Special Teams Analytics

I’m fascinated by the use of high level statistics in the NHL. I think it’s a great way to look at the game from a different angle and challenge previously held beliefs that are common in the game.

One example is special teams. Everyone in the hockey world looks at PP and PK% as an indicator of their success. While it is a measure of past success, it is not an accurate indicator of future success. Fear the Fin, a blog about the Sharks did a great job breaking down special teams from an statistics point of view. A fascinating look at the different factors that affect a power play (other than tactics and personnel) or penalty kill and how a team’s success or failure in those situations affects their place in the standings over the course of a season. Check out the full article here. The biggest takeaways for me are in the summary:

Summary Points

  • NHL teams spend a substantial portion of games on special teams. 5v4 time alone accounts for 10-20% of game time, and thus needs to be analyzed to further our prediction models.
  • Although we don’t have enough years of NHL RTSS (5 years and counting) data to conclude with statistical significance, it appears that Fenwick For/60, with misses and blocked shots adjusted for scorer bias, is the best predictor of power-play success (GF/60), and well correlated with winning (Pts/game).
  • The penalty-kill picture is less clear, likely the result of heavier regression to the mean. Presumably the heavier regression is because PK units spend much more time without the puck. Both Sv% (which is likely goaltender driven), and Corsi differential/60 (Corsi For – Corsi Against per 60) are predictive of future penalty kill success (GA/60), and winning (pts/game), but less powerful than the predictors of power-play success.
  • Shooting percentage on the power-play is negligible, regressing heavily to the mean, and shows at best very modest correlations with PP success and winning. Even if we attempt to attenuate Sh% (ie. control for how much it regresses to the mean), it still under-performs other more significant metrics like FF/60.
  • If we focus on which stat in theory has the strongest association with PP success (GF/60), we see that Fenwick For per 60 and Shots For per 60 are virtually indistinguishable. On the PK, however, Sv% becomes the strongest stat.

Looking a little closer at PP and PK is the Flyers blog Broad Street Hockey. In two articles, they dig into PP and PK and how you can look at future success for each of them. Their bottom line conclusion? PP success is indicated by shot rate while PK success is a little simpler, with just conversion rate. You can check out the PP article here: http://www.broadstreethockey.com/2011/5/22/2178537/zone-entries-what-drives-power-play-success and the PK article here: http://www.broadstreethockey.com/2011/5/14/2170957/what-makes-a-good-special-teams-unit.

What drove me to read these articles? A very interesting piece on TSN’s Analytics Blog by Travis Yost going into the details of Tampa Bay’s success at even strength and their relative weakness on special teams. Tampa has a power play that is clicking around 20% of the time and a PK that is successful about 85% of the time. Pretty good, no? Well looking deeper, they actually have a shot differential of -35.7 per 60 minutes of special teams play. What the article argues is that their current success rates are unsustainable based upon their shot differential. It will be interesting to see what happens moving forward with Tampa’s special teams units. You can find that article here: http://www.tsn.ca/special-teams-a-prime-concern-for-lightning-1.227049

My takeaways from these articles and insights into statistics? Well it also passes the “eye test” with hockey. Want to have a successful power play? It comes from shots. Shots come from faceoff wins and puck retrievals in the offensive zone. Want to have a successful penalty kill? A good goaltender goes a long way – as does limiting the quality of shots against. Finally, you can look at analytics to see if your success on special teams is due to random events or methodical success through statistical trends.

I’ll be honest, as a history major in college I never took a stats class and I couldn’t tell you what the actual equations and statistical work means. However, I trust that you need that data to prove your points and that the greater analytics community would immediately invalidate any argument that doesn’t pass the statistical data test. I try not to get too lost in the math but instead interpret the numbers and determine what they mean and what the effect on the game might be.


2 Responses to Special Teams Analytics

  1. Will McC says:

    Good Post Chris! My problem with Analytics is the apparent disconnect between what the stats tell us, and how those stats are achieved on the ice. Sure, a team consistently attempts more shots than their opponents; but LA does it in a much more ‘dump and chase’ style, while Chicago likes to enter the zone with more control. The temptation of many to simply say “we outshot them, we should have won, it was just bad luck that we didn’t”, is unsatisfying…
    Keep up the good work!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: