Monday, November 05, 2007

That's why they call them 'special'

Gabe Desjardins has cranked a few more numbers over on his site, and from these we can now determine which teams spend the most time on the power play, the penalty kill and even strength.

Obviously, if you're a coach, you're going to want your team on the power play more than the penalty kill because, well, goals don't usually go in the wrong net when you're up a man. (It's a little more complicated than that, so keep reading.) The average NHL game this season has broken down with 45 minutes of even-strength play, and 15 minutes on special teams, 7.5 on the power play for each side.

Those figures vary for each team, with Columbus spending the highest percentage of its minutes up a man (15.4% or 9:15/G) and Boston the lowest (10.2% or 6:05/G), which means the Blue Jackets have played more than three minutes a game more on the power play than the Bruins through the first dozen games.

Toronto, meanwhile, has spent the most time killing penalties (a whopping 16%), with Edmonton coming in at the low end at (10.3%) — about a 3:25 difference per game.

But here's what what you really came to see (figures from before Sunday games):

1 N.J 76.1% 13.1% 10.8% 2.4%
2 CHI 74.9% 13.6% 11.5% 2.1%
3 CBJ 71.3% 15.4% 13.3% 2.1%
4 MTL 75.9% 13.0% 11.1% 1.9%
5 L.A 74.5% 13.6% 11.9% 1.7%
6 PHX 76.2% 12.5% 11.3% 1.1%
7 PIT 74.6% 13.2% 12.2% 0.9%
8 CAR 72.8% 14.1% 13.2% 0.9%
9 S.J 74.0% 13.4% 12.5% 0.9%
10 STL 74.2% 13.4% 12.5% 0.9%
11 NYR 74.6% 13.1% 12.2% 0.9%
12 VAN 72.2% 14.3% 13.5% 0.8%
13 PHI 77.8% 11.5% 10.8% 0.7%
14 EDM 78.7% 11.0% 10.3% 0.6%
15 ANA 72.2% 14.2% 13.6% 0.5%
16 COL 75.6% 12.1% 12.3% -0.2%
17 T.B 78.5% 10.6% 10.9% -0.3%
18 BOS 79.4% 10.2% 10.5% -0.3%
19 NYI 73.9% 12.8% 13.3% -0.4%
20 CGY 74.6% 12.4% 12.9% -0.5%
21 DAL 77.1% 11.1% 11.7% -0.6%
22 FLA 74.6% 12.4% 13.1% -0.7%
23 NSH 74.9% 12.2% 12.9% -0.7%
24 OTT 75.6% 11.8% 12.5% -0.7%
25 ATL 74.8% 12.2% 13.0% -0.9%
26 MIN 75.8% 11.3% 12.9% -1.7%
27 WSH 72.1% 12.9% 14.9% -2.0%
28 BUF 71.7% 13.1% 15.2% -2.0%
29 DET 74.3% 11.4% 14.3% -2.9%
30 TOR 71.9% 12.0% 16.0% -4.0%

74.8% 12.6% 12.6%

What complicates all of this is that part of the reason teams spend a lot of time on the penalty kill is the fact they don't get scored on and therefore have more unbroken two minute chunks. Likewise, teams that have had difficulty scoring on the power play are more likely to be out there longer. (Columbus is a good example here, as they've got the league's top penalty kill but are middle of the pack when it comes to shorthanded situations. Because they've rarely been scored on, however, they're generally killing a full two minutes, which bumps them up in the 'percentage of time spent on the penalty kill' rankings.)

All that makes it more difficult to make blanket statements about New Jersey being a more "disciplined" team than the Red Wings — although even the NHL's own generic stats would bear that out.

The Behind The Net numbers also give you an idea of which teams play in more penalty filled games, with Columbus, Buffalo and Toronto all playing less than 72% of their games at even strength this season (~43 minutes). On the flip side, 79.4% of Boston's games have been at even strength, a difference of about five minutes a game.

There's a lot of data here I haven't even had a chance to look at; my guess is there's much more to be mined from these figures than what I've got here.



At 2:16 a.m., November 05, 2007, Blogger HockeyTownTodd said...

Nice number crunching, and interesting, if only to point out that short handed stats are worthless without breaking down the SH situations to 5 on 3 separate from 5 on 4. There are top teams at the top, middle, and bottom. 6 minutes of 5 on 3 would more than likely be more productive than 14 minutes of 5 on 4. And even these stats should be divided by situations where one of your top PK'ers are the 2nd man in the box.

At 9:32 a.m., November 05, 2007, Anonymous steve said...

One thing that I'd like to get rid of from traditional boxscores is the n/m method of reporting success on the power play (although over the course of the season it averages out). It really doesn't give enough information because it doesn't take into consideration 5 on 3, 4 on 3, or the length of the various power plays.

I think these numbers above are a good step in the right direction. I would like to see something like pp minutes played per goal.

At 9:36 a.m., November 05, 2007, Anonymous steve said...

Hehe... follow the link. Nice site

At 12:21 p.m., November 05, 2007, Anonymous Mogen_david said...

That was quick thanks. I guess one of the best ways to figure the impact of PK and PP on a teams performance would be PPG/G and PPGA/G as well as the % of each to a teams total goals. Although breaking down into situation would add information.
Oh, would it be worth adding Goalie pulled data or at least end of game goalie pulled information to tease things out a bit more?

At 12:41 p.m., November 05, 2007, Anonymous Mogen_david said...

Is their away to calculate TOI/oppurtunity? I dragged in the NHL data and they didn't look like they matched up well enough to just use those.
Although to be honest it feels like a dog chasing it's tail at this point. Between, TOI per situation, Goal per situation and the NHLs PPO and TS what more do we need to evaluate teams special teams performance. Perhaps penalties drawn taken in each situation?

At 2:02 p.m., November 05, 2007, Blogger Hawerchuk said...

I'll add more stats as the weeks go by, including for goaltenders. There's a ton of information that can be mined out of the NHL data.

Bonus: all of the statistics are finally automated. I skip over games where the goalie data is completely wrong and I'll go back and fix them later - that's about 2-3% of the games.


Post a Comment

<< Home

Links to this post:

Create a Link


Free Page Rank Checker
eXTReMe Tracker