Applying CUSUM to hockey prediction models

This slideshow requires JavaScript.

The NHL season is a long and grueling affair and most teams will experience some ups and downs over the course of 82 games. Even a team that had a 67% chance of winning every game it played, would still have a 20% probability of putting up a five-game losing streak. And this is just straight probability theory with fixed probabilities. What happens when you consider all of the factors that go into determining the probability of winning an individual game, let alone predicting performance over an entire season?

Well, I’m not here to answer that question.

What I am here to do is to try to apply an analytical technique that was developed in the 1950s for the purposes of quality control in industrial and manufacturing processes to the game of hockey. Continue reading

How does performance as of Dec. 1 relate to making the playoffs?

Every year right around December 1, NHL analysts and fans alike start thinking a little more seriously about playoff chances. Although in many ways this is an arbitrary point in time, particularly when linking it to US Thanksgiving, there’s also some logic to it as well.

By the end of November, we are two months in and most teams have played 20-25 games, or a quarter of the season. And from an analytics perspective, that’s about the point when you can start making use of most of the early season data.

So while most of the traditional playoff discussion at this time of the year tends to focus on points or even where teams are in the standings, let’s take a look at how early-season shot-based metrics relate to whether a team ultimately makes the playoffs.

Continue reading

Have you even used a calculator, jock?

20161118_125241

There has been quite a lot of talk over the last couple of days about how unhappy NHL players are with the escrow hit they take on every paycheque.

Unfortunately, when you have a collective bargaining agreement that specifies how hockey related revenues are to be split between owners and players, an escrow account is a necessary evil. Because the players’ share is paid out under the provisions of 700 or so individual player contracts throughout the season, there needs to be a mechanism to reconcile those payments with the total share following completion of the season, when all hockey related revenues have been accounted for.

Continue reading

Can defensemen control rebound opportunities? Putting the eye test to the test.

One of the attributes that is often attached to defensemen is the ability to clear rebounds. You hear this quite often on NHL broadcasts, usually after a flurry of rebound shots that ultimately wind up in a goal. The colour commentator will jump in and imagine the goalie is saying to his defensemen, “I’ll stop the first three, you get the fourth one.”

But just how much is there to defensemen’s ability to prevent rebound shots?
Continue reading

Hockey Talk: Shot Quality

Hockey Talk is a (not quite) weekly series where you will get to view the dialogue among a few Hockey-Graphs contributors on a particular subject, with some fun tangents.

This week we started from a Twitter conversation suggesting that expected goals calculations (xG) might underweight “shot quality”. A topic that HG contributors are hardly short of opinions on. Continue reading

Are there unintended consequences to the coach’s challenge?

On a Canucks broadcast earlier this season, Sportsnet’s John Garrett pointed out that we seem to be getting a lot more offside calls this year. And for once, I actually agreed with him.

Garrett theorized that maybe this was an unintended consequence of the coach’s challenge. Perhaps linesemen don’t want to be responsible for having a goal called back because they got the call wrong at the blue line. So if it’s a close call, just be conservative and whistle it down. Coaches can’t challenge an offside call, after all.

The NHL introduced a coach’s challenge to try and get more calls right. And clearly, it should be in everyone’s interest to do so. But what if we’re not just getting more calls right? What if we’re also getting more calls, period?

Ever since that broadcast, every game I’ve watched seems to be rife with offside calls on any play even remotely close at the blue line. It doesn’t matter the player, the team or the score. Bobble the puck? Offside. Drag the skate? Offside. Make an extra move? Offside.

At best this is slowing down the game, but could it also be contributing to the reduced scoring we’ve seen so far this season?

If so, this is an observation that both John Garrett and I picked up on just by watching the games. Maybe Brian Burke is right and hockey really is an eyeballs sport.

Let’s find out.

Continue reading

The relationship between competition and observed results is real and it’s spectacular

Raw Comp Impact

Abstract

There has been much work over the years looking at the impact of competition on player performance in the NHL. Prompted by Garret Hohl’s recent look at the topic, I wanted to look at little deeper at the obvious linear relationship between Quality of Competition and observed performance.

The results are a mathematical relationship between competition and observed, which could provide insight into player performance over short time frames. In the long run, the conclusions drawn by Eric Tulsky still hold. The impacts of facing normally distributed Quality of Competition (QoC) will wash out the effects over time. But this should not preclude consideration and even adjustments for QoC when looking at smaller sample sizes.

Continue reading