NRL Predictions for Round 14
Team Ratings for Round 14
The basic method is described on my Department home page.
Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.
| Current Rating | Rating at Season Start | Difference | |
|---|---|---|---|
| Panthers | 12.54 | 8.77 | 3.80 |
| Warriors | 6.75 | -1.18 | 7.90 |
| Sharks | 5.94 | 7.25 | -1.30 |
| Roosters | 5.83 | 9.50 | -3.70 |
| Sea Eagles | 5.19 | 0.21 | 5.00 |
| Dolphins | 4.75 | 1.85 | 2.90 |
| Storm | 2.33 | 6.96 | -4.60 |
| Cowboys | -0.29 | -2.69 | 2.40 |
| Rabbitohs | -0.37 | -5.05 | 4.70 |
| Broncos | -0.77 | 7.06 | -7.80 |
| Raiders | -1.33 | 1.62 | -2.90 |
| Bulldogs | -4.42 | 2.13 | -6.50 |
| Knights | -4.71 | -14.06 | 9.30 |
| Wests Tigers | -4.98 | -7.26 | 2.30 |
| Titans | -6.75 | -8.02 | 1.30 |
| Eels | -6.86 | -0.37 | -6.50 |
| Dragons | -12.84 | -6.72 | -6.10 |
Performance So Far
So far there have been 100 matches played, 61 of which were correctly predicted, a success rate of 61%.
Here are the predictions for last week’s games.
| Game | Date | Score | Prediction | Correct | |
|---|---|---|---|---|---|
| 1 | Sharks vs. Sea Eagles | May 29 | 28 – 22 | 4.40 | TRUE |
| 2 | Knights vs. Eels | May 30 | 28 – 22 | 6.20 | TRUE |
| 3 | Wests Tigers vs. Bulldogs | May 30 | 22 – 16 | 2.70 | TRUE |
| 4 | Storm vs. Roosters | May 30 | 18 – 4 | -1.80 | FALSE |
| 5 | Broncos vs. Dragons | May 31 | 26 – 30 | 19.30 | FALSE |
| 6 | Raiders vs. Cowboys | May 31 | 26 – 12 | 1.00 | TRUE |
| 7 | Panthers vs. Warriors | May 31 | 20 – 18 | 11.20 | TRUE |
Predictions for Round 14
Here are the predictions for Round 14. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.
| Game | Date | Winner | Prediction | |
|---|---|---|---|---|
| 1 | Sea Eagles vs. Rabbitohs | Jun 04 | Sea Eagles | 9.60 |
| 2 | Storm vs. Knights | Jun 05 | Storm | 11.00 |
| 3 | Raiders vs. Roosters | Jun 05 | Roosters | -3.20 |
| 4 | Cowboys vs. Dolphins | Jun 06 | Dolphins | -1.00 |
| 5 | Broncos vs. Titans | Jun 06 | Broncos | 10.00 |
| 6 | Wests Tigers vs. Panthers | Jun 07 | Panthers | -21.50 |
| 7 | Sharks vs. Dragons | Jun 07 | Sharks | 22.80 |
| 8 | Bulldogs vs. Eels | Jun 08 | Bulldogs | 6.40 |
David Scott obtained a BA and PhD from the Australian National University and then commenced his university teaching career at La Trobe University in 1972. He has taught at La Trobe University, the University of Sheffield, Bond University and Colorado State University, joining the University of Auckland, based at Tamaki Campus, in mid-1995. He has been Head of Department at La Trobe University, Acting Dean and Associate Dean (Academic) at Bond University, and Associate Director of the Centre for Quality Management and Data Analysis at Bond University with responsibility for Short Courses. He was Head of the Department of Statistics in 2000, and is a past President of the New Zealand Statistical Assocation. See all posts by David Scott »
Good Morning
A few questions.
How is this year’s prediction rate compared to previous years?
Given the initial estimate at the start of the season, does that mean the prediction success rate will increase as the season goes along?
The NRL has the lowest current successful prediction rate this year. Why do you think that is?
Thanks for your time
2 months ago
I don’t answer many questions since most are not statistically interesting. I had to go back a little while to answer this one.
Without actually checking the data my gut reaction would be to say NRL success rates are typically less than those of the various rugby ones, in the 60’s with rugby competitions being largely in the 70s. I did for some reason achieve better success rates in the covid years. Also I got the impression that sometimes results early in the season were not so good. What then does the data say?
I have been doing predictions since 2012 for the NRL. Past results over that time averaged 64% correct, median 62.7%, minimum 58.5%, maximum 72.5%. This year’s results (currently 64.2%) seem to be distinctly average.
I only looked at the first week success rates for past years. It was highly variable as you can imagine. The average success rate improved from 59% to 64% but the difference in the median success rates was only 0.2%, and the median of the differences was -0.5%.
The highest success rates were achieved in the covid era, rates for 2020, 2021, and 2022 were 69%, 72.5% and 68%.
For comparison, I only looked at Super Rugby. Past results from 2012 to 2025 averaged 71% correct, median 70.8%, minimum 63.8%, maximum 79.4%. This year’s results are 73.2%, not too far off the long term average.
Regarding changes over the season, for Super Rugby the average success rate improved from 68% to 71% but the difference in the median success rates was -0.6%, and the median of the differences was -3%.
So, overall it doesn’t seem as if there is much change on average over the season, just the initial results are a bit more variable. Rugby union may be more predictable than NRL generally. As point of interest I started AFL predictions more recently and NFL last year. AFL seems to be a bit less predictable than rugby union, while NFL might be the hardest to predict since I got only 60.6% correct last year.
One guide to how predictable a competition might be is to look at the range of ratings: if it is large then there are big differences between teams so that more games might be predicted correctly. When comparing different codes though the overall level of scores needs to be taken into account: AFL has much higher scores in general than rugby union or rugby league which I think are broadly comparable.
Why should a particular sport be more predictable than another sport, specifically rugby league compared to rugby union?
Exponential smoothing prediction will be most effective when ability persists from one time period to the next. Rugby union relies to a great extent on set piece performance, namely scrums and lineouts, and these disciplined, trained aspects of the game might be expected to persist more than individual pieces of skill. Scrums in rugby league are some form of bad joke, and with exception of a 40-20 the team that last touched the ball when it goes over the sideline essentially hands over possession to the other team. So effectively no set pieces in my view.
Some competitions try to make teams more equal. Salary caps are one method. The NRL has a fairly firm salary cap while those in the various rugby union leagues are a bit looser, Super Rugby in particular. A number of teams have been penalised for salary cap breaches in the NRL but not in Super Rugby. (I have no knowledge of other leagues.)
One thing I am unable to rule out is the existence of match fixing. Gambling, including by players, and more recently by a referee raises the probability that there could be corruption. The Ryan Tandy case was a confirmed case of spot-fixing.
Overall, why are rates different? Different form persistence, more or less closely matched team abilities, and possible corruption.
3 weeks ago