1 / 6

An Iterative Strength Based Model for the Prediction of NCAA Basketball Games

An Iterative Strength Based Model for the Prediction of NCAA Basketball Games. Jeff Harrison & Philip Tan. Motivations. Money Betting Pattern Prediction Economic Scientific Extrapolation to Future. Research Questions & Method. Questions:

zazu
Télécharger la présentation

An Iterative Strength Based Model for the Prediction of NCAA Basketball Games

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. An Iterative Strength Based Model for the Prediction of NCAA Basketball Games Jeff Harrison & Philip Tan

  2. Motivations • Money • Betting • Pattern Prediction • Economic • Scientific • Extrapolation to Future

  3. Research Questions & Method Questions: • What method of predicting college basketball games should we use to obtain the best results? • Can we alter the basic algorithm to produce more accurate predictions of the NCAA tournament? Methods: Ranking Systems • ISR System • Tweaking the Standard Determining the Winner • "Winner Takes All"- Higher Ranking = Better Team • Problem: Does not consider how close the rankings are. • Markov Chain • Determining win probability as a function of difference in ranking • If the rankings are close, there is a probability that the lower ranked team will win

  4. Iterative Strength Ranking Overview • Set all teams to an initial ranking. • Go through every game of the season. • Give winner their opponents ranking + a constant bonus • Give the loser their opponents ranking - a constant bonus • Use the ranking generated by this iteration as the starting point for another iteration. (Recursion!) • When two successive iterations yield the same ranking, You're Done!

  5. Results Winner Takes All: • Close game 79.37% Smart Winner: • Standard 82.54% Biases: • Data already known Applicability: • Difficult to apply to future • NCAA basketball volatile

  6. End

More Related