February 14, 2014

Super 15 Predictions for Round 1

Team Ratings for Round 1

The basic method is described on my Department home page. I have made some changes to the methodology this year, including shrinking the ratings between seasons. Still the Crusaders have a high rating. Despite being tempted to arbitrarily change their rating, I have stuck with the formula. You may wish to downgrade them. The replacement of the Kings by the Lions provides another problem. In the past I have given a rating of -10 to an unknown team entering the competition, and this has seemed to provide reasonable predictions. The Lions do have a rating from when they were previously in the competition so I have used that rating. Again, you may wish to lower their rating.

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
Crusaders 8.80 8.80 -0.00
Bulls 4.87 4.87 -0.00
Sharks 4.57 4.57 0.00
Stormers 4.38 4.38 -0.00
Chiefs 4.38 4.38 0.00
Brumbies 4.12 4.12 0.00
Waratahs 1.67 1.67 0.00
Reds 0.58 0.58 -0.00
Cheetahs 0.12 0.12 -0.00
Hurricanes -1.44 -1.44 -0.00
Blues -1.92 -1.92 -0.00
Highlanders -4.48 -4.48 0.00
Force -5.37 -5.37 -0.00
Rebels -6.36 -6.36 -0.00
Lions -6.93 -6.93 0.00

 

Predictions for Round 1

Here are the predictions for Round 1. 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 Cheetahs vs. Lions Feb 15 Cheetahs 9.60
2 Sharks vs. Bulls Feb 15 Sharks 2.20

 

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
    Willis du Plessis

    Thank you

    10 years ago

  • avatar
    Kim Selwood

    Hi David

    Although I am far from a stats expert, and don’t understand all the methodology, I followed your predictions for the Super 15 in 2013, and found your accuracy pretty good. So I will be following you again this year. The only difference being that I will not always agree with your predictions for our local South African teams, and perhaps I will go with my knowledge of the local teams. Especially the Sharks, who I support. Thank you for an informative web site. Regards

    Kim

    10 years ago

  • avatar
    Rob Mai

    Hi David,

    Seems you have quite a following from South Africa.

    I’ve only recently found your site and am keen to follow your predictions more closely.

    Following on from the comment (Kim) above, I’m intrigued by the following statement regarding your methods:
    “Obviously there is some initial estimation of team ratings, and some choice of the optimum parameter values for the exponential smoothing constant and the home ground advantage.” … To what extent are your choices here subjective? I.E. Could they be influenced by your knowledge (or lack therefore) of the local factors at play in each country? AND/OR are these ratings based (primarily) on past results? How far back?
    Overall, I probably think the positions of the teams in your ratings are pretty accurate, but some of the actual ratings scores (and the amount by which they differ) are a bit of a surprise to me.

    Will continue to follow with interest.

    Cheers
    R

    10 years ago

    • avatar

      My parameter values are all chosen by examining the data: I try to be totally objective. The only point where the country is relevant is there are two different home ground advantage values, one when the away team is from the same country, and another when it from a different country. I have data back to 2006.

      That said, it is difficult to choose the parameters for the model, because the optimum values change from year to year.

      10 years ago