IPP for Defensemen - Western Conference
Yesterday, I began looking at individual point percentage for defensemen, and to no one's surprise, the numbers tend to be lower for defensemen than forwards. Whereas the average forward earned a point on about 69.5% of the goals scored while he was on the ice, the average defenseman only got a point on about 28.5%. That lower number comes as no surprise; forwards, after all, have a lot more to do with goals being scored than do defensemen. But it also means that we're more likely to see a defenseman put up a very good number based on luck since each point is worth relatively more compared to the average. That's important to keep in mind as we now turn to look at the results for each team in the Western Conference.
In case you need a refresher, the results here are supposed to give us a better idea of which guys are handling the puck a lot in the offensive zone and helping to create goals. In the case of defensemen, that most likely takes the form of being willing to jump into the rush or pinch down from the point when the opportunity arises. When we look at the results over a long period of time, the results generally make sense, but in any one year, an unlikely candidate might put up a particularly good or bad season before regressing to their personal average. The raw data here all comes from Gabriel Desjardins' behindthenet.ca. Now on to the teams (in alphabetical order by city including every player who was on the ice for at least ten goals-for):
Cam Fowler has a promising rookie season here, and I expect that he'll end up being a very good offensive defenseman based on the tremendous success he had in the OHL in addition to his NHL debut. But his teammate, Lubomir Visnovsky, is one of the best offensive defenders in the game right now. His 35 five-on-five points is the highest total in the league, and he did it while playing against tough opposition night-in and night-out. I think he probably deserved the Norris Trophy last season, but I'm not at all surprised that he didn't get it. Andreas Lilja and Andy Sutton, meanwhile, show you what can happen when a small sample and luck come together.
Jay Bouwmeester isn't bad exactly, but he isn't good either, and when you're paying the guy $6.68M per season, you kind of need him to be more than "just" an elite shut-down defenseman. If this was just a one-off, it wouldn't be a big deal, but Bouwmeester has been at about this level in at least each of the last four seasons. He's still a very good player, and the Flames only spent money (and a third-round pick) to acquire him, but I can't help but think that the team was expecting him to be a lot better offensively.
With Chris Campoli and Brian Campbell moving to other organizations, the Blackhawks will be relying heavily on their top two defensemen (and new addition Steve Montador) to provide offense at even strength. The numbers from this year suggest that there's a gap between Keith and Seabrook, but those are probably misleading. If you look at the longer track record for each player, Keith actually does quite a bit better than his partner.
Kyle Cumiskey didn't get a lot of reps this year, and really hasn't had a lot of reps in the NHL, but he's a guy I really like to watch. He's probably one of the fastest skaters I've seen, and in 2010-11, he showed some flashes of offensive ability. Avalanche fans have discussed the possibility of him going on waivers. So should the Oilers put in a claim? Probably not. He's not a very big guy, he's had to deal with some injuries, and prior to 2010-11 he was just average offensively (an IPP of 27.8% on 54 goals-for from 2007-08 to 2009-10). Still, I just like to watch the guy play, so I'd probably be excited to have him around.
Kris Russell looks good here and that number is very close to his four-year total of 34.1% (on 126 goals-for). But that number may actually be underselling Russell's offensive ability. Russell was a very good offensive defensemen in the WHL, but had a difficult transition to the NHL. In his first season, Russell only got about ten minutes per night at even strength, and scored just three points five-on-five. Over the last three years, he's earned 40 points on 107 goals-for five-on-five for a very impressive IPP of 37.4% even as his ice time (and responsibility) has increased since his time as a rookie.
Trevor Daley is one of those guys who seems to have snuck into the league without anybody (or at least me) noticing. He's been quietly excellent in Dallas for a few years now, posting excellent even strength offense in a top four role. The Stars recently locked him in with a six-year deal with a $3.3M cap hit, a deal that I think they'll be very happy with.
I know that he doesn't need the money, but it's hard to believe that Brian Rafalski retired after playing so well in 2010-11. The big surprise for me on this list is the guy who will probably takes some of his minutes, Niklas Kronwall. Not only was he good in 2010-11, he's been good for a long time now. Kronwall is actually one of the league leaders in IPP for defensemen over the last four years at 41.4% on 198 goals-for. With Rafalski moving on, I think there's a good chance that Kronwall will get a spot on the first unit power play for the first time and have an absolutely huge offensive season with the Red Wings.
Theo Peckham's presence near the top of this list comes as something of a surprise even though I watched him all year. His partner, Tom Gilbert, had a very poor year offensively by his standards, and it seems like some of that was probably bad luck, with a whole lot of "too much responsibility" mixed in. Ryan Whitney is unsurprisingly leading the pack with (yet another) percentage that looks like it'll be coming back down this season. I wonder how much of this natural regression will get blamed on that injury of his?
Matt Greene? Really? No. Not really. In his previous three seasons, he has a combined IPP of 23.5% on 98 goals-for, which is a lot more like the Matt Greene I remember. He's definitely been better in Los Angeles than he was in Edmonton, and may even be league average offensively at this point (which is astonishing), but this is probably the last year he leads the Kings in this category.
Brent Burns is a pretty impressive player, and his departure really hurts the Wild. On the other hand, his arrival in San Jose makes the Sharks a whole lot better, and they're one of the teams I'm counting on to beat Vancouver, so I think that trade worked out great! As for Cam Barker... his departure doesn't hurt the Wild quite as much. It's just one season, but that's pretty sad stuff for an offensive defenseman. Thankfully, his total over the last four years - 30.4% on 115 goals-for - is much better.
I know he's only got a couple of years in the league, and that the club needs to save money where they can, but the Predators might really regret giving Cody Franson to Toronto just to have them take on a bad contract, especially if they end up moving one of Shea Weber or Ryan Suter in the next year or two. Franson will regress substantially and may end up regressing all the way back to average, but his strong offensive seasons in the AHL suggest that he may just be coming into his own as an NHL defenseman. What a great trade by Brian Burke.
None of the team's forwards were very far above average, so it comes as no surprise that a bunch of defensemen do well, and that Keith Yandle in particular has a pretty dominant performance. I haven't watched a lot of the Coyotes, but it makes me wonder how much of Yandle's numbers relate to the team's system, which (in the little bit I've seen) seems to emphasize a defensive shell and a strong counter-attack that includes defensemen jumping up into the play to create odd-man situations when the opposition is less organized. Any Coyotes fans (or observers) care to chime in?
Well... at least everyone that managed a minimum of thirty events is where they should be. Dan Boyle is consistently among the league's top men by this measure with an IPP of 36.3% over the last four season (212 events). The big loss is Ian White, but he was replaced by the much better Brent Burns. Put Boyle and Burns alongside Vlasic and Murray, and the Sharks now boast one of the very best top four groups in the league for just $15.82M in cap space. That number will be heading upward next season, but they'll be getting tremendous value in 2010-11.
The Blues sure did buy high when they traded for Chris Stewart and Kevin Shattenkirk, and Erik Johnson is a high, high price, especially if Alex Pietrangelo regresses toward league average too. The club is putting a lot of faith in percentages here, which is always a little bit scary, but at least the guys they've chosen are all very young, so the possession metrics might improve even as the percentages decline.
Vancouver's best defenseman is Andrew Alberts, and they were one of just two teams to have a regular player register zero points five-on-five (the other was the Flames). What a terrible team.
14 comments
|
1 recs |
Do you like this story?
Comments
Just a general question, why do (almost) all of the articles on this site focus (or use exclusively) 5-on-5 numbers? I get that the PP isn’t “real” hockey I suppose, but games are decided by who scores the most goals, regardless of situation.
I think it’s perfectly reasonable (and smart) to separate different situations, however I think it is just as pertinent to look at PP and SH numbers. This isn’t directed specifically to you or this article, just something in general about this site I’ve noticed and always kind of wondered about.
If you guys have some sort of reasoning for focusing on 5v5 I’d be more than willing to listen though.
I don’t want to answer for Scott, but last year the average team took 291 penalties. If we assume the average penalty lasts 90 seconds, that’s about 436 total minutes. That’s only equal to about 7 and a quarter games over a season. So 7.25 games on the PP, 7.25 games on the PK, what’s left is about 65+ games spent solely at even strength.
I also think that in general we know way less about the power play and penalty kill – there’s not enough data over a single season to make conclusions.
http://drivingplay.blogspot.com - The blog with three first lines
Two interrelated reasons, which Triumph touches on above:
1.) ES is about 75% of the game in total. That makes it far, far more important.
2.) Because of the relatively small sample size, isolating ability versus variance in special teams is problematic. What’s more, only a select segment of each team’s roster plays on the PP or PK.
Triumph and Kent have both pretty much answered your question, but I will say that I’m hoping to do more work on odd-man situations (we started to do it a bit with the Hemsky stuff, but came to a lot of murky non-conclusions).
The biggest fanana of the Havana Bananas.
by Scott Reynolds on Aug 29, 2011 9:21 PM MDT up reply actions
Oh yeah, obviously I know that ES makes up the vast majority of the hockey season, however I would also guess that the percentage of time spent at even strength is greater than the percentage of goals scored at ES, so just comparing straight across time of each is not as relevant (IMO) as looking at the number of goals scored at uneven strength. I’m sure EV is still the vast majority of the goals, I was merely pointing out that EV is often the only situation shown on this blog, and I find it important to look at everything, as that will give us the best picture.
The PP work done about Hemsky was very intriguing, and I think it would be beneficial to at least attempt to better understand the stats behind odd man situations.
All that said, you guys do great work and I love reading your articles, so keep it up!
The scoring rate at ES is significantly lower than the scoring rate on the PP. Now that everyone is finished saying “Well Duh!”, I think the reason that so much time gets devoted to ES over PP/PK is that there is far more data in order to compare with.
When a guy shoots 4% over his career average at ES, it has a big impact. ES shooting % is remarkably stagnant (7.2-7.8% over the past few seasons). When he shoots 4% over his career average on the PP that’s almost normal because there is a much wider variance on power play shooting.
Just as an example, if you compare ESP and PPP per 60 over several seasons (for players who play both), there’s a much bigger swing in PP scoring than ES scoring.
For instance, Ovechkin and Hemsky have the following numbers over the past 4 years at 5v5 and 5v4 (all are per 60):
Hemsky
10-11 – 2.88 ES, 3.47 PP
09-10 – 2.85 ES, 2.85 PP
08-09 – 2.08 ES, 5.00 PP
07-08 – 2.36 ES, 5.93 PP
Ovechkin
10-11 – 2.59 ES, 3.48 PP
09-10 – 3.70 ES, 5.80 PP
08-09 – 2.86 ES, 6.09 PP
07-08 – 3.00 ES, 4.55 PP
PP scoring is much harder to find a pattern with, simply because it’s such a small portion of the TOI and a goal impacts it much more significantly.
Same thing on the PK. It’s not unusual to see a goaltender post a significantly higher or lower sv% on the PK, which skews the numbers significantly (think Garon in 08-09 and Khabibulin last year). 2 very opposite seasons but it just kills a lot of what you can pull from it.
In theory, there is little difference between practice and theory, but in practice there is!
Oh yeah, I completely understand that, smaller sample size leads to greater influence by luck and so we can tell less from the stats. I just find it interesting looking at the more “mystical” stats and trying to figure out what they’re telling us, or develop new metrics to accurately (as accurately as we can anyways) measure a players performance in odd-man situations.
Why are we assuming that 5 on 5 point percentage and offensive ability are that highly correlated? It would seem to have a lot of year to year variation and probably variation with zone start as well. I would imagine that the Preds getting rid of Franson is a sell high.
http://drivingplay.blogspot.com - The blog with three first lines
There is a lot of year to year variation (as with many statistics), but over the long haul (i.e. several years of data), I think we see the cream rise to the top. It definitely passes the smell test (at least for me), and makes sense philosophically as well (the guys who have the puck the most in the offensive zone tend to get the most points and thus have the most offensive ability).
As for the specific transaction between the Leafs and Preds, I agree that they “sold high”, but it’s a sale that doesn’t gets them much. Lebda is overpaid for 2011-12 and Franson is underpaid, so that lessens the financial gain from ditching Lombardi substantially. I understand that they don’t have the leeway to gamble much money, but I would think that a goodly portion of Lombardi’s salary is covered by insurance if he stays out of the lineup, and if he makes it back, there’s a decent chance that he’ll end up providing decent value.
The biggest fanana of the Havana Bananas.
by Scott Reynolds on Aug 29, 2011 9:33 PM MDT up reply actions
well that deal’s foolishness was compounded by the fact that the preds forgot to buy lebda out. remember when the red wings kept derek meech and brett lebda over kyle quincey?
as for point production, yeah, i agree, but you were speaking as if 1 year worth of data were particularly meaningful, when i’m not sure that it is.
http://drivingplay.blogspot.com - The blog with three first lines
You’re referring to Franson here, correct? I listed just the one season above, but he has played two NHL seasons and was very good offensively in both. He had 14 points on 35 five-on-five goals-for in 2009-10 (40.0%), which brings his two-year total to 37 points on 76 goals-for (48.7%). We also know that he scored 24 even strength points in his last season in the AHL, which makes for a third consecutive excellent offensive season. Over time, his IPP is very likely to come down some, but I think we have quite a bit of evidence that Franson is going to be a pretty good offensive defenseman at the NHL level.
The biggest fanana of the Havana Bananas.
by Scott Reynolds on Aug 30, 2011 10:30 AM MDT up reply actions
That Sharks top four is outstanding. They and the Pens are just loaded on D
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.

by 


























