I only saw the last half of last night’s Oilers game, as I had tickets to the Leafs-Sharks game. All sorts of interesting stuff happened but one of the things that caught my attention was the length of Phil Kessel’s shifts. I don’t watch a ton of Maple Leaf games, but I’m familiar with beer league hockey players who stay on the ice too long and I recognize the signs - standing away from the scrum for the puck and just sort of hoping that it will pop out in your direction, resulting in a scoring chance. It’s not a great tactic, as if the other team has fresh legs out there, you end up with the opposition winning more pucks.
I went and looked after the game and, sure enough, Kessel is amongst the league leaders in shift length for forwards, with an average length of 52 seconds. Then I noticed something weird: the list is decidedly skewed in favour of the Eastern Conference. So far this year (min. 40 GP), the Eastern Conference has 26 of the top 30. Only Brad Richards, Corey Perry, Ryan Getzlaf and Jarome Iginla prevent the sweep. Lots of guys who aren’t stars take long shifts too - Doug Weight, Kyle Okposo, Brendan Morrison and Brooks Laich are all taking long shifts.
Turns out that this is something of a trend. In 2008-09, the East had 24 of the top 30 spots (Mike Ribeiro, Richards, Getzlaf, Iginla, Steve Ott and Loui Ericksson; Dave Tippett was out of step with the other Western coaches). In 2007-08, the East had 26 of the top 30 (Ribeiro, Iginla, Joe Thornton and Brenden Morrow); Sam Gagner was actually tied for thirtieth). In 2006-07, the East had 24/30 (Ribeiro, Petr Sykora, Joe Thornton, Patrick Marleau, Iginla and Jonathan Cheechoo). In 2005-06, the East went 24/30 again (Nikolai Zherdev, Doug Weight, David Vyborny, Petr Sykora, Sergei Fedorov, Rick Nash). There are some ties at the bottom of the top 30 but the trend seems to be pretty clear. Interestingly, if you look at TOI/G, the split is smaller. Counting backwards from 2009-10, I get the East with 17/30, 15/30, 19/30, 18/30 and 22/30.
It seems to me that there are a couple of potential reasons for this. These could include Eastern coaches letting their players stay on the ice longer or Eastern players tending to have more defined roles - it seems to me that long shifts tend to really occur on the power play and if you only play ES and the PP, your average shift length may well be longer than someone who plays the same amount of ice time per game but who kills penalties - purely on the basis of observation, it seems that penalty killers are more aggressive about getting changes.
I am inclined to think however that, for whatever reason, Eastern coaches are more permissive when it comes to shift lengths. I’m also inclined to think that it’s a bad idea. There’s some interesting work that could be done here to see if anything good happens when players have been on the ice at evens for more than a minute. My guess would be that the numbers fall through the floor, for everyone - if the other coach is getting his changes, and you’re out there with a minute’s worth of lactic acid in your legs, it seems to me that even if you’re Alexander Ovechkin and better than the guys who just came over the boards, that difference is going to be swamped by the fact that you’ve been skating for a minute and they haven’t.
One wonders (this is the speculative part of the post) whether this might account for some of the difference between the East and West. If the Eastern teams all play this way with one another, there’s effectively no cost to them for doing so - Kessel’s 1:30 shifts are balanced by Ovechkin’s, in a manner of speaking. If they then go west and try to do the same thing, with coaches who don’t play that way, it might start to bite them a little bit. I’d be very interested to see the data on this but it seems to me that being on the ice after a minute is sort of like being in a bar after 1:00 a.m. - there’s no guarantee that something bad will happen, it’s possible that something good will happen but the odds are slanted heavily in favour of something bad. It’s an interesting area for someone to dig into.
this is going to sound nutty, but how about the elevation in some of the western conference cities playing a part? every time someone out east plays the avalanche, people always talk about getting shorter shifts and not getting caught out too long. i don’t know how high up phoenix, calgary, and edmonton are, and i’m too lazy to look to support my unfounded theory, but that could certainly make a difference.
fwiw, as a devils fan seeing kovalchuk take super-long power play shifts, he “seems” to have the same jump at the first second as he does at 1:30, but that’s because he’s skating as little as possible throughout the shift.
Perhaps it’s disparate travel schedules causing the difference?
I meant to mention this the other day somewhere but where does our own 78 stand in terms of shift length?
I know that Bruce still isn’t over the time he stayed out too long in Nov of ‘08 against the Wings;) and here lately I’ve been noticing two little things about Pouliot’s game: his shifts seem to almost always last longer than his wingers and he gets kicked out of the faceoff circle more than anyone I can remember.
I doubt there’s a stat that backs up my second hunch but I think the first one’s correct; at least in terms of his shift vs his linemates.
I don’t think anyone has done a thorough study of it but J.P. looked at the issue as it pertained to the Caps in the playoffs last year (http://www.japersrink.com/2009/5/4/859460/short-shifts-and-long-runs) and came to somewhat similar conclusions: more bad stuff is happening on long EV shifts than good stuff. The part about Tortorella game-planning for tired legs in the NY-Wsh series is also interesting. It seems that at least Tortorella believes there was an advantage to be had there. Jonathan also looked at the issue (http://www.coppernblue.com/2009/5/9/870697/shift-length-boudreau-tippett-and) but gave a more positive spin on long shifts.
One of the guys Jonathan identifies as a “long shift” guy is Dave Tippett who moved over to Phoenix. So far they’re not taking longer shifts under Tippett than they were under Gretzky, so it would seem Tippett has changed strategies with a new roster. But the change from Tippett to Crawford in Dallas has resulted in some big swings. Ribeiro is down from 0:58 to 0:49, Morrow from 0:59 to 0:46 with everyone else coming down a few seconds and the whole range moving from 0:44-0:59 to 0:39-0:52.
Interesting stuff, Tyler.
I would venture a guess that a large part of this hinges on the stylistic differences between the conferences. For the last several years, the Eastern Conference has been all about scoring lots of goals and not playing defense. In the West, you see a lot more defense-oriented hockey. Teams like the pre-lockout Flames, the Ducks from 03 to this year, the 06 Oilers, the pre-Crawford Stars, the Wild… None of those kinds of teams exist in the East.
As a Wings fan, following one of the higher scoring teams in the West, I’ve noticed that no one in the East plays like those teams that used to knock Detroit out earlier in the decade. The 03/07 Ducks, the 04 Flames, the 06 Oilers… They were all about defense first, and if you came near their net, they’d knock you flat on your ass. Compare that to the cakewalk that the Pittsburgh Penguins have enjoyed the last two years in the playoffs. I mean, honestly, that Carolina/Pittsburgh series was just plain embarassing. If the Penguins had to go through a series like the Wings/Ducks did last year, I think the SCF turns out different. But everyone in the East lets it happen, because that’s where the marketable faces are. Because those guys can go to the front of the net and not pay a price, because they’re not having the energy physically ground out of them, they can afford to stay out longer, especially when everyone plays by those same rules.
All of this is purely anecdotal and based on observation, as well as a definite bias in that I can’t stand the Eastern Conference for this reason. But I think there’s some truth to it.
Re:Pouliot
So far this season he’s on the low end in terms of shift length for the Oilers (0:43 for Pouliot, the team range is 0:43-0:53 for guys that have played more than a few games). He’s been short-to-medium relative to teammates over the last three seasons.
Scott: That’s a big jump up from MacTavish.
I don’t think it is Jon; they look pretty similar to me. Last year’s range was 0:39 to 0:58 and the year before was 0:42 to 0:58. The top end guys are all PP guys but that’s true for Quinn too.
Calgary’s elevation is around 1,100 metres. Not as high as Denver, but in the neighbourhood compared to a lot of coastal or near-coastal cities out east.
When I’ve looked at the East/West records, the East gets hammered at home too (relative to what we’d expect the home team to do). I don’t think it’s an altitude or travel thing.
I feel like Vic has talked about this a few times. In fact I think he’s posted about it and I think he mentioned that Neilson used to chase after tired legs on a regular basis.
Definitely an interesting topic.
Ah I remember. I think the Nilsson post has something to that effect - MacT used to give Nilsson shifts against tired legs IIRC.
Randomly, I have a post almost done on this topic for tomorrow. Shift length has been slowly decreasing since they started collecting the data. Harry Sinden won with long shifts though, so maybe it’s a good idea.
it’s cuz the East is a weaker conference. the est has a
38% win percentage vs the west
When I’ve looked at the East/West records, the East gets hammered at home too (relative to what we’d expect the home team to do). I don’t think it’s an altitude or travel thing.
Well, it would probably rule out the travel factor, but if the western teams are training in higher altitudes then it would make sense that they’d have an extra bit of gas in the tank at sea level.
Ok, I hit you up with some data on optimal shift lengths:
http://www.behindthenethockey.com/2010/2/11/1305245/optimal-even-strength-shift-length
If I’ve read this right, the list you had included PP situations? If so, Weight’s is probably explainable by the fact he plays the point on the PP, creating very long shifts. Then again, when not on the PP he’s getting spot duty on essentially a checking line that is more prone to getting hemmed in.
Come to think of it, during the many times Weight has been out this year, Okposo replaces him on the PP point. But I think Okposo is getting overused, period, and getting a long leash or being pushed out there for long shifts would be another indicator of that.
What about the possibility that there’s more depth in the West, so less incentive to keep your best players out as long as possible? Maybe you keep Kovy/Ovie out in the hopes of catching their fourth line out and making hay, but against anybody but Edmonton that doesn’t work against Western teams? Just a thought based on the inter-conference records.