I read somewhere once (I think from Jesse Marshall) that around 450 minutes of even-strength game time, or around 9 games, was a reasonable sample size to start drawing conclusions about a team based on their advanced stats in that time.
So now that the Penguins are 10 games and 480 5v5 minutes into the season, I wanted to take a look at their fancy stats to see what we can see.
|
Percentage |
Rank |
Corsi For |
48.84 |
20th |
Fenwick For |
49.41 |
19th |
Shots For |
48.25 |
25th |
Scoring Chances For |
51.05 |
11th |
High Danger Chances For |
51.31 |
14th |
Expected Goals For |
49.23 |
21st |
Those numbers are largely not good and, with the exception of scoring chance generation/prevention, rank near the bottom third of the league.
Of course, anyone who is actually watching the games would point out that the Pens have looked different and, in fact, better over the last five games, coinciding with the Western Canada road trip. So let's take a look at the numbers in smaller, five game increments.
First 5 Games
|
Percentage |
Rank |
Corsi For |
46.26 |
22nd |
Fenwick For |
45.87 |
23rd |
Shots For |
44.00 |
26th |
Scoring Chances For |
47.63 |
22nd |
High Danger Chances For |
51.36 |
12th |
Expected Goals For |
47.34 |
N/A |
Last 5 Games
|
Percentage |
Rank |
Corsi For |
53.26 |
9th |
Fenwick For |
54.05 |
5th |
Shots For |
53.48 |
9th |
Scoring Chances For |
55.6 |
4th |
High Danger Chances For |
51.68 |
19th |
Expected Goals For |
51.59 |
N/A |
These numbers are adjusted for score and venue because in many cases the Pens were either being blown out (5-1 against MTL, 6-3 against NYI) or blowing the other team out (5-0 against VAN, 9-1 against CGY). When one team has a comfortable lead and can afford to sit back more, it can influence the way the stats look at the end of the game. We want to make sure we're accounting for that.
So what do we see? Well, the Pens of the first 5 games just stunk. The numbers reflect what we saw on the ice: a listless, unmotivated team with little care for details or defense.
But the Pens of the last 5 games look like the real thing, posting top-10 numbers in most categories. And that, too, matches what we saw on the ice. Even when adjusting for the blowout nature of some games, they were controlling play effectively.
There are, of course, some causes for concern.
First, the High Danger Chances number has not changed much (even though where they rank league wide did). And while they're still generating more than their opponents, the fact that it didn't increase in this recent 5 games stretch seems to indicate that they're still sloppy in their own zone and better goaltending may have served to cover that up.
Additionally, some of these games came against expected non-playoff teams in Vancouver, the Islanders, and possibly even Edmonton. Will the Pens still be able to play at the same level against better competition?
Finally, we should look at the Pens' PDO. It's at 1.037 for the season and ranks 4th in the league. Over the last 5 games, it was even higher at 1.067. Both of those numbers (especially the second one) are likely unsustainable, even for a team with this much offensive talent.
If and when those numbers come down, however, the Pens should still be able to win games if they can drive play as they have over the last 5 games.
But will they? 5 games is too small a sample size on its own. But, if you're looking for optimism, if those 5 games are a sign of things to come, the Pens may have fixed some of their early season problems and be headed in the right direction.
All data comes from Corsica.Hockey and naturalstattrick.com. xGF rank for 5 game increments is not available due to the nature of the Corsica site.