This is one of the best article I read recently. I found it from playerrs.com
How did NHL teams perform last year on special teams and at even strength? San Jose had the best PK unit and Montreal had the best PP unit in terms of percentages, but there's no established way of bringing the PK and PP numbers together to get a picture of overall special team performance. I've seen the two numbers added together, but that seems just a little too improvised. Is there a better way?
Team plus-minus
I suppose you could consider something like a team plus-minus to measure performance at even strength. A team plus-minus would let you see which teams were giving up more than they were getting as far as goals were concerned. You'd have to assume all goals are equal (which I don't), and you'd still be hard pressed to make comparisons between special team and even strength play, because, on the one hand, you're using a percentage as a measure of effectiveness, while on the other, you're using net goals.
If you've seen some of my earlier posts on how to value goals, you know the approach I took to this problem. Goals scored at competitive points in a game (a tie score or a one-goal game) are more valuable because they have a larger influence on the points that the teams involved can expect to collect. These competitive situation goals matter more than goals scored when the game's pretty well decided.
Teams don't often come back from three goals down. Even if they score two to draw close, neither of these goals is as important as the third one that actually ties the score. Without that third goal, the other two are just moral victories; they make the game more entertaining without altering its eventual result. It's the third and tying goal that's, in fact, game-altering. Game-altering goals are the ones to focus on to effectively measure player and team performance.
Making the comparison
So what I did was go through all the even strength and special team goals that were scored last season, taking into account the context in which they were scored (the score and the time remaining), and assigned a value to each goal based on the change in the points that each team could expect to collect after the goal had been scored. Then, it was just a matter of adding the sum of the value of all the goals that a team scored and subtracting the sum of the value of all the goals that a team allowed both in even strength situations and during special team play to get the numbers behind the following chart.
NOTE: Strong even strength units are to the right and strong special team units are on top.
Some observations
It turns out the Canadiens and the Sharks did in fact have the two best special team units in the league, with the Flyers not far behind. At even strength, however, all three teams tended to give up more valuable goals than they scored. I'll try to elaborate on this on team-specific posts in the coming days.
The Red Wings and the Ducks, meanwhile, were the class of the league at even strength. But, whereas the Red Wings gained value from their special teams as well, the Ducks were let down last year in this department, faring not much better than the lowly Leafs and Blues. This was a big drop from the team's performance in the previous (Stanley Cup winning) year, in which their special teams ranked among the league's best. Apparently, Mathieu Schneider does not equal Scott Niedermayer, and Todd Bertuzzi does not equal Teemu Selanne.
