Let's talk about shot blocking

Cam Charron
May 29 2013 03:38PM

It's not that blocked shots aren't important, it's that the team with the most blocked shots isn't necessarily the best team at shot blocking—the team with the most blocked shots is often the one trying to clear the puck out of its own end. Like in that PK above. It's an excellent shift from Tyler Carroll, but I'd bet that Guelph would rather not be down 3-on-5 in that situation.

So a new statistic has popped up: "Percentage of shots blocked" and it's a little dicey as well. Generally speaking, it's just not good to block a lot of shots or to be in situations where you have to block a lot of shots. "Percentage of shots blocked" has been kicking around but I've seen no evidence that it's a repeatable statistic that correlates with winning.

Read Article | 8 Comments

Machine Learning and Hockey Predictions: Part II

Josh W
May 28 2013 12:57PM

 

 

Background

Last time I tried to use Machine Learning to create a simple classifier that can predict which of two teams is more likely to win a hockey game.  Machine Learning is a class of artificial intelligence that can take a large amount of data, learn from it and then make future decisions.  A simple description of it is that the algorithms it uses (such as Neural Networks, Decision Trees, and Support Vector Machines) act like a black box, you feed it in data, it learns from it, and then it can make decisions on new and future data. 

Read Article | 7 Comments

Who Are the Edmonton Oilers' Best Scouts?

RexLibris
May 24 2013 10:48AM

I’d begun to notice something of a trend recently amongst the Oilers’ prospects and decided to take a closer look. Excluding 1st overall picks, is there a one area in which the Oilers are collecting the majority of their successful draft picks? In other words, who, based on draft pick performance to-date, are the Oilers’ best amateur scouts?

The Oilers scouting group is headed by Stu MacGregor, but also features others among the ranks, including Frank Musil, Brad Davis (son of former scout Lorne Davis), Bob Brown, Kent Hawley, Jim Crosson and others. Given the importance of the draft in the modern NHL and the Oilers’ dependence on that aspect of talent acquisition lately, which of these scouts is delivering at peak performance?

Read Article | 12 Comments

Kings' assistant Bill Ranford cites Corsi to TSN's Ryan Rishaug

Cam Charron
May 21 2013 09:28PM

I thought this was real cool. At the start of the second period between the Los Angeles Kings and San Jose Sharks, Chris Cuthbert threw it down to reporter Ryan Rishaug for a quick hit where he actually gave some pretty good info as a result of a mid-game interview:

Chris, I spoke with Kings' assistant Bill Ranford at the break I asked 'which part of the stats page makes you the most upset?' He said 'their number of attempts and our lack thereof'. If you consider that San Jose had 15 shots on net, they missed the net eight times and had 14 blocked. That's 37 attempts on net. The Kings had just seven.

There you see the zone time, not even close. Los Angeles with half the offensive zone time that San Jose had. Ranford says 'we are losing every battle in our own end and we are not competing'.

I've screencapped the accompanying graphic, which is some real hipster stuff because you can't find "zone time" anywhere online. It has to be manually tracked. We use Corsi, basically shot attempt differential, as a proxy for zone time, but only because the zone time numbers aren't available. TSN did a great job at showcasing the Sharks' advantage in the first period.

Kudos to Rishaug and TSN there.

Read Article | 3 Comments

Can we make predictions in hockey with Machine Learning? A simple experiment

Josh W
May 18 2013 11:23AM

 

 

Introduction

You probably don't know me, my name is Josh Weissbock and I am a Masters student at the University of Ottawa up in Canada and I study Machine Learning.  Many of the other authors on the Nations Network, who use advanced stats, typically use statistical models to analyze hockey where as I use algorithmic models. What I can do is use Machine Learning techniques which is an application of these same statistical ideas.  Machine Learning is simply the ability for a computer to take a whole bunch of data, learn from it and be able to make decisions. 

Read Article | 8 Comments