March 8, 2022

Top 14 Predictions for Round 21

By

Team Ratings for Round 21

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
La Rochelle 7.43 6.78 0.60
Bordeaux-Begles 5.90 5.42 0.50
Stade Toulousain 5.60 6.83 -1.20
Clermont Auvergne 5.26 5.09 0.20
Racing-Metro 92 5.26 6.13 -0.90
Lyon Rugby 4.92 4.15 0.80
Montpellier 4.40 -0.01 4.40
Castres Olympique 1.44 0.94 0.50
Stade Francais Paris 1.41 1.20 0.20
RC Toulonnais 0.47 1.82 -1.30
Section Paloise -1.25 -2.25 1.00
USA Perpignan -3.33 -2.78 -0.60
Brive -3.56 -3.19 -0.40
Biarritz -6.58 -2.78 -3.80

 

Performance So Far

So far there have been 136 matches played, 98 of which were correctly predicted, a success rate of 72.1%.
Here are the predictions for last week’s games.

Game Date Score Prediction Correct
1 Biarritz vs. RC Toulonnais Mar 06 17 – 45 1.20 FALSE
2 Bordeaux-Begles vs. Section Paloise Mar 06 16 – 23 15.00 FALSE
3 Castres Olympique vs. Montpellier Mar 06 25 – 9 2.60 TRUE
4 Clermont Auvergne vs. Lyon Rugby Mar 06 25 – 16 6.60 TRUE
5 La Rochelle vs. Brive Mar 06 41 – 15 16.50 TRUE
6 USA Perpignan vs. Racing-Metro 92 Mar 06 34 – 13 -3.60 FALSE
7 Stade Francais Paris vs. Stade Toulousain Mar 07 23 – 16 1.80 TRUE

 

Predictions for Round 21

Here are the predictions for Round 21. 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 Brive vs. Castres Olympique Mar 27 Brive 1.50
2 La Rochelle vs. Racing-Metro 92 Mar 27 La Rochelle 8.70
3 Montpellier vs. Biarritz Mar 27 Montpellier 17.50
4 Section Paloise vs. USA Perpignan Mar 27 Section Paloise 8.60
5 Stade Francais Paris vs. Bordeaux-Begles Mar 27 Stade Francais Paris 2.00
6 Stade Toulousain vs. Lyon Rugby Mar 27 Stade Toulousain 7.20
7 RC Toulonnais vs. Clermont Auvergne Mar 27 RC Toulonnais 1.70

 

avatar

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 »

Comments

  • avatar
    Frédéric Leroy

    Dear David,
    I tried to reconstruct the reasoning to obtain the predicted ratings from the information on your Department page. Thank you for providing us with it. However, I have a problem when calculating the actual rating after a given round so I can obtain the difference between the predicted and actual value. I have tried to infer a formula from your data, to no avail so far. Perhaps could you give some pointers on how you calculate actual ratings? Many thanks in advance. Kind regards from France.

    3 years ago

    • avatar
      David Scott

      There are some tweaks to the basic exponential smoothing algorithm due to the error distribution of scores being long tailed, so insufficiently normal. I choose a threshold value where if the size of the error (positive or negative) is greater than the threshold, I shrink the error towards zero.

      Note that it is also necessary to choose values for all the parameters in the model: the smoothing parameter, the threshold, the shrinkage, two home ground advantage parameters when teams are from different countries (e.g. URC, Super Rugby), and also a shrinkage parameter for shrinkage between seasons.

      3 years ago