Monday, January 17, 2011

Price vs. Price

40 Games By Stats

Carey Price just played his 40th game of the season on Saturday night. An accomplishment in itself for the oft-beleaguered netminder of 2009. As it happens 40 games provides an interesting point to have a more in-depth look at Carey's stats.

As a whole, we rated Carey as a solid A last week, and heard arguments for A-. Overall, I think we'd all have to admit there's been some brilliant goaltending along the way. That's not to say there aren't lingering concerns from our previous roller coaster seasons with the young goalie. Is this an excellent season? Or did Carey just have one excellent stretch? On to the stats...

The 20-20 dichotomy

The 20th the game of the season was Carey Price's 4th shutout. It propelled him to unthinkable numbers: 2.05 GAA and 0.932 Sv%. It was his 19th dome in 19 starts. 12 wins from 19. Critics, we heard were silenced.

At the turn of the 40th, that 31st save clinched a 22nd win and firmed his above average output of 2.37 in GAA and 0.919 in Sv%. At 40 games it is still critic-silencing stuff. He's a winning goalie challenging in important stats categories. Yet his early excellence prompts some examination.

What does it take to get from 0.932 to 0.919? How many goals a game does a goalie have to allow to slip from 2.05 every 60 minutes to 2.37?

Team output?

This is the favourite of many. While it is certainly true that in some games Carey received incredibly poor support from his teammates. the averages tell us that it went both ways. In fact, if you look below, you can see that the Habs scored more goals, allowed fewer shots and fewer chances over the 20 games. If all else was equal, one might have expected a stabilization in stats, if not an improvement.

Looking at things this way is far too simplistic, i realise. But it remains that this is one of the things to examine in this puzzle. It is not the main part of any explanation it seems, from a quick check at least.

A return to earth?

Some stats, no matter how much we want to believe otherwise, are probably just too good to be true. In stats circles, there's endless talk about regression to the mean (they don't like Tim Thomas all the time). In Carey's case, the stat that stood out like a sore thumb was his SHSv%. Over the years I have kept an eye on these things, and can tell you that anything over 0.850 in this category is exceptional. For a goalie to post 0.917 over a season would thoroughly smash established marks and norms. All to say, it probably wasn't going to last.

Since the 20 game mark, Carey has posted a very very good 0.869 on the PK. However, the return to very good from out of this world is at least partially responsible for the drop in all numbers.

A slip in preparation?

At the beginning of the season (at least after those preseason debacles), I marvelled at Carey Price's preparation. Every game, he would come in and just look an impenetrable fortress. His numbers tell the tale.

Below are two charts. The first shows his stats by period. One can see clearly here that in the first twenty games, Carey was truly outstanding at the beginning of games. Over 0.948 for two periods. The second chart shows the stats I have been tracking on when goals go in. Again, Carey was just amazing at Game 20 here. He was allowing the first goal only after an average of nearly 15 shots were taken (and nearly 7 quality scoring chances) -- this gave him game starting numbers of 0.937 and 0.869 for chances. Both outstanding.

But has something happened? The second twenty show little change in play after the second period, but now Carey is a 0.910ish goalie in the first frames. What's more, he's been allowing goals after just 9 shots instead of 15, and 3.6 chances instead of 6.65, dropping his starting game numbers significantly.

Perhaps this is only luck, or a regression to the mean as well. But I'd swear that he's looked different.

The next 20 games are critical

This is all very interesting, but I don't think this sample answers many questions for us just yet. I find it too difficult to dissect from the numbers just now. The slip in preparation teamed with a return to earth seem most likely to hold the key explanation, but we've yet to see.

The biggest question has to be about what that first stretch of the season was. Was it a pleasant surprise? A fortuitously timed one-off streak? Or was it Carey taking his play to a new level? The next 20 games will help us answer which streak (the incredible one or the latest below average spell of ten games) is the real outlier in his play.

For anyone who watched those first 20-30 games as closely as I did, it is tempting to suggest that the real Carey Price played in October and November. But getting real answers takes time. This is an interim update, the final analysis is yet to come.

No comments:

Post a Comment