NBA presentation1[1]

NBA presentation1[1] - champs Often had 100% accuracy We...

Info iconThis preview shows pages 1–16. Sign up to view the full content.

View Full Document Right Arrow Icon
P redicting NBA Champ ion s Jonathan Bush, Mike Denny, and Nicholas Sulli van
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Outline § Introduction § Methods § Problems § Success! § Conclusion § Future Opportunities § Questions Outline Introduction Questions Methods Problems Successes Conclusion
Background image of page 2
A Sure Bet? The 2007 Dallas Mavericks Incredible winning percentage of . 8171 They were the overall favorite Got knocked out in first round VS.
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
More than Winning % Season record seems like good predictor This predictor has sensitivity of 58.2% Classification Confusion Matrix Predicted Class Actual Class 1 0 1 32 23 0 31 1172
Background image of page 4
The Data Wrote program to get every team’s data Added two columns manually: Point difference and Champ (1) or not (0)
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Analyses MLR Helped determine which stats were important Logistic Regression Classified teams into champs (1) or not (0) Proved to be the most valuable to us
Background image of page 6
Problems Model great at predicting non-
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 8
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 10
Background image of page 11

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 12
Background image of page 13

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 14
Background image of page 15

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 16
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: champs Often had 100% accuracy We tried oversampling, unsuccessfully Results were garbage Could not do better than 28% sensitivity Beating heads against wall Work Smarter Improve model through domain knowledge The NBA has changed multiple times Divide data into decades Recent data is more relevant to today Use MLR to find important variables Success at Last Started getting models surpassed 58% Logistic Regression was most successful Important Variables List most important variables A Useful Model Classified 17 of 323 as champs and got 10 right Place Your Bets The model has the Heat winning it all this year 2007 Revisited Our model chose the Spurs over the Mavericks Conclusion Future Research? Question s?...
View Full Document

Page1 / 16

NBA presentation1[1] - champs Often had 100% accuracy We...

This preview shows document pages 1 - 16. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online