What I Read or Listened to This Week

 Posted on April 29, 2018  |  2 minutes  |  Darryl Blackport

Here is some non-basketball content I read or listened to this week that I found interesting: Glitch Capitalism: How Cheating AIs Explain Our Glitchy Society - When you train a machine learning model, sometimes it comes up with a model that solves the problem by finding a loop hole that doesn’t really solve the problem. Endure - A book by Alex Hutchinson , a former national level distance runner in Canada, on pushing the limits of human performance. [Read More]
Reading 

What I Read or Listened to This Week

 Posted on April 22, 2018  |  2 minutes  |  Darryl Blackport

Here is some non-basketball content I read or listened to this week that I found interesting: Everyone Wants To Go Home During Extra Innings — Maybe Even The Umps - Analysis showing that in extra innings the team closer to winning gets more favourable calls on balls and strikes. I would guess some unconscious biases are at play here. Science’s “Reproducibility Crisis” Is Being Used as Political Ammunition - This part stood out to me ‘…the reproducibility debate has already been exploited by political activists. [Read More]
Reading 

Digging Deeper Into Pace in the NBA

 Posted on April 18, 2018  |  4 minutes  |  Darryl Blackport

In NBA stats pace is the number of possessions a team gets per 48 minutes. It is used as a measure of how fast a team plays. While it is generally a good measure, it is flawed in a few ways. The first reason is that it doesn’t separate offensive pace and defensive pace. When pace is brought up, often people are talking about the pace of a team’s offense. A team that plays fast on offense and forces their opponents to play slow on defense may not have a fast pace by the traditional definition. [Read More]
Basketball 

NBA Possession Data

 Posted on April 17, 2018  |  3 minutes  |  Darryl Blackport

I have decided to share my possession details dataset for the 2017-18 regular season. This data is the starting point for all stats on PBPStats.com . I have parsed out possession details from play by play data to create detailed stats for each possession. I spent dozens of hours putting this together and I think there is a lot that can be done with this data. By making it public I’m hoping people will do some stuff with it that I would have never thought of doing. [Read More]
Basketball 

NHL Prediction Model

 Posted on April 14, 2018  |  4 minutes  |  Darryl Blackport

Last summer I was invited by Emmanual Perry , who created corsica.hockey , to participate in a prediction contest for the upcoming NHL season. Having never built a prediction model, and not really following the NHL, I figured it would be a fun challenge undertake. Manny was kind enough to share a dataset with over 5000 variables for all games since the 2007-08 season. This saved me a lot of time that would have been spent gather and cleaning data. [Read More]
Models