NRL Predictions for the Grand Final
Team Ratings for the Grand Final
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 | |
|---|---|---|---|
| Storm | 8.13 | 16.73 | -8.60 |
| Roosters | 7.37 | 0.13 | 7.20 |
| Sharks | 4.08 | 2.20 | 1.90 |
| Rabbitohs | 3.73 | -3.90 | 7.60 |
| Broncos | 2.73 | 4.78 | -2.10 |
| Raiders | 1.85 | 3.50 | -1.70 |
| Panthers | 0.87 | 2.64 | -1.80 |
| Cowboys | 0.13 | 2.97 | -2.80 |
| Dragons | -0.06 | -0.45 | 0.40 |
| Warriors | -0.74 | -6.97 | 6.20 |
| Bulldogs | -0.82 | -3.43 | 2.60 |
| Titans | -4.06 | -8.91 | 4.90 |
| Wests Tigers | -5.43 | -3.63 | -1.80 |
| Sea Eagles | -5.47 | -1.07 | -4.40 |
| Eels | -5.98 | 1.51 | -7.50 |
| Knights | -8.66 | -8.43 | -0.20 |
Performance So Far
So far there have been 200 matches played, 126 of which were correctly predicted, a success rate of 63%.
Here are the predictions for last week’s games.
| Game | Date | Score | Prediction | Correct | |
|---|---|---|---|---|---|
| 1 | Storm vs. Sharks | Sep 21 | 22 – 6 | 2.10 | TRUE |
| 2 | Roosters vs. Rabbitohs | Sep 22 | 12 – 4 | 2.90 | TRUE |
Predictions for the Grand Final
Here are the predictions for the Grand Final. 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 | Roosters vs. Storm | Sep 30 | Roosters | 2.20 |
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 »
Hey.
Greetings from Bulgaria!
I am watching the NRL grand final atm, and amazed at what is going on ( Roosters up 12 p ATM). I am curios, have ou published the math behind the method you use, I would be curious if it works for the NFL as well (also huge fan)
Regards
6 years ago
The basic method I use is the error correction form of exponential smoothing. The specific application to sports prediction is due to Stephen Clarke from Swinburne University in Melbourne and was originally developed for AFL predictions. He has published a number of papers describing his approach.
Yes, it would work for NFL, for basketball and a number of other sports, but not soccer or hockey. There is an implicit requirement that the scoring system produces score differences which may be taken to be normal or at least continuous and symmetric. Sports where scores are very low such as soccer don’t satisfy that requirement.
The main point of publishing these predictions is to show the value of statistics even with very simple methods and limited data. In particular, exponential smoothing produces startlingly good predictions. In addition though, followers of the predictions should also realise the inherent variability in sporting contests. Take for instance, the recent Mitre 10 Cup results: after a number of weeks of getting most results correct, this week there were less than 50% correct predictions.
I don’t suggest that my predictions are the best that could be produced, betting companies use much more data and more complicated approaches and should do better, provided they don’t overfit.
One additional note concerning the NRL Grand Final, where I predicted the Roosters as winners. That was down to me deciding to give them home ground advantage based on them playing in Sydney against a team from Melbourne, even though ANZ Stadium is not their home ground (listed as Allianz Stadium and Central Coast Stadium). Situations such as that are the only time I use personal judgement rather than a straight statistical algorithm.
6 years ago
Thank you for your quick and detailed answer. Home advantage is a key factor, but so are missing key players and others. In NFL missing QB is really important, if he is a good one. As for basketball,fivethirtyeight have a very complex prediction system, based on ELO, but incorporating things like altitude, miles traveled by team before that, player contribution to the team, etc.
I will have a look at the papers you pointed out and comment if something interesting comes out.
Kind regards.
6 years ago