the Blues’ 2000 sweep brought in just under 59,000. The 75,000 that turned up last Wednesday suggests the media’s lack of access to players has had precisely zero impact on fans engagement with the sport, a truly shocking turn of events for the ‘coverage is what makes Origin special’ crowd. Dead rubbers at Suncorp tend to still crack 50k.

• Sami Sauiluma is the greatest try scorer in Burleigh Bears’ history. His 72nd try, scored against the Hunters two weeks ago, pushed him past previous record holder Kurtis Rowe. He scored his 73rd against the Dolphins in Miles on the weekend, the Bears’ only try of the afternoon.

• My hunch was that with 17 teams in the comp, making the top eight would require a minimum 13-11 record. With the disparity between the very bottom teams (Tigers, Dragons, Bulldogs) and the lengthy peloton strung out behind the Panthers, the price of admission to the post-season might be 14 wins (34 competition points, including three byes).

• Since the NRL moved to a 24 game schedule in 2002, the highest barrier to finals entry was 2018, with a minimum 15-9 record required to the make the top eight. That year, the Tigers finished 12-12 in ninth, which would normally be enough, although their points difference of -83 makes that hypothetical more marginal.

• The minor premier in 2018 (Roosters) only had 16 wins, the lowest number in the NRL era, equal with 2014 (also Roosters) and 2005 (Eels). In ‘14 and ‘05 only two teams finished 16-8. In ‘18, four teams finished 16-8.

• The world is a lot more flammable than you think thanks to climate change and a dependence on petroleum products.

NRL Stats Drop

Scroll to the end for the explanatory notes or visit How It All Works.

Queensland Cup Stats Drop

Explanatory notes

Elo Ratings

Elo ratings are a way of quantitatively assessing teams, developing predictions for the outcomes of games and then re-rating teams based on their performance, home ground advantage and the strength of their opposition. Form Elo ratings are optimised for head-to-head tipping and tend to reflect the relative strengths of each team at that particular point in time, although there are many factors that affect a team’s rating.

Pythagorean Expectation and Winning Percentage Comparison

The black dots on the winning percentage comparison are each team’s actual win-loss record to this point in the season. The coloured dots represent what the stats say about the team’s underlying performance, i.e. how many games they should be winning. Wins and losses are binary and can be prone to good and bad luck in a way that other stats that correlate to wins are not, so we have other metrics to help see through the noise to good teams, rather than just good results. While each metric has strengths and weaknesses, it helps paint a more complete picture of team performance.

Pythagorean expectation (gold) relies on points scored and conceded by estimating a team’s number of wins based on their for and against. Where there is a deviation between a team’s actual record and their Pythagorean expectation, we can ascribe that to good fortune, when a team wins more than they are expected to, or bad fortune, when a team wins less than they are expected to. 2nd order wins (silver) relies on metres and breaks gained and conceded. Elo ratings (maroon) rely on the margin of victory and strength of opponent.

Dots should tend to gravitate towards each other. If a team’s dots are close together, that means their actual results are closely in line with their underlying metrics and represents a “true” or “fair” depiction of how good the team is. If a team’s coloured dots are clustered away from their actual record, then we should expect the actual and the coloured cluster to move towards each other over time.

If the black dot is well above gold, that team is suffering from good fortune and may mean regress to more typical luck in the future (vice versa also holds). The silver dots will tend to hover around .500, so if gold is between silver and .500, the team could have an efficiency issue. On the balance, I would expect more often the actual percentage will move towards the cluster but the opposite is also possible.

Player Leaderboards

Production the amount of valuable work done by a team as measured in counting statistics that correlate with winning. These statistics are converted to a single unit called Taylors. Taylor Player Ratings (TPR) are a rate metric that compares an individual player’s production, time-adjusted, to that of the average player at their position, with a rating of .100 being average (minimum 5 games played). Wins Above Reserve Grade (WARG) is a volume metric that converts player’s production over a nominal replacement level into an equivalent number of wins they contribute to their team.

Team Strengths

A chart that tabulates each team’s WARG, both as a team total and broken down into platoons by listed position, to identify each team’s relative strengths and weaknesses.

The WARG is the sum produced at the position, not by specific players, e.g. if Kalyn Ponga produces a total of 1.5 WARG comprising 1.0 WARG while named at fullback and 0.5 WARG named at five-eighth, then the WARG produced while named at fullback will be added to the 1-5 platoon and the WARG created while named in the halves will be added to the 6+7+9 platoon.

The top team in each category is rated 100, the bottom team is rated 0 and every team in between is scaled accordingly.

Team Efficiencies

This table compares the SCWP produced and conceded by each team (a product of their metres and breaks gained and conceded) against the actual points the team scores and concedes to measure which teams are most efficiently taking advantage of their opportunities. A lack of efficiency here could be the result of bad luck and poor execution - sometimes you have to watch the games.

Thanks for reading The Maroon Observer. If you haven’t already, you can subscribe below to receive all the latest about Queensland rugby league.

Subscribe now

If you really enjoyed this, please forward the email on to someone who might also enjoy it.

Share

When you really think about it, homeground advantage is weird. There’s plenty of data to prove that it exists. Covid showed that it’s at least partly the crowd. Travel for the away team also seems to be a component. The precise ratio of these factors and the impact of others, like familiarity with the ground, is not always clear. Whether the advantage is constant through time, or variable, has probably not been tested. From year-to-year there’s a large amount of inexplicable noise1.

While over the long run, playing at home is worth about a try all other things being equal, trying to determine which teams have a greater homeground advantage than others is difficult.

The noise is made louder by the performance of the individual teams in question. A very good team wins both home and away games, largely in equal measure, which tells you nothing about their home ground advantage. A very bad team loses both home and away games, which doesn’t offer many clues as to whether their home results would be worse without the in-built advantage. Everyone in between wins and loses in differing ratios, of which homeground advantage has some, but not all, explanatory power.

For example, we would expect that North Queensland enjoys a significant homeground advantage due to a decent attendances, a high level of parochiality and Townsville is a long way from anywhere else. But if the Cowboys are bad (or good!) for many years, then how much of their results at home can be attributed to the home ground advantage and how much to the team itself?

To cut through some of the noise caused by performance, I looked at the long term ratio of wins collected at home as a proportion of a team’s total wins.2

There’s something here. We would expect the Warriors and Hunters to have the toughest road trips in their respective leagues but also probably struggle most with the travel demands for their away games.

But there are oddities. Mackay and Townsville are much lower than we’d expect given their relative remoteness, although they are at different ends of the performance spectrum. Manly and St George Illawarra are much higher than we’d expect, given their location in the eastern half of Sydney and most of their rivals having less than two hours on the road to reach them.

It’s also strange to contrast the Broncos, who have exactly one home ground and have occupied that ground since 2003 and it’s so popular, other teams take their home games there, and the Tigers, who have multiple home grounds, both in Sydney but also in regional New South Wales and, sometimes, New Zealand. Despite all these differing factors, their respective ratios are pretty much identical.

Perhaps this is a mystery that will never be fully solved.

Trivia

• People aren’t high enough on the Eels and are too high on the Warriors. Elo don’t lie3.

• Canberra are currently on pace to finish with 3.3 wins over their Pythagorean expectation. Only the 2003 Panthers (+3.5), 2006 Storm (+3.6) and 2007 Cowboys (+4.4) had bigger outperformances. Two of them topped the ladder and one won a premiership. There are seven teams that underperformed by a greater margin, the best being the 2002 Eels and the most recent, the 2008 Cowboys. We haven’t seen a discrepancy like this in 15 years.

• Ipswich are currently winless through 15 games. One of the footnotes on Monday covered the six winless teams in Queensland Cup history but the Jets are projected by Elo to win one (1) game. There have been 8 one-win teams in Queensland Cup: 96 and 14 Sunshine Coast (1-14 and 1-23 respectively), 97 Ipswich (1-16-1), 99 Souths (1-21) and 05, 15, 19 and 21 Central (1-17-2, 1-22, 1-21-1 and 1-14-2 respectively).

• Tony Stewart killed a guy.

NRL Stats Drop

Scroll to the end for the explanatory notes or visit How It All Works.

Queensland Cup Stats Drop

Explanatory notes

Elo Ratings

Elo ratings are a way of quantitatively assessing teams, developing predictions for the outcomes of games and then re-rating teams based on their performance, home ground advantage and the strength of their opposition. Form Elo ratings are optimised for head-to-head tipping and tend to reflect the relative strengths of each team at that particular point in time, although there are many factors that affect a team’s rating.

Pythagorean Expectation and Winning Percentage Comparison

The black dots on the winning percentage comparison are each team’s actual win-loss record to this point in the season. The coloured dots represent what the stats say about the team’s underlying performance, i.e. how many games they should be winning. Wins and losses are binary and can be prone to good and bad luck in a way that other stats that correlate to wins are not, so we have other metrics to help see through the noise to good teams, rather than just good results. While each metric has strengths and weaknesses, it helps paint a more complete picture of team performance.

Pythagorean expectation (gold) relies on points scored and conceded by estimating a team’s number of wins based on their for and against. Where there is a deviation between a team’s actual record and their Pythagorean expectation, we can ascribe that to good fortune, when a team wins more than they are expected to, or bad fortune, when a team wins less than they are expected to. 2nd order wins (silver) relies on metres and breaks gained and conceded. Elo ratings (maroon) rely on the margin of victory and strength of opponent.

Dots should tend to gravitate towards each other. If a team’s dots are close together, that means their actual results are closely in line with their underlying metrics and represents a “true” or “fair” depiction of how good the team is. If a team’s coloured dots are clustered away from their actual record, then we should expect the actual and the coloured cluster to move towards each other over time.

If the black dot is well above gold, that team is suffering from good fortune and may mean regress to more typical luck in the future (vice versa also holds). The silver dots will tend to hover around .500, so if gold is between silver and .500, the team could have an efficiency issue. On the balance, I would expect more often the actual percentage will move towards the cluster but the opposite is also possible.

Player Leaderboards

Production the amount of valuable work done by a team as measured in counting statistics that correlate with winning. These statistics are converted to a single unit called Taylors. Taylor Player Ratings (TPR) are a rate metric that compares an individual player’s production, time-adjusted, to that of the average player at their position, with a rating of .100 being average (minimum 5 games played). Wins Above Reserve Grade (WARG) is a volume metric that converts player’s production over a nominal replacement level into an equivalent number of wins they contribute to their team.

Team Strengths

A chart that tabulates each team’s WARG, both as a team total and broken down into platoons by listed position, to identify each team’s relative strengths and weaknesses.

The WARG is the sum produced at the position, not by specific players, e.g. if Kalyn Ponga produces a total of 1.5 WARG comprising 1.0 WARG while named at fullback and 0.5 WARG named at five-eighth, then the WARG produced while named at fullback will be added to the 1-5 platoon and the WARG created while named in the halves will be added to the 6+7+9 platoon.

The top team in each category is rated 100, the bottom team is rated 0 and every team in between is scaled accordingly.

Team Efficiencies

This table compares the SCWP produced and conceded by each team (a product of their metres and breaks gained and conceded) against the actual points the team scores and concedes to measure which teams are most efficiently taking advantage of their opportunities. A lack of efficiency here could be the result of bad luck and poor execution - sometimes you have to watch the games.

Thanks for reading The Maroon Observer. If you haven’t already, you can subscribe below to receive all the latest about Queensland rugby league.

Subscribe now

If you really enjoyed this, please forward the email on to someone who might also enjoy it.

Share

A couple of years ago, I looked at which clubs had had the most State of Origin caps. That yielded one of my favourite pieces of esoteric rugby league trivia that the Panthers took until 1995 to have as many State of Origin caps as Wynnum Manly had had by 1987.

Now that Redcliffe have joined the NRL, and Hamiso Tabuai-Fidow and Tom Gilbert are the first Dolphins to have been selected for Queensland since Greg Conescu and Bryan Niebling in 1987, I thought it was time to update the dataset.

Broadly, Queensland’s two strongest periods in Origin were the 80s, when there were multiple BRL clubs to draw upon, and the 10s, when there were multiple NRL clubs and the Storm to draw upon. It’s clear that we’re not returning to a 90s and early 00s level of Broncos dominance any time soon for a variety of factors (e.g. enforcement of the salary cap, other potential employers for players, generalised cheating by other clubs, etc) and that’s probably for the best from a whole of state perspective. Brisbane’s selections have at least bounced back from their 2020 wooden spoon lows to return to something close to the recent historical trend.

Interestingly, the club that is neither in Queensland nor in New South Wales alread has the fourth most all-time caps. The Storm will likely never be able to catch the Broncos but are 27 caps adrift of the Roosters’ second place. Despite the obvious geographical conundrum and not entering the competition until nearly 20 years into the concept, Melbourne’s players have been a noticeable presence in the Origin arena. That this has mostly come through Queensland is no accident, as is the popularity of the Storm north of the Tweed, although it can also be partly explained by their asterisk seasons.

Like most clubs, the Cowboys’ selections rise and fall in line with their performance on the field, peaking around 2005 and then the mid-10s Thurston-led North Queensland sides. It’s difficult to say whether the Titans’ caps mirrors their performance because the Gold Coast are more often poor than not. It’ll be interesting to monitor how their respective contributions change with their pathways plans.

Trivia

• The selections of Gilbert and Tabuai-Fidow have pushed the Dolphins (44) past the Steelers (43) into 20th place. There’s a chance they pass the Wests Tigers (48) in the next couple of years into 19th. Then begins the chase to Wynnum Manly’s 70 caps.

• The BRL’s Big 8 clubs have a combined 180 caps, which would be 12th on the all-time list, between St George Illawarra and Cronulla. That’s not bad for a bunch of clubs that no caps for 35 series and are considered irrelevant to the history of “the game”.

• Between 1980 and 1987, there were 300 Queensland Origin caps. Of those, 177 were occupied by BRL players, or an average of 59%. That peaked in 1983 when only ten caps were given to interstate players and the remaining 35 were held by local players. I’ll wait here while you look up the results of that series and ponder whether the received narrative about the relative strengths of the Brisbane and Sydney competitions is accurate or simply based on the last available example.

• Rohan Hancock was picked out of the Toowoomba competition to play for the Maroons in the 1980, 1981 and 1982. Greg Conescu was playing for Brothers Gladstone in 1984, the last player to be selected for Queensland from the bush.

• 17 years before Alfie was brought home for one last mission, Bob Kellaway was selected from Bradford Northern to play in game 3 of the 1984 series.

NRL Stats Drop

Scroll to the end for the explanatory notes or visit How It All Works.

Queensland Cup Stats Drop

Explanatory notes

Elo Ratings

Elo ratings are a way of quantitatively assessing teams, developing predictions for the outcomes of games and then re-rating teams based on their performance, home ground advantage and the strength of their opposition. Form Elo ratings are optimised for head-to-head tipping and tend to reflect the relative strengths of each team at that particular point in time, although there are many factors that affect a team’s rating.

Pythagorean Expectation and Winning Percentage Comparison

The black dots on the winning percentage comparison are each team’s actual win-loss record to this point in the season. The coloured dots represent what the stats say about the team’s underlying performance, i.e. how many games they should be winning. Wins and losses are binary and can be prone to good and bad luck in a way that other stats that correlate to wins are not, so we have other metrics to help see through the noise to good teams, rather than just good results. While each metric has strengths and weaknesses, it helps paint a more complete picture of team performance.

Pythagorean expectation (gold) relies on points scored and conceded by estimating a team’s number of wins based on their for and against. Where there is a deviation between a team’s actual record and their Pythagorean expectation, we can ascribe that to good fortune, when a team wins more than they are expected to, or bad fortune, when a team wins less than they are expected to. 2nd order wins (silver) relies on metres and breaks gained and conceded. Elo ratings (maroon) rely on the margin of victory and strength of opponent.

Dots should tend to gravitate towards each other. If a team’s dots are close together, that means their actual results are closely in line with their underlying metrics and represents a “true” or “fair” depiction of how good the team is. If a team’s coloured dots are clustered away from their actual record, then we should expect the actual and the coloured cluster to move towards each other over time.

If the black dot is well above gold, that team is suffering from good fortune and may mean regress to more typical luck in the future (vice versa also holds). The silver dots will tend to hover around .500, so if gold is between silver and .500, the team could have an efficiency issue. On the balance, I would expect more often the actual percentage will move towards the cluster but the opposite is also possible.

Player Leaderboards

Production the amount of valuable work done by a team as measured in counting statistics that correlate with winning. These statistics are converted to a single unit called Taylors. Taylor Player Ratings (TPR) are a rate metric that compares an individual player’s production, time-adjusted, to that of the average player at their position, with a rating of .100 being average (minimum 5 games played). Wins Above Reserve Grade (WARG) is a volume metric that converts player’s production over a nominal replacement level into an equivalent number of wins they contribute to their team.

Team Strengths

A chart that tabulates each team’s WARG, both as a team total and broken down into platoons by listed position, to identify each team’s relative strengths and weaknesses.

The WARG is the sum produced at the position, not by specific players, e.g. if Kalyn Ponga produces a total of 1.5 WARG comprising 1.0 WARG while named at fullback and 0.5 WARG named at five-eighth, then the WARG produced while named at fullback will be added to the 1-5 platoon and the WARG created while named in the halves will be added to the 6+7+9 platoon.

The top team in each category is rated 100, the bottom team is rated 0 and every team in between is scaled accordingly.

Team Efficiencies

This table compares the SCWP produced and conceded by each team (a product of their metres and breaks gained and conceded) against the actual points the team scores and concedes to measure which teams are most efficiently taking advantage of their opportunities. A lack of efficiency here could be the result of bad luck and poor execution - sometimes you have to watch the games.

Thanks for reading The Maroon Observer. If you haven’t already, you can subscribe below to receive all the latest about Queensland rugby league.

Subscribe now

If you really enjoyed this, please forward the email on to someone who might also enjoy it.

Share

It bugs me when a NRL commentator rattles off a piece of statistical trivia and says it’s the worst/best in the “history of the game”. The “game”, in this context and by extension the context used by most people, is the Sydney comp from 1908 to 1994, the short lived ARL from 1995 to 1997, the NRL from 1998 onwards and maybe Australian Super League if they’re feeling generous.

It is an incredibly dull view of the history of rugby league. It is a view that I believe is promulgated because of Australian rugby league’s inherent Sydney-centrism but also the relative availability of records or the Sydney comp. This continuity of record keeping gives the NSWRL a sheen of historical legitimacy, much in the same way that societies that build monuments out of stone traditionally have had more cultural cachet with historians than societies that build their monuments out of more perishable materials.

Ultimately, emphasising one aspect of the sport’s history while ignoring the rest because it is convenient, only serves to flatten and obscure huge swathes of rugby league history, to the detriment of everyone.

In Brisbane, the 1974 floods destroyed records stored in the low-lying QRL/BRL offices on Castlemaine Street. To my knowledge, no one has yet compiled a full list of BRL results, although the team at Rugby League Project (subscribe to their Patreon) are currently working on it1.

Without these records, it is not as easy to point to the history of the sport other parts of the world - whether it be Brisbane, Auckland or the bush - and it is unfairly given a lesser status as a result, which then makes it hard to put the career of someone like Eric Weissel - Hall of Famer, 18 years playing in the Riverina, never played for a Sydeny club but made eight appearances for Australia - into some sort of context.

I won’t proceed much further because I find it difficult to articulate the nature of the problem, why it matters and come up with anything approaching a rational solution. However, I’ve never been a fan of “first grade” as the moniker, which should denote the highest level of game played in a geographical area (e.g. first grade in Townsville and Sydney are different competitions in 1923 but in 2023, it’s just the NRL). My preference is to adopt the concept of “first class” matches from cricket - which has its own definitional problems - except that I think what constitutes first class is flexible and changes over time2.

I’ve made a start by compiling all of the ladders I could find for first division English rugby league (mostly from Wikipedia), the NSWRL (from RLP), the BRL (up to 1987, manually entered from Steve Haddan’s Our Game), the Super League and the NRL (both from RLP) into a single spreadsheet.3

Over 1.4 million points were scored in the top flight Sydney and Brisbane comps and the national English and Australian leagues from 1895 to 20224 across 61,000-odd matches.

While it’s not complete without finals results, the data are good for providing some appropriate historical context to numbers. For example, 2023 Penrith’s current rate of conceding 11.9 points per game seems impressive and is a mark only bettered by three NRL-era teams5. But 1944 Western Suburbs (Brisbane) conceded 117 in 10 games (11.7) while finishing second last on that season’s BRL ladder. The ‘44 Mud and Bloods were 5.5 wins clear of cross-town rivals and last placed Eastern Suburbs, who managed to concede 25.8 points per game. In fact, over a quarter of historical teams conceded fewer points per game. It doesn’t seem so impressive if clubs of the calibre of Ebbw Vale and Natives and Merthyr can better it.

Trivia

• In the 1940 BRL premiership, the Eastern Suburbs District Rugby League Football Club conceded 509 points in ten games for what was the worst defensive performance of any club known in first class rugby league. The next worst team, 1906/07 Liverpool City, conceded four points fewer per game at 46.6. 1898/88 St Helens conceded just 18 points in 26 games (0.7 points per game).

• The most points scored per game was St Helens in 1996 at 43.2. The 1917/18 Rochdale Hornets managed to score 1.0 points per game in a 7 game, and winless, season. Liversedge in 1900/01 had the worst non-wartime record of 43 points scored in 30 games (1.4 points per game).

• Valleys, including Valleys-Toombul, won 21 premierships between 1909 and 1987. Souths, Wests and Wynnum-Manly received 13 wooden spoons each (excluding their progenitor clubs).

• After 13 games, the ‘23 Dolphins are 7-6 (-7). The ‘88 Broncos were 9-4 (+71), the ‘95 Cowboys were 1-12 (-230) and the ‘07 Titans were 7-6 (+1).

NRL Stats Drop

Scroll to the end for the explanatory notes or visit How It All Works.

Queensland Cup Stats Drop

Explanatory notes

Elo Ratings

Elo ratings are a way of quantitatively assessing teams, developing predictions for the outcomes of games and then re-rating teams based on their performance, home ground advantage and the strength of their opposition. Form Elo ratings are optimised for head-to-head tipping and tend to reflect the relative strengths of each team at that particular point in time, although there are many factors that affect a team’s rating.

Pythagorean Expectation and Winning Percentage Comparison

The black dots on the winning percentage comparison are each team’s actual win-loss record to this point in the season. The coloured dots represent what the stats say about the team’s underlying performance, i.e. how many games they should be winning. Wins and losses are binary and can be prone to good and bad luck in a way that other stats that correlate to wins are not, so we have other metrics to help see through the noise to good teams, rather than just good results. While each metric has strengths and weaknesses, it helps paint a more complete picture of team performance.

Pythagorean expectation (gold) relies on points scored and conceded by estimating a team’s number of wins based on their for and against. Where there is a deviation between a team’s actual record and their Pythagorean expectation, we can ascribe that to good fortune, when a team wins more than they are expected to, or bad fortune, when a team wins less than they are expected to. 2nd order wins (silver) relies on metres and breaks gained and conceded. Elo ratings (maroon) rely on the margin of victory and strength of opponent.

Dots should tend to gravitate towards each other. If a team’s dots are close together, that means their actual results are closely in line with their underlying metrics and represents a “true” or “fair” depiction of how good the team is. If a team’s coloured dots are clustered away from their actual record, then we should expect the actual and the coloured cluster to move towards each other over time.

If the black dot is well above gold, that team is suffering from good fortune and may mean regress to more typical luck in the future (vice versa also holds). The silver dots will tend to hover around .500, so if gold is between silver and .500, the team could have an efficiency issue. On the balance, I would expect more often the actual percentage will move towards the cluster but the opposite is also possible.

Player Leaderboards

Production the amount of valuable work done by a team as measured in counting statistics that correlate with winning. These statistics are converted to a single unit called Taylors. Taylor Player Ratings (TPR) are a rate metric that compares an individual player’s production, time-adjusted, to that of the average player at their position, with a rating of .100 being average (minimum 5 games played). Wins Above Reserve Grade (WARG) is a volume metric that converts player’s production over a nominal replacement level into an equivalent number of wins they contribute to their team.

Team Strengths

A chart that tabulates each team’s WARG, both as a team total and broken down into platoons by listed position, to identify each team’s relative strengths and weaknesses.

The WARG is the sum produced at the position, not by specific players, e.g. if Kalyn Ponga produces a total of 1.5 WARG comprising 1.0 WARG while named at fullback and 0.5 WARG named at five-eighth, then the WARG produced while named at fullback will be added to the 1-5 platoon and the WARG created while named in the halves will be added to the 6+7+9 platoon.

The top team in each category is rated 100, the bottom team is rated 0 and every team in between is scaled accordingly.

Team Efficiencies

This table compares the SCWP produced and conceded by each team (a product of their metres and breaks gained and conceded) against the actual points the team scores and concedes to measure which teams are most efficiently taking advantage of their opportunities. A lack of efficiency here could be the result of bad luck and poor execution - sometimes you have to watch the games.

Thanks for reading The Maroon Observer. If you haven’t already, you can subscribe below to receive all the latest about Queensland rugby league.

Subscribe now

If you really enjoyed this, please forward the email on to someone who might also enjoy it.

Share

Lineups for the first game of the men’s Origin series were announced yesterday, largely confirming the Newscorp leaks of the days prior.

Here’s how the sides stack up using each player’s season to date Taylor Player Rating (TPR), Wins Above Reserve Grade (WARG) and the expected Taylors (xTy) each player is predicted to produce.

Before we get too far into the analysis, it’s worth pointing out that this form of Taylor-based analysis is fine for the NRL but it has been extremely bad at predicting the actual outcome of State of Origin matches.

Since 2014, with the exception of two games, NSW has always been favoured, typically by 5% to 10% winning percentage (or a margin of about three to five points). In that time, the Blues lost five of nine series, including losing 14 of 27 games, albeit out-scoring the Maroons 508 to 4271. Perhaps the Taylors just need a 5% adjustment due to the Maroons just getting Origin?

Even the two games in which Queensland were favoured (game 2, 2016 and game 3, 2017), it was by a combined total of nine (9) Taylors, or about as slim a margin as is possible to estimate. In 2023, Queensland have a clear, and unprecedented, advantage in lineups based on their component players’ expected production.

In total, the Maroons have a 26 Taylor advantage, which is good for about a 56% win percentage at this level of production (about four points in margin terms), but also have the edge in the three platoons2.

Still, the Maroons have historically struggled at neutral site venues, losing in Los Angeles, getting smashed on both trips to Perth and batting a bit under .500 in Melbourne (3-0 at Docklands, 1-4 at the MCG and 0-1 at Olympic Park). There’s only ever been a single Origin game in Adelaide. It was in 2020. The worst Queensland side of all time won 18-14 and the rest is extremely familiar history.

Queensland’s Elo rating - an extremely stupid thing to track in a two team series that has three games each year - is even pointing up.

It’s getting hard to find an underdog storyline. Someone’s going to have to get covid.

Trivia

• The peak Origin Elo rating was after game 1 of the 1912 Interstate series, which New South Wales won 65-9 at the RAS Showgrounds in Sydney. This is somehow not the highest winning margin in Interstate history, which was 69-5 in game 3 of the 1957 series, played at the SCG. Queensland’s peak rating was after game 1 of the 1989 Origin series, which Queensland won 36-6 at Lang Park. The 1988 and 1989 series were back-to-back sweeps for the Maroons.

• The narrative that people had lost interest in the Interstate Series, precipitating the switch from residential selections to State of Origin - itself derided as a gimmick that would never take off by the southern authorities, was driven primarily by Sydney’s lack of interest. Attendances at Interstate games in Brisbane in the 1970s averaged over 21,000, up on the 17,000 averaged through the 1960s and only bested by the 24,000 in the 1950s. Meanwhile, attendances in Sydney dropped from 34,000 in the 50s, to under 18,000 in the 60s and bottoming out under 11,000 in the 70s. While no one wants to go back to residential selection rules, not least because Queensland wins more than once a decade under Origin rules, it’s clear that New South Wales’ inability to get Origin has its genesis in the latter days of the Interstate Series.

• Project Genetrix was an American intelligence program during the Cold War to release reconnaissance balloons over the Soviet Union. The program was largely a failure, producing just 31 useable images from 500 balloons, as the Russians used the balloons for target practice. However, the Soviets took the temperature-resistant and radiation-hardened film from the balloons recovered and used the film to take the first photos of the dark side of the moon in 1959.

NRL Stats Drop

Scroll to the end for the explanatory notes or visit How It All Works.

Queensland Cup Stats Drop

This is still in testing

Explanatory notes

Elo Ratings

Elo ratings are a way of quantitatively assessing teams, developing predictions for the outcomes of games and then re-rating teams based on their performance, home ground advantage and the strength of their opposition. Form Elo ratings are optimised for head-to-head tipping and tend to reflect the relative strengths of each team at that particular point in time, although there are many factors that affect a team’s rating.

Pythagorean Expectation and Winning Percentage Comparison

The black dots on the winning percentage comparison are each team’s actual win-loss record to this point in the season. The coloured dots represent what the stats say about the team’s underlying performance, i.e. how many games they should be winning. Wins and losses are binary and can be prone to good and bad luck in a way that other stats that correlate to wins are not, so we have other metrics to help see through the noise to good teams, rather than just good results. While each metric has strengths and weaknesses, it helps paint a more complete picture of team performance.

Pythagorean expectation (gold) relies on points scored and conceded by estimating a team’s number of wins based on their for and against. Where there is a deviation between a team’s actual record and their Pythagorean expectation, we can ascribe that to good fortune, when a team wins more than they are expected to, or bad fortune, when a team wins less than they are expected to. 2nd order wins (silver) relies on metres and breaks gained and conceded. Elo ratings (maroon) rely on the margin of victory and strength of opponent.

Dots should tend to gravitate towards each other. If a team’s dots are close together, that means their actual results are closely in line with their underlying metrics and represents a “true” or “fair” depiction of how good the team is. If a team’s coloured dots are clustered away from their actual record, then we should expect the actual and the coloured cluster to move towards each other over time.

If the black dot is well above gold, that team is suffering from good fortune and may mean regress to more typical luck in the future (vice versa also holds). The silver dots will tend to hover around .500, so if gold is between silver and .500, the team could have an efficiency issue. On the balance, I would expect more often the actual percentage will move towards the cluster but the opposite is also possible.

Player Leaderboards

Production the amount of valuable work done by a team as measured in counting statistics that correlate with winning. These statistics are converted to a single unit called Taylors. Taylor Player Ratings (TPR) are a rate metric that compares an individual player’s production, time-adjusted, to that of the average player at their position, with a rating of .100 being average (minimum 5 games played). Wins Above Reserve Grade (WARG) is a volume metric that converts player’s production over a nominal replacement level into an equivalent number of wins they contribute to their team.

Team Strengths

A chart that tabulates each team’s WARG, both as a team total and broken down into platoons by listed position, to identify each team’s relative strengths and weaknesses.

The WARG is the sum produced at the position, not by specific players, e.g. if Kalyn Ponga produces a total of 1.5 WARG comprising 1.0 WARG while named at fullback and 0.5 WARG named at five-eighth, then the WARG produced while named at fullback will be added to the 1-5 platoon and the WARG created while named in the halves will be added to the 6+7+9 platoon.

The top team in each category is rated 100, the bottom team is rated 0 and every team in between is scaled accordingly.

Team Efficiencies

This table compares the SCWP produced and conceded by each team (a product of their metres and breaks gained and conceded) against the actual points the team scores and concedes to measure which teams are most efficiently taking advantage of their opportunities. A lack of efficiency here could be the result of bad luck and poor execution - sometimes you have to watch the games.

Thanks for reading The Maroon Observer. If you haven’t already, you can subscribe below to receive all the latest about Queensland rugby league.

Subscribe now

If you really enjoyed this, please forward the email on to someone who might also enjoy it.

Share

The captain’s challenge was introduced into the NRL in 2020, an initiative the previous Greenberg administration. By and large, it’s been an interesting improvement to the game and something the administration has more or less gotten right out of the box.

The details are sometimes hard to fathom, but so are many of the intricacies of the NRL, and the challenge hasn’t been without its controversies. No one should forget the end of the Cowboys and Tigers game last year where North Queensland pulled off the most audacious heist in the history of the sport by challenging the very concept of full time1.

I’ve never seen any stats compiled on the use and successes of the captain’s challenge, which is strange when you think about it, but not hard to imagine why. I compiled the numbers for the season-to-date and it makes for interesting reading.

All four Queensland clubs are in the top six for success rate. The league leader, the Manly Sea Eagles, are succeeding two out of three times and are captained by the incumbent Maroons halfback. No one ever said Daly Cherry-Evans was dumb.

Obviously, there are issues using this analysis as a proxy for refereeing competence. It’s only 88 games, or at most 11 per referee, so a tiny sample size. An early unsuccessful challenge deprives the team of the ability to question any absolute howlers that come later in the game, although that seems to be largely the remit of the Bunker, which is blessed with papal infallibility2. Most of the unsuccessful challenges are time wasters in the last ten minutes of the game so the challenging team can catch their breath and reset their defence.

Even given all of the above caveats, I don’t think players should be overturning referees’ rulings at a rate of 57%. That goes double if you are the equal most challenged referee in the league. I haven’t checked the historical rates of challenge successes, which are probably much lower than what we’ve seen from this limited sample size, and referees can’t see everything and often have to make educated guesses about what they’re seeing due to the limitations of physics and the human mind, but that just can’t be right. A professional referee shouldn’t be outperformed by a coin toss, even within the subset of calls that captains tend to challenge.

Trivia

• The Raiders challenge in roughly half their games, equal lowest in the league with the Dolphins, the Titans and the Dragons. With their five challenges, Canberra have been successful exactly zero times. To be clear, I’m not saying Jarrod Croker or Elliott Whitehead are stupid in case that makes them want to beat me up.

• The Raiders are currently seventh on the ladder with a -29 points difference. In fourteenth, the Eels have a points difference of +24. In between them, in tenth, are the Roosters with a difference of -65. The Roosters have challenged in every single game this year.

• For the third year running, the minor premier of the QRLW has not won the grand final. Souths Logan (7-0) didn’t even make it to the final this year, after the Capras (6-1) lost by two points last year and Valleys (7-0) by four the year before.

• The Broncos need 281 metres before the Panthers get 655 to be the first NRL team to clock 20,000 running metres this season.

• It has been traditionally held that the city of Rome was named for its legendary founder and first king, Romulus. Modern scholars believe that it was, in fact, the other way around: Rome, the city, was first named by its inhabitants and that Romulus, the legendary founder, was named after the city as part of the creation of that myth. Whether there was a real Romulus on which the legend was based, we’ll probably never know.

Form Elo Ratings

Elo ratings are a way of quantitatively assessing teams, developing predictions for the outcomes of games and then re-rating teams based on their performance, home ground advantage and the strength of their opposition. Form Elo ratings are optimised for head-to-head tipping and tend to reflect the relative strengths of each team at that particular point in time, although there are many factors that affect a team’s rating.

Pythagorean Expectation

Pythagorean expectation estimates a team’s number of wins based on their for and against with a reasonable degree of accuracy. Where there is a deviation between a team’s actual record and their Pythagorean expectation, we can ascribe that to good fortune, when a team wins more than they are expected to, or bad fortune, when a team wins less than they are expected to.

There’s always one or two teams each year that win more than their Pythagorean expectation and one or two that grossly underperform and it’s difficult to tell in advance which is which, but over the long run, Pythagoras remains undefeated and always demands his tribute.

Win Percentage Comparison

The black dots are each team’s actual win-loss record to this point in the season. The coloured dots represent what the stats say about your team’s underlying performance, i.e. how many games they should be winning. Wins and losses are binary and can be prone to good and bad luck in a way that other stats that correlate to wins are not, so we have other metrics to help see through the noise to good teams, rather than just good results.

Pythagorean expectation (gold) relies on points scored and conceded. 2nd order wins (silver) relies on metres and breaks gained and conceded. Elo ratings (maroon) rely on the margin of victory and strength of opponent. Each metric has strengths and weaknesses.

Dots should tend to gravitate towards each other. If a team’s dots are close together, that means their actual results are closely in line with their underlying metrics and represents a “true” or “fair” depiction of how good the team is. If a team’s coloured dots are clustered away from their actual record, then we should expect the actual and the coloured cluster to move towards each other over time.

If the black dot is well above gold, that team is suffering from good fortune and may mean regress to more typical luck in the future (vice versa also holds). The silver dots will tend to hover around .500, so if gold is between silver and .500, the team could have an efficiency issue. On the balance, I would expect it’s more the actual moving towards the cluster but the opposite is also possible.

Team Efficiency

This table compares the SCWP produced and conceded by each team (a product their metres and breaks gained and conceded) against the actual points the team scores and concedes to measure which teams are most efficiently taking advantage of their opportunities. A lack of efficiency here could be the result of bad luck and poor execution - sometimes you have to watch the games.

Player Leaderboards

Production the amount of valuable work done by a team as measured in counting statistics that correlate with winning. These statistics are converted to a single unit called Taylors. Taylor Player Ratings (TPR) are a rate metric that compares an individual player’s production, time-adjusted, to that of the average player at their position, with a rating of .100 being average (minimum 5 games played). Wins Above Reserve Grade (WARG) is a volume metric that converts player’s production over a nominal replacement level into an equivalent number of wins they contribute to their team.

Queensland In Focus

Thanks for reading The Maroon Observer. If you haven’t already, you can subscribe below to receive all the latest about Queensland rugby league.

Subscribe now

If you really enjoyed this, please forward the email on to someone who might also enjoy it.

Share