NHL Mid-Season Statistics Leaders, Volume I
by Hawerchuk on Feb 15, 2010 9:00 AM EST
in Statistics
Email
PrintI thought I'd take a look at the statistical leaders in some very slightly advanced metrics. First, for some perspective, the league leaders in Points per 60 Minutes of 5v5 Ice Time over the last three seasons:
| 2009-10 | 5v5 P/60 | 2008-09 | 5v5 P/60 | 2007-08 | 5v5 P/60 |
| Ovechkin | 4.12 | R. Bourque | 3.20 | Crosby | 3.38 |
| H. Sedin | 3.79 | Semin | 3.16 | Malkin | 3.20 |
| D. Sedin | 3.77 | Malkin | 3.07 | Stastny | 3.12 |
| Crosby | 3.35 | Crosby | 3.00 | Alfredsson | 3.12 |
| Semin | 3.33 | Savard | 2.99 | Ovechkin | 3.00 |
| Backstrom | 3.01 | D. Sedin | 2.97 | Heatley | 2.95 |
| Kovalchuk | 2.89 | Krejci | 2.95 | Spezza | 2.93 |
| Gaborik | 2.88 | Parise | 2.93 | Dumont | 2.88 |
| Knuble | 2.85 | Datsyuk | 2.91 | Radulov | 2.86 |
| Erkisson | 2.84 | Havlat | 2.89 | Roy | 2.86 |
Raise your hand if you knew Rene Bourque was one of the most efficient players in the NHL last season. Here are the 11th-40th players in that category this seasons:
| Player | P/60 | Player | P/60 | Player | P/60 |
| Marleau | 2.83 | Parise | 2.62 | Ponikoraovsky | 2.46 |
| Thornton | 2.82 | Horton | 2.61 | Wolski | 2.46 |
| Kane | 2.76 | Antropov | 2.61 | Sharp | 2.42 |
| Malkin | 2.75 | Lecavalier | 2.56 | Upshall | 2.42 |
| B. Richards | 2.73 | Neal | 2.52 | Heatley | 2.40 |
| C. Stewart | 2.70 | Fleischmann | 2.51 | Parse | 2.39 |
| Fehr | 2.70 | Stamkos | 2.51 | R. Bourque | 2.38 |
| Burrows | 2.69 | Simmonds | 2.48 | Asham | 2.36 |
| Getzlaf | 2.63 | Van Riemsdyk | 2.47 | Stastny | 2.35 |
| St. Louis | 2.62 | E. Staal | 2.46 | J. Jokinen | 2.34 |
A lot of good lines, good offensive players, and a few guys, like Scott Parse, who you've probably never heard of. And, of course, Rene Bourque. What about On-Ice/Off-Ice +/-, which I've called a player's "Rating":
| NAME | RATING | NAME | RATING | NAME | RATING |
| Ovechkin | 2.83 | H. Sedin | 1.86 | Cammalleri | 1.52 |
| Parise | 2.80 | C. Franson | 1.85 | Holmstrom | 1.5 |
| D. Sedin | 2.55 | J. Jokinen | 1.81 | Knuble | 1.49 |
| Fistric | 2.38 | Lidstrom | 1.79 | E. Staal | 1.49 |
| Downie | 2.34 | Bailey | 1.74 | Backstrom | 1.47 |
| Zajac | 2.24 | Brookbank | 1.65 | Smyth | 1.46 |
| Antropov | 2.16 | Schultz | 1.64 | Rafalski | 1.46 |
| Foligno | 2.04 | Smid | 1.61 | Simmonds | 1.45 |
| Gaborik | 1.99 | Vrbata | 1.59 | Prust | 1.44 |
| Penner | 1.87 | Wolski | 1.52 | Hornqvist | 1.41 |
It's a real mixed bag, mostly because players draw widely-varying matchups and don't line up against similar opponents. In that vein, the league leaders in Quality of Competition:
| NAME | QoC | NAME | QoC | NAME | QoC |
| Ward | 1.931 | Jovanovski | 1.655 | Heatley | 1.328 |
| Volchenkov | 1.891 | Michalek | 1.596 | Pandolfo | 1.323 |
| Seabrook | 1.831 | Smithson | 1.584 | Zanon | 1.321 |
| Jackman | 1.779 | Legwand | 1.565 | Thornton | 1.295 |
| Phillips | 1.777 | Grossman | 1.540 | Marleau | 1.288 |
| Alfredsson | 1.755 | Robidas | 1.436 | Lidstrom | 1.283 |
| Polak | 1.740 | Gleason | 1.392 | Girardi | 1.269 |
| Keith | 1.717 | Leopold | 1.360 | Thompson | 1.255 |
| Mitchell | 1.711 | Zetterberg | 1.348 | Sutton | 1.241 |
| M. Staal | 1.665 | Brown | 1.343 | Ohlund | 1.233 |
The Quality of Competition metric here is "Relative Corsi" - in other words the average opponent's Corsi number relative to his team. Quality of Competition is an imperfect metric, but using opponent shot totals reduces the role of individual scoring luck, or a tougher schedule of opponents.
There are a lot of lines and defensive pairs on this list with very few singletons - Carolina's Tim Gleason is probably the only player on this list who doesn't have a teammate within striking distance. The Quality of Competition leaders always pose an interesting question for me: are these players at the top of this list because their coaches send them out against the other team's best players? Or because opposing coaches send out their best players against them.
Next time, a run through the leaders a few more obscure statistics...
12 comments
|
0 recs |
Do you like this story?
Comments
Also for QoC...
Or, for the defensemen, could it be because they log more minutes than forwards, with the exception of the best forwards on the opposing team? We know that coaches like to match up their players to the other team’s, but with the new icing rule there’s some overlap.
Bettman's Nightmare: A Blog Where Hockey Aficionados Dismantle That Mighty Empire, One Balsillie at a Time
http://bettmansnightmare.blogspot.com/
by Bettman's Nightmare on Feb 15, 2010 10:06 AM EST reply actions
I just realized
the above comment made little sense. That was a pre-coffee comment. What I meant to say was that a potential reason that defensemen tend to “dominate” QoC could be “because they log more minutes than forwards…etc., etc. (see above comment)”.
Bettman's Nightmare: A Blog Where Hockey Aficionados Dismantle That Mighty Empire, One Balsillie at a Time
http://bettmansnightmare.blogspot.com/
by Bettman's Nightmare on Feb 15, 2010 12:25 PM EST up reply actions
I've always been curious
…why you do value out of 60 minutes instead of something closer to what an average defensemen or forward plays per game. I understand that average TOI has fluctuated over the years, but you could always adjust for those shifts, too.
Bettman's Nightmare: A Blog Where Hockey Aficionados Dismantle That Mighty Empire, One Balsillie at a Time
http://bettmansnightmare.blogspot.com/
by Bettman's Nightmare on Feb 15, 2010 10:09 AM EST reply actions
I think it makes sense, since we think of things in games (60 mins), right? Like the 5v5 pts/60 says basically that for a games’ worth of ES time for Ovechkin, he gets a little over 4 points. It’s easier to compare them relative to the team then, I guess. No dividing by three needed.
Ovechkin = Green Backs
by red army line on Feb 15, 2010 11:18 AM EST up reply actions
To me, everything makes sense on a per-60 basis. We know how many goals teams score per 60 minutes (2-3). But Ovechkin’s line scores 4-5. So I know intuitively how much better he is than average.
Do you think it's too hypothetical?
Meaning, points per 60 minutes (of 5v5) presents what Ovechkin’s line would do if they played 60 minutes of even strength and fatigued to the same levels of energy they do at 20-24 minutes per game. I completely understand the reasoning of using 5v5, and I like the notion of “if Ovechkin’s line was out there all game” but I wonder if its too much credit and hypothesizing.
Bettman's Nightmare: A Blog Where Hockey Aficionados Dismantle That Mighty Empire, One Balsillie at a Time
http://bettmansnightmare.blogspot.com/
by Bettman's Nightmare on Feb 15, 2010 12:30 PM EST up reply actions
I don’t think so. The point is to judge a player’s effectiveness against other players’ and teams’ effectiveness over a standard time frame. Teams play 60 min games, generally, so that makes 60mins an easy standard of time for comparison. We’re not saying that Ovechkin will score 4 points if he plays 60 mins in a game, but he scores 4 points every 60 mins he plays (just under 3 games’ worth).
Ovechkin = Green Backs
by red army line on Feb 15, 2010 3:15 PM EST up reply actions
gabe,
while it would be mostly for fun/approximation, you should take a kick at coming up with some sort of corsi rating that takes relative corsi and adjusts for faceoffs and RelCorsiQoC.
I have put it together, but it is an enormous piece of shit that makes me unhappy when I look at it. Seriously, Bobby Holik is the #1 guy over the last three seasons.
I have to agree that anything that makes Bobby Holik look that good is bad.
Bettman's Nightmare: A Blog Where Hockey Aficionados Dismantle That Mighty Empire, One Balsillie at a Time
http://bettmansnightmare.blogspot.com/
by Bettman's Nightmare on Feb 16, 2010 9:12 AM EST up reply actions
re: Holik Problem
This is one flaw in using start of faceoffs as a proxy variable for shift location. In Atlanta Holik was sent over the boards to take many faceoffs but often would immediately exit the ice and allow another Center to step out there. Ergo, his Start of Shift numbers are significantly biased and likely over state the extent to which Holik spent time in his own defensive end. If only we had the old “Zone Time” stat for each player. Shift Starts and Shift Ends are only an approximation and they may not capture what we really want under some circumstances.
All things Thrashers + stats: www.birdwatchersanonymous.com
by The Falconer on Feb 19, 2010 12:08 PM EST up reply actions
Comments For This Post Are Closed
Secondary Sidebar
User Tools
Related FanPosts on Arctic Ice Hockey
FanPosts
Community blog posts and discussion.
