DataTaunew | comments | leaders | submitlogin
NBA score prediction by matrix completion (github.com)
11 points by lab_Fly 2684 days ago | 4 comments


3 points by rcompt 2684 days ago | link

Do you have any evaluation metrics to report on this? Really curious how this method performs

-----

2 points by lab_Fly 2683 days ago | link

I can copy and paste my response to the github issue:

Not yet. This weekend I'm going to sit down and do some back-testing. One can easily leave values out of the known matrix, and see how well they are predicted in the rank-reduced SVD matrix. I can say, however, that I have tested it against the last ~100 or so games, and I've beat the line (The spread, not winner/loser) about 73% of the time so far.

Thanks for the feedback!

-----

2 points by rcompt 2683 days ago | link

Thanks for the reply, that is rather good results for the spread. Looking forward to any updates

-----

1 point by lab_Fly 2683 days ago | link

Yeah, The model has a single parameter that I need to determine with cross-validation. I'm excited to see how that helps.

-----




RSS | Announcements