Welcome to the first day using my Rebuilt Machine Learning Algorithms for the NHL. I’m not sure how these updated models will perform since I have never done NHL before, but we will find out starting with today’s games.
I have run tests on the models from this weeks games and will continue to make make edits and tweaks as needed throughout the remainder of the season, but hopefully this performs well and doesn’t require any significant changes.
I do not bet NHL and I do not follow it extremely close either, so right now the plan is to keep it simple and post just the top performing model. I don’t have the time to go more in-depth like I do with NBA. If you have any suggestions please let me know!
Hope you all can get something useful from this and, as always, best of luck everyone!
Before building my models to be more accurate my models were on average ~75% ATS, 45% Moneyline, and 35% O/U. Hopefully the rebuild makes the moneyline and O/U projections more accurate without hurting the ATS accuracy.
Top Performing Models
Based on Average Win % between ATS, Moneyline, and O/U.
Updated after todays games.
- kNN – TBD
- Random Forest – TBD
- Ada Boost – TBD
- Neural Net – TBD
- SVM – TBD
- Basic Model – TBD
- Linear Regression – TBD
- SGD – TBD
Summary of Projections
1. k-Nearest Neighbor (kNN)
The lazy learner. Nearest Neighbor has performed best with 60 Neighbors.
Record: TBD ATS, TBD ML, TBD O/U.
This is the first time I am using any of these algorithms for NHL.
If you don’t understand what you are looking at, please read my post about betting tips. The percentages represent value for that particular number. Good value alone doesn’t mean you should bet it (To quantify, >35% is great value, 20-35% is really good, 10-20% is decent, <10 is ok value, blanks are negative value).