## Penalty Killing Attributes

Efficient penalty killing is important for every team. Not only can it stifle your opponent's momentum, but it gives your team something to build on going forward. I wanted to see which statistics were related to good penalty killing, so I compiled some in excel and ran a regression. Method and results after the jump.

The statistics I used for my analysis are for the 2009-2010 regular season. They were collected from the database at NHL.com.

I input the data into excel and ran a simple multi-variable regression. The dependent variable was a team's penalty kill rate. The independent variables were hits, blocked shots, faceoff win percentage, shots against on the pk, giveaways, takeaways, and the number of penalties taken on the year. I thought all of these variables might be related to a team's ability to kill penalties. If anyone has some ideas about others to include, I'm all ears.

One important note is that shots against was the only statistic used that was broken down by man-power situations (even strength, short-handed, or powerplay). Optimally, I would have liked all of the independent variables to have this breakdown, but the data for the other six variables is simply the total from the year, which includes even strength, short-handed and powerplay situations.

The results were not very helpful. I put them into a Google doc sheet for anyone who wants to take a closer look. The first sheet open is the regression results, and the second sheet (sheet1) is the data. None of the independent variables were statistically significant. Blocked shots had the biggest t stat (-1.15), but it still doesn't even approach marginal significance. And the R squared is about 0.3, so most of what makes a penalty killing unit good is not being captured by the statistics I chose.

I did have one idea about a new statistic that could be used to aid in an assessment of a penalty kill unit. I call it the set-up rate, and it would measure how well a penalty kill unit prevents the opposing powerplay from setting up in their zone. A set-up is defined as when the team on the powerplay attempts at least one shot (it can be missed, blocked or on goal) in the offensive zone before the puck is cleared out. So if a team on the pp wins a faceoff in the offensive zone, but doesn't attempt a shot before the puck is cleared down the ice by the pk, the pk has a 0% set-up rate because it prevented the set-up.

UPDATE: I just crunched some numbers and replaced faceoff winning % with shorthanded faceoff winning % in the regression. Unfortunately, the results were the same. Since I have shots against and faceoff % for SH situations, but no statistically significant variables, it's unlikely that getting shorthanded stats for all of my independent variables will make much of a difference. My R squared was once again about 0.3. Here's the Google doc sheet with my new regression results.

Thoughts?

If this FanPost is written by someone other than one of the blog's editors, the opinions expressed in it do not necessarily reflect those of this blog or SB Nation.

## Trending Discussions

forgot?

We'll email you a reset link.

Try another email?

### Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

### Join Arctic Ice Hockey

You must be a member of Arctic Ice Hockey to participate.

We have our own Community Guidelines at Arctic Ice Hockey. You should read them.

### Join Arctic Ice Hockey

You must be a member of Arctic Ice Hockey to participate.

We have our own Community Guidelines at Arctic Ice Hockey. You should read them.