Game projection model – In-season Model (Part II)
In this article I will start building my game projection model. I’d recommend reading the two first articles of the series before you continue reading this post: The model construction The idea is to build a game projection model that actually consists of two separate models: A pre-season model An in-season model The goal ofContinue reading “Game projection model – In-season Model (Part II)”
Hockey Deployment Management analysis using Machine Learning
This article is written by Chris from ImPCT Sport. Follow them on Twitter: @ImpctSport. Here’s the previous articles written by ImPCT Sport: Lineup deployment management is a subject of debate in the sport community. By lineup deployment, I mean the manner in which ice time and starts are divided (or spread) between players during aContinue reading “Hockey Deployment Management analysis using Machine Learning”
Game projection model – The Variables (Part IB)
After I published the last article, I had a great conversation @IanGraph. Based on this conversation, I’ve decided to make a follow-up article. The main question is: Why does my findings on predictability differ from other research on the subject? Tests typically show that corsi and expected goals are better predictors than actual goals. However,Continue reading “Game projection model – The Variables (Part IB)”
Game projection model – The Variables (Part I)
The goal with this series is to build a new game projection model. In this first article I will focus on the different variables/metrics. I want to test how well they predict future results. The old model But first a few comments on the old model. It was built on the basis of Evolving-Hockey’s GARContinue reading “Game projection model – The Variables (Part I)”
Talent distribution – Contract value (Part V)
This is the fifth part of the talent distribution series. You can find the parts here: Percentiles, Predictability, Forwards vs. Defenders and goaltending. In this post I will focus mainly on contracts and contract value. The data set All contract data for this article is from capfriendly.com, and the data goes back to the 2013-2014Continue reading “Talent distribution – Contract value (Part V)”
Deep Dive into Offensive and Defensive Quality in the NHL
This article is written by ImPCT Sport. Make sure to give them a follow on Twitter @ImpctSport. In my previous publication, I covered the ImPCT Engine. The ImPCT engine is a machine learning based algorithm that was developed to process large amounts of sport related data and provide meaningful intelligence on the performance of players.Continue reading “Deep Dive into Offensive and Defensive Quality in the NHL”
Talent distribution – Goaltending (Part IV)
This is the 4th article in the talent distribution series. You can find the other articles here: Part 1, Part 2 and Part 3. This piece will focus on goaltending. What is rink bias? You can’t really discuss goaltender statistics without also mentioning rink bias. Very basically, you can say that rink bias is shotContinue reading “Talent distribution – Goaltending (Part IV)”
Deep Learning Modeling of Hockey Game Contribution
The article below is written by Chris Tremblay from ImPCT Sport. Make sure you’re following them on Twitter @ImpctSport. They’ve created a player evaluation model based on machine learning, and here’s a short introduction to the model. More to come from ImPCT Sport in the future. Player profiles using the ImPCT engine Assessing player’s entireContinue reading “Deep Learning Modeling of Hockey Game Contribution”
Talent distribution – Forwards vs. Defenders (Part III)
In this piece of the talent distribution series, I will compare forwards and defenders. I recommend reading Part 1 and perhaps Part 2 first. How the sGAA model is designed Before we get started it’s important to understand how my sGAA model is calibrated. It’s designed to describe goal differential at the team level. So,Continue reading “Talent distribution – Forwards vs. Defenders (Part III)”
Talent Distribution – Predictability (Part II)
I didn’t really plan to discuss predictability in this series about talent distribution, since the two are not directly connected. However, I was asked about it, and it does open up for a good discussion about descriptive vs. predictive modelling. I recommend reading part 1 before you read this piece. Descriptive vs. Predictive models TheContinue reading “Talent Distribution – Predictability (Part II)”
Loading…
Something went wrong. Please refresh the page and/or try again.
Hockey-Statistics 