February 14, 2013

Super 15 Predictions, Round 1

Team Ratings for Round 1

Welcome to the new Super Rugby season. This year the predictions have been slightly changed with the help of a student, Joshua Dale. The home ground advantage now is different when both teams are from the same country to when the teams are from different countries. The basic method is described on my Department home page.

The introduction of a new team causes problems. I have arbitrarily assigned a rating of -10 to the Kings. This value worked reasonably well when the Rebels were introduced but obviously there will be uncertainty about games involving the Kings until they have some history.

Here are the team ratings prior to Round 1, along with the ratings at the start of the season.

Current Rating Rating at Season Start Difference
Crusaders 9.03 9.03 0.00
Chiefs 6.98 6.98 -0.00
Sharks 4.57 4.57 0.00
Hurricanes 4.40 4.40 0.00
Stormers 3.34 3.34 0.00
Bulls 2.55 2.55 0.00
Reds 0.46 0.46 -0.00
Brumbies -1.06 -1.06 -0.00
Blues -3.02 -3.02 0.00
Highlanders -3.41 -3.41 -0.00
Waratahs -4.10 -4.10 0.00
Cheetahs -4.16 -4.16 -0.00
Force -9.73 -9.73 0.00
Kings -10.00 -10.00 0.00
Rebels -10.64 -10.64 0.00

 

Predictions for Round 1

Here are the predictions for Round 1. The prediction is my estimated 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 Rebels vs. Force Feb 15 Rebels 1.60
2 Brumbies vs. Reds Feb 16 Brumbies 1.00

 

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
    TashNz

    Hi David! I may have discovered somehting amazing here :) Did you do this for last year and if so do you have an overall percentage of picking the correct outcomes? Thanks. Tash

    11 years ago

    • avatar

      I predicted Super Rugby and NRL last year. Super Rugby is more predictable than the NRL. I got 71% correct for Super Rugby, 61% for the NRL.

      11 years ago

      • avatar
        TashNz

        That’s fantastic. I make my Jimungo pics on stats and everyone at work laughs at me, but I came 3rd in our work comp (1 point away from the 2nd placer lol).

        11 years ago

  • avatar
    Jarred

    Hi, when do you post the predictions for the next round? I can’t believe you predicted kings to win, no one I know thought they’d win, you made me a believer.

    11 years ago

    • avatar

      I try to post on a Wednesday.

      Don’t believe too much. The Kings were playing the Force who I rate very lowly and were also at home, with the force having to travel to a different country. Rating the Kings is basically a lottery until they have played a few games. As a new team I arbitrarily gave them a low ranking (-10), which had worked reasonably well in the past when the Rebels started as a new team.

      I have had a couple of good rounds though. Eight correct out of nine, only missing the Blues has been good. The Blues may cause me problems again this year if they improve a lot. Last year they went bad and caused me problems.

      11 years ago

      • avatar
        Gary

        How did you come up with the initial ratings? I note that they are slightly different than the ratings at the end of the previous season.

        I’d also be interested in more detail about the different home advantage weightings. Do you know what your predicition rate would have been for the previous season if you had used the different home advantage weightings?

        11 years ago

        • avatar

          Deciding on the actual parameters to use is the hardest part of the exponential smoothing approach. I changed the smoothing parameter a little between the years which explains the difference in the ratings between the end of last year and the beginning of this. (I am surprised that someone actually noticed.) It is not appropriate to choose the optimum parameter values from the previous year or years because the optimum parameter values change quite a lot from year to year so you would really be overfitting which is a bad thing when predicting. I basically choose a reasonable value based on a number of previous years for each of the parameters I need.

          I don’t know offhand what I would have got if I had used this year’s home advantage weightings, but my guess would be I might have got one or two more games correct. Improvements are really quite hard to obtain, and can’t be expected to be dramatic. If I look at my predictions, and think whether I would like to change them given my general assessment of the game, I am rarely unhappy with what the model has come up with. I am of the view that the predictability of the result of Super 15 games is of the order of 70%. Anybody achieving much above that is relying on luck (assuming there is no match fixing occurring).

          11 years ago

  • avatar

    I hardly create responses, but I browsed a few of the responses on Super
    15 Predictions, Round 1 | Stats Chat. I do have a couple of questions for you if
    it’s okay. Is it just me or does it give the impression like some of these comments come across like they are written by brain dead individuals? :-P And, if you are writing at other online sites, I’d like to
    follow you. Could you list of every one of all your social community pages like
    your twitter feed, Facebook page or linkedin profile?

    11 years ago

  • avatar
    Gary

    Thanks for your detailed response above, David.

    11 years ago