Reason for Pessimism

I've finally been moved to post by something that has been as rare as hen's teeth this season: a whale of a hockey game. There is nothing quite so youthfully exuberant as a teenager screaming expletives in ecstasy on a HD broadband broadcast. Shootouts still suck, though.

Anyone who watched that game will recall a 4-on-3 PP in overtime. Ryan Parent was the right defenseman for Canada for nearly the whole penalty kill. He had several opportunities to clear the puck, but couldn't because he was on his backhand. A right-handed shot in that position could have easily relieved the pressure. That, my friends, goes a long way in explaining why Bob Gainey valued David Fischer so highly in the last draft.

Now down to brass tacks.

With roughly half the NHL season gone, the Habs appear to be in very good shape. They've put up the second most points per game in the East. Does that make them contenders? Let's see...

There is an item at The Hockey Project (RIP) that rates teams across eras using a measure called Pythagorean Dominance:

Pythagorean Dominance (PDOM) = (PW% - .500) / CB

The PW% is the Pythagorean win% of the team in question, (GF² / (GF² + GA²)). CB is the competitive balance, which is simply the standard deviation of the league-wide win percentages.

I tinkered with the formula a bit (used the season's average PW% instead of just .500, and used the standard deviation of the PW% instead of the win%) to come up with this list of recent Cup champs and runners-up:

No champ in the last 17 seasons has had a PDom less than .55. The average is 1.15, so let's call any team with a PDom over 1.0 a serious contender.

Here are this season's teams with PDom over 0.55:

At 0.80, which is good for 6th in the league, the Habs look like dark horses. They're better than the average Cup loser since 1989. Sound encouraging Habs fans? Well, unfortunately the Canadiens' goal differential has been propped up by an outrageous differential in empty net goals. They have scored 9 and allowed only 1 this season.

Here are the PDom ratings of the last eight Cup finalists, once ENGs are removed:

As expected, they're not too different from their PDom ratings when ENGs are included. Now, here are the PDom ratings of teams this season when ENGs are omitted:

The top five all retain the same positions, and all remain above 1.0. The mighty Canadiens drop below the 0.55 threshold, all the way down to 11th place at 0.45.

All of the best team ranking systems out there utilize goal differential. Right now, the Canadiens' freakish ENG differential is making them look better than they really are. It's not unusual for good teams to have a positive ENG differential. After all, a good team is more likely to be leading by a goal late in the game than a bad team. But a +8 after half a season is huge. In the last four seasons, the biggest ENG differential over a whole season was +11 ('04 Isles and '03 Sens). Only four teams have had a +10 or better. This cannot continue.

Although they're an amazing +21 (50 GF, 29 GA) on special teams so far this season, Montreal is -12 even strength (60 GF, 72 GA). Even if they can maintain the red-hot special teams (which is unlikely), that does not bode well for playoff hockey when the number of penalties drops.

Don't get me wrong - we're seeing an above-average team this season. But unless we see some serious improvement at even strength, I can't consider them serious contenders.

Time for a commercial break.


Blogger E said...

i'm thrilled that you've posted again, really, but the last thing i needed was more reasons for pessimism. does anyone who follows the habs need encouragement to be negative?

honestly, though, i believe your analysis, but i think you're basic position (i.e. that the habs aren't as good as the surface stats might suggest) is more or less the majority position these days. outside of montreal, i don't think anyone does now nor ever did consider them a serious contender- just a team doing better than predicted.

1/05/2007 1:26 a.m.  

Post a Comment

<< Home