Wednesday, October 14, 2009

Simple stats that I don't see often: Special Teams Plus/Minus and Net Goals

When it's midnight and you need a photo of a cranky journalist ...

There was a point when I was like those many curmudgeon old sports columnists who looked down on complex statistics. Thankfully, though, there have been enough great stat blogs to finally teach me my lesson: some stats just don't tell you as much as you might originally think.

Still, there is a part of me that always searches for the right cross-section between simple and deep. I'd imagine that these kind of stats have been around plenty of times before, but I've never seen them expressed before in official sites' standings sections (at least on ESPN or NHL.com).

So, with all the Power Rankings out there, I thought it would be fun to occasionally take a look at NHL teams through completely objective - and obnoxiously simple - statistics. Obviously, this first few weeks' worth of stats will be greatly altered by outliers but you know that already, right?

(A few more notes: these stats were taken from before tonight's games and I'm dumb and don't know a better way to put up an Excel spreadsheet on blogspot, so please tolerate my sloppy use of screenshots)

ANYWAY, first the NHL's leaders in Net Goals.

This takes a SUPER sophisticated formula of Goals For minus Goals Against (although the spreadsheet also provides even strength goals ... which I might incorporate into something more elegant and interesting in the future).

If you're a Maple Leafs fan, well, at least you aren't running your car in your garage right now. Small victories. feel free to avert your eyes.

(Click to enlarge)

So those of you who hate simple arithmetic, you're welcome. The real reason I thought to do this, though, was to take a look at special teams numbers.

It's always bothered me that such an emphasis has been made on Power Play and Penalty Kill percentages but who gives a rat's ass about that? To me, PP effectiveness has always been about a) sheer quantity of goals and b) timeliness. There's no doubt in my mind that I'd take a powerplay that scored 2 out of 10 than one that scored 1 out of 4.

I'm aware that is an overly-simplistic criticism, but work with me here.

To take a more "big picture" look at special teams, I think it is also important to compare teams' PP and PK together. If your team can eek out a substantial amount of PP goals while keeping PK goals under control, you'll have a major advantage while attempting to make the playoffs.

So, I've come up with (OK, I bet someone else has done this too since it's super-simple) "Special Teams Plus-Minus."

The formula's almost as simple as "Net Goals"

PP Goals Scored + SH Goals Scored - PP Goals Allowed + SH Goals Allowed* = Special Teams Plus-Minus.**

* Just realized I didn't include SHG allowed but I'll do it next time. Promise!

** - However, if I've come up with something stupidly original feel free to call it a Jimbo Score. :)

4 comments:

Navin Vaswani (@eyebleaf) said...

I refuse to look at the spreadsheet.

Joe said...

XD @ the tags for this post

It's always bothered me that such an emphasis has been made on Power Play and Penalty Kill percentages but who gives a rat's ass about that? To me, PP effectiveness has always been about a) sheer quantity of goals and b) timeliness. There's no doubt in my mind that I'd take a powerplay that scored 2 out of 10 than one that scored 1 out of 4.

Uhhhhhh wut?

You'd prefer a 2/10 powerplay, as opposed to a 2.5/10 powerplay? Where were you when they were explaining fractions in school dude? Thats why rate statistics > raw numbers, almost every time. The only problem with PP% as a stat is that they count shortened PP's. Mickey Redmond once suggested something when I was kid about tracking it based on PPG/PPmins, in a game where the Wings had a bunch of abbreviated power plays. I always thought that would be a good idea.

CLS said...

I prefer 2 out of 9 over 1 out of 4 in a single game. My point was that I'd rather look at total season PPG instead of total season PP percentage.

I can see why you would get confused with the wording but that's what I meant.

CLS said...

To expand on that idea, a team might also create more powerplay opportunities, which might water down their percentages.