How different are NRL regular season games from finals games statistically?

The meaningless games are over and the NRL finals kick off this weekend, with not much separating the top seven teams on the field (as I broke down last week).

With that in mind, I wanted to have a look at what changes statistically between the regular season and finals games? Are these games played as differently as you’d think watching them? Do teams focus more on defense and try to play conservatively?

Since I know attention spans are short these days, the tl;dr version is there’s no dramatic shift, and the changes are in line with what you’d expect as the weaker teams are eliminated. There is a focus on safer play and defenses contain their opponents slightly more, but it doesn’t markedly change.

That’s not a bad thing though, and “the art of being less wrong” comes into play here. Confirming what you think you already know can be a positive step. Analytics isn’t always about solving a problem (and you’ll never solve a sport, let alone one like rugby league, with data); it’s about imrpoving and informing decision making.

To do this, I’ve taken a look at the average per game statistics from every regular season game from 2014 to 2021, and compared them to every finals game during the same period. Yes, the number of finals games during that time is a smaller sample, but it’s sufficient enough to compare trends.

We’re going to show the data this week in the barbell charts I’ve been using to compare coaching changes over the last few seasons. This allows us to use percentage changes between statistics instead of raw values, as three extra broken tackles per game doesn’t fit on the same scale as 120 extra kick metres per game. By using percentage changes, everything fits on the same scale. All data in this post is sourced from Fox Sports Stats.

These numbers are also possession adjusted. As we’ll get to shortly, there is a difference in the number of play the balls in regular season games compared to finals. By normalising possession, we can remove any influence the extra play the balls have on generating more statistics and isolate what actually changes in these numbers without the influence of increased opportunity to accumulate them.

Below is a chart of the first group of statistics, showing possession and scoring numbers percentage change from regular season games to finals games (blue dot) against the changee from finals games to regular season games (orange dot). The line in the middle represents the gap between both data points. The further to the right the data point is, the higher the increase, the further to the left, the lower the increase.

The first thing to notice is the pace of the game. State of Origin is notoriously faster than regular NRL games, which is no surprise as the the best players are involved and the game speeds up dramatically. In 2022, the average number of play the balls in a regular season NRL game was almost 280 per game. For State of Origin this season, the average number of play the balls was 316, with 344 in game one alone.

For finals games from 2014-2021, the average number of play the balls increases but not by the huge amount we see in Origin. There’s only a 2% increase up to 280 per game, over the 274 seen in regular season games from 2014-2022. So whilst Finals games are “faster”, they’re not as fast as Origin which have 15% more play the balls than regular season games, and 13% more than finals games.

Despite the increase in opportunities to produce statistics purely by the way of more possession, there’s 7% fewer points scored, at 39.7 compared to 41.6.

Interestingly there’s no significant change the number of touches per position, except for hookers having a higher share of touches in finals games.

Number 9s end up with nearly 24% of touches, increasing from their 22.2% during the regular season. Second rowers and locks see a very minor increase in touches, and fullbacks touches increase by 0.4%. All other positions see a drop, especially interchange players who move from 13.3% to 11.7%.

These changes are largely due to hookers playing more minutes in finals games. They increase from 66.7 minutes to 70 per game in finals, and we see a similar drop in interchange minutes from 32.6 to 30.9.

This all makes sense – you want your best players on the field during finals football and there’s no point giving them an extended spell with a utility backup when your season is on the line.

Second rowers are the only other position to change their minutes played by more than sixty seconds, up to 74 minutes from 72.2. The real life eye test would corroborate this, but with average front rower minutes dropping by 30 seconds, the extra interchanges are being used to rotate middle forwards more frequently than during the regular season. Of course, this assumes that bench composition is the same throughout the season, which it isn’t.

Given the above mentioned drop in scoring points you wouldn’t be surprised to then see the fall in other scoring or scoring related statistics. Try assists tumble by 12%, line breaks decline by 9.8%, and line break assists fall by 5.3%. The latter is an interesting one and indicates that getting through the line is more due to individual effort or brilliance than relying on being put into hole.

Possession adjusted goal attempts are practically even between regular season and finals games, dropping just 0.8% for makes and 4% for attempts. Given that the overall rate of scoring drops by 7%, these are actually over indexing and can be accounted for the extra penalty goal attempts in finals matches due to the closer nature.

The goals that are kicked are done so at a higher rate too, rising from 75% to 78% in finals matches.

Looking at runs of the ball, again even with more possession there’s fewer runs, at -1.89% after adjusting for possession.

This can be explained through a number of statistics. With fewer runs, comes fewer total run metres, 2.6% down on the regular season. The fact the total run metres are down more than the number of runs also indicates better defense, with metres per run also dropping from 8.96 to 8.89.

Part of this is that post contact metres become harder to accumulate with the better defense, as they drop by a significant 16% from regular season to finals. This is also exhibited when you look at the length of runs. Those longer than 8 metres fall off by 6.7% in finals matchups, whilst runs shorter than 8 metres only wane by 2%.

Interestingly there are more broken tackles in finals football, at +2.1% which is an increase of around 3 per game. But going on the above post contact metre data, these broken tackles aren’t creating more metres downfield, reducing their effectiveness.

In summary, there’s more possession, but fewer things happening on each play the ball and defenders restrict gains with the ball far better than in the regular season. Parramatta fans who follow the Eye Test on Twitter will have seen my weekly tweets about how the Eels struggle to control metres per set to their opposition.

In finals games, teams don’t have the same opportunities to push the ball downfield with strong post contact runs, which probably explains why the Eels have had a ceiling on their post season progress with Brad Arthur’s current gameplan that relies on having a net possession advantage. 

Compounding these running stat differences is that there are 5% fewer general play passes, as teams eschew expansive passing games for a simpler style of play.

This is also seen with more than 11% one pass runs (standard hit ups) in a finals game than a regular season game. Ball carries are far more likely to lock up the ball rather than look to promote it, with 11.6% fewer offloads, and 16% fewer effective offloads. Dummy halves are given less freedom as well, as there are 4% less runs straight out of the play the ball.

Next lets look at passing, errors and negative play.

A reduction in offloads and general passing translates into better possession, with errors dropping 9%, incomplete sets falling by 7.6%. This results in the increase in play the balls as noted above, however it creates longer sets, not more of them. Total sets drop by 0.9% and complete sets drop by 2.2%, with the average number of play the balls per set rising from 3.66 in the regular season to 3.78 in finals.

This points to teams making errors fewer errors, and when they are they’re making them later in the tackle count than they do in the first 25 rounds of the season. This aligns with the way teams approach finals games, play a safe style of football and wait for your opponent to make a mistake rather than gifting them possession early.

This simpler style of play results in more of the game being played between the 20 metre areas, again something you’d expect with better defensive teams matching up rather than getting a free hit at the likes of the Tigers, Knights or Warriors. The number of play the balls inside 20 by 9%, which equates to about four fewer opportunities per game in that area.

What this results in is more importance on a team’s kicking game.

The number of kicks per game increases by 3.2%, and more long kicks (+6.5%). Naturally this results in more kick metres gained, at +11% and an average kick length of 27.5 metres in finals games vs 25.9 metres per kick in the regular season. There’s also a 2.9% increase in forced dropouts, which could indicate a more precise kicking game.

Lastly there’s also less infringements called, which anyone using the real life eye test would have noticed. There are approximately 14% less penalties awarded and set restarts (from 2020 onwards obviously) are called on 19% fewer occasions than the regular season.

As noted in the introduction, there’s no groundbreaking differences between regular season games and finals football. There is more of a focus on defense, and a smarter, simpler game plan than reduces errors made and conservative running of the ball.

Which is why, as friend of the site Jason Oliver of the wonderful Rugby League Writers has pointed out recently, that only one team outside the top three defensively has won an NRL premiership since 2005.

If you enjoyed this post please consider supporting The Rugby League Eye Test through one of the links below.

Please Add coin wallet address in plugin settings panel