Mining NHL Draft Data and A New Value Pick Chart

For every hockey player, getting drafted to the NHL is a dream come true, but the real goal is to reach 160 games played (GP). A simple model is fit to NHL draft years 1998 to 2009 with an aim at predicting the proportion of players who will play 1 or 160 games. For individual players drafted between 1998 and 2011, predictive models (GLM, Neural Network, SVM and LOESS) are created to predict a player's career GP, then combined to create a voting model to decide whether or not a player will reach 160 GP. Players drafted from 1998 to 2008 were analysed using non-linear multi-variate model, with a modified weighted least squares, to predict a player's Time-On-Ice (TOI) for their first seven seasons. The results were analysed for positional advantage, then converted to value pick charts (used by teams when choosing a draft pick or considering a trade).

Name thesis:degree_name

Master of Science (M.Sc.)

Level thesis:degree_level

Master's

Discipline thesis:degree_discipline

Applied Mathematics

Grantor dc:publisher

Carleton University

Year dc:date.issued

2016

Author dc:creator

• Wilson, David

Statement dc:rights

• Copyright © 2016 the author(s). Theses may be used for non-commercial research, educational, or related academic purposes only. Such uses include personal study, research, scholarship, and teaching. Theses may only be shared by linking to Carleton University Institutional Repository and no part may be used without proper attribution to the author. No part may be used for commercial purposes directly or indirectly via a for-profit platform; no adaptation or derivative works are permitted without consent from the copyright owner.

Language dc:language.iso

en

DOI

10.22215/etd/2016-11431

OAI identifier oai:identifier

oai:carleton.scholaris.ca:20.500.14718/38918