Tuesday, April 6, 2010

Habs a strong even strength team?

In the current context, with the Canadiens struggling to score goals at any time, let alone 5-on-5, to suggest this is a strong even strength team appears quite ludicrous. But when you're good at math and know which statistics to look for, that seemingly absurd notion stops sounding so crazy.

Alas, I'm not good at math, nor do I know which statistics to look for. But Timo Seppa, a writer at Puck Prospectus, has crunched the numbers and decided that Montreal's play since the Olympic break makes the Habs a prime darkhorse contender to come out of the east. His logic appears pretty sound, namely that since Andrei Markov's return from the injury the Canadiens have become a much stronger defensive team at even strength, which offsets some of the current offensive woes.

Seppa also goes through an exercise I tried to do about a month ago, projecting what a fully healthy Habs squad would like. Except he uses something called GVT, which stands for Goal Versus Threshold. It's a statistic that Puck Prospectus author Tom Awad devised in order to be able to compare offensive players with defensive players and goaltenders to know who is most valuable to a team. Or at least that's what I can understand. The explanation of GVT by Awad is in two parts, part one is here and part two is here.

Seppa's number crunching also presumed that Jaroslav Halak would have started 10 more games at this point, which brings us to today's news that he was named the NHL's first star of the week in a bit of a no-brainer decision.

The crux of Seppa's thinking is that if the Canadiens had no injuries to their key players, and if Benoit Pouliot spent all season in Montreal and if Halak had played more often, the Canadiens goal differential would improve by about 30 goals. That's a very significant jump. He uses that to determine, seeing as all of the factors he used are currently in place, that the Canadiens will be a very difficult out in the first round of the playoffs and beyond. He concludes his piece like this:

"With all of their stars healthy and with Jaroslav Halak in goal throughout the postseason, there’s a strong case to be made for Montreal being the fifth best team in the playoffs and the second best team in the Eastern Conference. Translation? The second through fourth seeds––Pittsburgh, New Jersey and Buffalo––should be wishing for anyone but Montreal in the first round, because those are upsets in the making. And, believe it or not, the Habs might just be making the Capitals sweat out a long series in round number two. And from there on…You never know.
Montreal? Nous croyons."
The methodology appears sound, but I'm not sure if there's a way to statistically project how a team will respond in the playoffs. Still, I thought Seppa's work would be of significant interest to all of you.


Dann said...

I`m game with that. I`ve been trying to hold off on my analysis of Scott Gomez and his team until watching them in the playoffs. Cheers to hoping the best is still ahead.

Bryne said...

The way the Habs have been playing since the Olympic break should be a concern to any first round opponent.

According to ESPN's stats (they have a quick post-olympic split option) Montreal (at + 0.69) is second only to Detroit (0.83) in GF/game vs GA/game differential.

At .931 The Hab's have the best save% in the league since the break.

And somewhat surprisingly though quite possibly a huge factor in the post-Olympic success is the turnaround in penalties. Currently the Canadiens have the 4th best differential, having enjoyed a man-advantage for 33 minutes more than they have been short-handed.

Perhaps there is some method to JM's madness .... system .. afterall.

ayc said...

The biggest drawback to such analysis is that the same exercise must be applied to all other teams as well.

Sens lost Alfie and Spezza for a combined 33 games and then we could assume that Elliott would have played 15 more games than Leclaire did.

There is no doubt that Boston has been greatly affected by injuries as well, with Savard and Lucic out for nearly half the season.

Nevertheless, I'm sure the Habs would still rank high, but it might not lead us to conclude that they could be legitimate contenders.

MathMan said...

And meanwhile another respected stats guy, Gabe Desjardins, is making the case that the Habs aren't very good if they can't beat Toronto. ;)

I'm all for optimism and it's certainly the case that the Habs have been playing better lately -- though I do agree with 'ayc' that the projection exercise would need to be done for other teams. I'm just a little concerned in that I'm not entirely confident that the vast improvement in ES defense has been done by reducing ES chances.

But I'll certainly buy the possibility of a first-round upset... assuming the Habs make the playoffs. ;)