In this project I collected a number of traditional and advanced statistics,over a period of 10 weeks, in the National Hockey League. I use this data totrain a number of diﬀerent classiﬁers to forecast success in an NHL game. Inthe ﬁrst half I compare classiﬁers using traditional stats, advanced stats andboth. Then I use these results and expand upon the advanced stats, usingvariants of the PDO stat, and train new classiﬁers. The best classiﬁer is aNeuralNetwork able to accurately predict success 84.33% of the time using10-fold cross-validation and the PDO of each team over the last 3 games.