Explaining the Drop in Crosby & Malkin's Ice Time

One thing that's new in this young season is that Sidney Crosby and Evgeni Malkin are seeing less ice time per game. It's certainly an open question whether their usage will stick, but I want to take a stab at explaining what might be behind the limited ice time for our top two centers. Part of this entails looking at different game strengths, and another part involves looking at what the Penguins do when playing with the lead.

Before we dive in, below is a graph that charts their overall time on ice per game (in minutes) over the last six years.

You can see the obvious dip this year. Other than in 2011-12 (when Crosby was coming back from a concussion) he's always been around 21-22 minutes per night. Malkin's ice time has actually decreased over the last five years, but he's dropped nearly two minutes this year. To figure out where this change lies, we need to break down their ice time by game state.

The chart below shows shorthanded ice time per game.

This doesn't explain anything. Malkin was getting literally no PK time per game after 2010, and Crosby only saw between 30-45 seconds per game on the penalty kill. Thus far, their usage while shorthanded this year is consistent with year's past.

Now onto the powerplay.

Again we see that Johnston is using Crosby and Malkin no less on the powerplay than Bylsma did. Through the process of elimination, that leaves even strength play as the source of Crosby and Malkin's drop in total ice time this year.

The drop-off this year is transparent. Crosby was typically around 16 minutes (17 last year), but he's only averaging roughly 14.5 even strength minutes per game this year. Malkin was closer to 15 even strength minutes per game in previous seasons, but now has less than 14 minutes per game.

I hinted above that I've wondered if this drop in ice time for Crosby and Malkin is due to the Penguins playing a lot with the lead this year. The Penguins have spent more time leading at 5v5 hockey this year than any other team by a substantial margin. And that's with some teams having played three more games than Pittsburgh. It's also not the case that the Penguins have simply led in a lot of close games--they've spent the most time leading by two or more goals. The inverse of this proposition is that the Penguins have spent very little time trailing during 5v5 play.

One way to see if this is really the answer to the diminished ice time for Crosby and Malkin is to look at what percentage of the team's total 5v5 ice time in a particular game-state Crosby or Malkin is taking. If the Penguins have 50 min of tied 5v5 hockey, and Crosby has 20 minutes of 5v5 tied hockey, then he's got 40% of the team's total tied ice time.

To make this simple, I picked each player's most recent season where they played nearly 82 games. For Crosby, it was last year when he played 80 games, and for Malkin, it was 2011-2012 when he played 75 games. I went to War on Ice and pulled the team numbers from the games that Crosby and Malkin played in for those seasons. Since they didn't miss much time in each year I've selected, this process didn't take long.

Look at that. Crosby's usage at 5v5 when broken down by game-state is astonishingly similar thus far. The reason he's got less even strength ice time per game is because the team has spent so little time trailing or tied this year. Indeed, Johnston is actually using Crosby more than Bylsma when the score is tied or when the the Penguins are playing catch-up. He is, though, being used less when the team has the lead, but if the Penguins played more hockey in different game states, Crosby could be much closer to his ice time from previous years.

A bigger sample (40+ games) for this season might paint a different picture, but for now there's no evidence that Crosby's even strength usage is actually a commitment to using him less rather than the fact that the Penguins have played well enough to jump out to early leads.

On to Malkin's numbers.

Malkin's graph looks more like genuine limited usage. You can see that he's skating a smaller share of the team's total even strength ice time in each game state. The trailing numbers are weird, though; you want Malkin playing way more than only 1/4 of the ice time your team spends trailing. I suspect this oddity is more a product of the limited time the Penguins have spent trailing so far than an intentional decision on Johnston's part to limit Malkin's play when the team needs goals.

Probably the best indication of how Malkin will be used is his 5v5 tied numbers, and they aren't that different from how he was used in 2011-2012. Indeed, you can comb through games last year and see that Crosby and Malkin played limited even strength minutes in games where the Penguins spent a lot of time leading.

	Crosby EV Ice Time	Malkin EV Ice Time
11/20/13 @ WSH	15:43	14:35
11/29/13 @ TBL	16:45	12:37
12/19/13 vs MIN	16:06	Did not play
1/3/14 vs NYR	15:19	Did not play
1/22/14 vs MTL	13:41	12:21
1/5/14 @ BUF	15:58	15:00
3/4/14 vs NSH	15:51	13:36
3/18/14 vs DAL	12:58	12:47
Average	15:16	13:29

I think it's also worth considering how injuries affected the team. Injured forward lines make it difficult to roll out depth lines consistently. As top nine forwards like Dupuis and Bennett go down, and their backups from the AHL suffer injuries as well, you'll naturally lean on your best players a little harder since you don't have as much scoring depth.

And think about what happens to injured teams. Injured teams are not as good, and teams that aren't good tend to spend more time trailing in hockey. What happens when teams trail? Your superstars (i.e. Crosby and Malkin) see more ice time as the game wears on. In other words, these injury effects combine to create a vicious feedback loop that will force your best players into more and more ice time

***

As I mentioned above, we're still dealing with a small sample, so we can't confidently say how Crosby and Malkin will be used as the year progresses. Our uncertainty is compounded by the fact that the Penguins have spent the overwhelming majority of their games with a lead, making it tough to figure out how Johnston will use his players when his team is tied or trailing. Regardless, we know that Crosby and Malkin are being used the same on the PP and the PK, and that so far, their even strength ice time has decreased.

For Crosby, it looks like his smaller portion of even strength ice time is due to the fact that the team hasn't trailed that much. For Malkin, he does seem to be used less, though his 5v5 tied usage is similar. We'll need more data to be sure of what's happening. But for now, this isn't something we haven't seen before. If the Penguins spend little time trailing--and remain healthy--you won't have to rely on your best players all that much.