Who Were The Best Shot Blockers In The NHL In 2011-12?

Derek Zona
May 13 2012 02:51PM

 Photo by Saruwine, via Wikimedia Commons , commons license.

If blocked shots are your thing, and going by the number of readers and analysts who use them to judge defensemen, there are a ton, you're probably used to hearing how essential shot-blocking is to sucessful NHL teams. You've also probably looked at shot block totals, or listened to an analyst discuss shot block totals and laud the players with the most blocks. The danger of using raw blocked shot totals as a measurement of effectiveness is that the players who see the most icetime and/or allow the most shot opportunities are natually going to block the most shots. 

Noted tactics writer Dawgbone has written about what happens when blockers get in the way, and the results haven't always been pretty. Sunny Mehta showed a small team skill in shot blocking and Desjardins showed an even smaller individual skill in the same. While shot blocking is a skill, or an art, for a very small segment of the NHL player population, talking heads espouse it as yet another magical part of the game, dictated by hard work and grit. In reality, a large quantity of blocked shots simply means the team, or player, is being dominated and forced to spend their time in their own end blocking rubber rather than possessing the puck and forcing the other team to block shots.

Earlier this season, I was in the midst of a discussion with the incomparable George Ays who turned me on to the idea of re-measuring shot blocking with context.  Ays re-created a formula used by Desjardins (we think) to determine which players were blocking the most shots, and which players were giving up a bunch of shots and blocking some.

In the tables below, I've listed the ESBS Ratio (% of even strength shot attempts blocked by an individual player) of all NHL players, as I'm calling it until someone comes up with a better name. One caveat to these numbers - I'm using total even strength blocks, not away even strength blocks, which may leave a heavy tinge of scorer bias, especially for any Rangers.  Since they block the preponderance of shots, let's start with the defensemen first:

Greg Zanon BOS 0.1455
Brett Clark T.B 0.1405
Keaton Ellerby FLA 0.1351
Josh Gorges MTL 0.1321
Anton Volchenkov N.J 0.1291
Steve Staois NYI 0.1286
Nate Prosser MIN 0.1272
Andy Greene N.J 0.1246
Colin White S.J 0.1234
Mark Stuart WPG 0.1228
Marc-Andre Bourdon PHI 0.1224
Roman Hamrlik WSH 0.1218
Ron Hainsey WPG 0.1216
Zbynek Michalek PIT 0.1189
Mike Komisarek TOR 0.1185
Ladislav Smid EDM 0.1184
Niklas Kronwall DET 0.1182
Niklas Hjalmarsson CHI 0.1167
Douglas Murray S.J 0.1161
Stu Bickel NYR 0.1159
Andy Sutton EDM 0.1150
Bryan Allen CAR 0.1148
Mark Fistric DAL 0.1138
Roman Josi NSH 0.1123
Andrew MacDonald NYI 0.1112
Kris Russell STL 0.1108
Raphael Diaz MTL 0.1108
Kevin Klein NSH 0.1101

Names like Schenn and Orpik drop off the list, while the very underrated Roman Josi and Andy Greene appear.  Does this alter the way we should view blocked shots?  Should it?

Though they block far fewer shots, it's interesting to see the forwards list:

David Steckel TOR 0.0708
Darroll Powe MIN 0.0707
Nick Bonino ANA 0.0678
Tim Connolly TOR 0.0659
Mike Fisher NSH 0.0638
Logan Couture S.J 0.0638
Andrew Miller DET 0.0631
Michal Handzus S.J 0.0626
Patrice Bergeron BOS 0.0622
R.J. Umberger CBJ 0.0612
Daymond Langkow PHX 0.0611
Matt Hendricks WSH 0.0597
Marc-Andre Gragnani VAN 0.0595
Ryan Callahan NYR 0.0595
Patrick Kaleta BUF 0.0591
Ruslan Fedotenko NYR 0.0574
Ryan Getzlaf ANA 0.0574
Vernon Fiddler DAL 0.0567
Matt Duchene COL 0.0565
Brooks Laich WSH 0.0561
Boyd Gordon PHX 0.0560
Brandon Sutter CAR 0.0560
David Backes STL 0.0555
Erik Condra OTT 0.0552
Brian Boyle NYR 0.0551
Brandon Prust NYR 0.0551

I expect to see defense-first and defense-only forwards like David Steckel, but Daymond Langkow, Logan Couture and Tim Connolly are a bit of a shock.

Which names stick out?  Is Brandon Sutter, Carolina's mule and should-be Selke finalist, a surprise?  Does his presence on the list alongside Patrice Bergeron change his place on the Selke list?  What about Ryan Getzlaf, a power play specialist most known for passing to Corey Perry? 

E976572df3ea71edf6bafb00adc1c546
Avatar
#1 Corey S.
May 13 2012, 02:59PM
Trash it!
0
trashes
+1
0
props

Good stuff. I've been monitoring this with the Hurricanes players so I'm not too surprised to see Allen and Sutter on this list. Although, I thought Dwyer would be on here too since he led forwards in ESBS% the last time I checked. Things may have changed since then, though.

Something that I've been meaning to look at is the percentage of shots that a player blocks on the PK.

Avatar
#3 Derek
May 13 2012, 03:27PM
Trash it!
0
trashes
+1
0
props

Great stuff - Josi/Klein are a Corsi horror show as a pairing but I guess this shows that they aren't totally useless.

As for Corey's comment regarding PK blocks, I can't remember who it was (might have been Vic or George Ays...all I remember is that it was in the comments section on Gabe's site years back) who showed that there's actually a negative correlation between blocked shots on the PK and scoring chances against the PK. That seems to jive with JLikens' findings that mere SA/60 on the PK is a better prediction of future GA/60 than Fenwick Against or Corsi Against per 60. So it's a possibility using BS% to evaluate shorthanded shot-blocking undersells some guys.

Avatar
#4 Eric T.
May 13 2012, 07:19PM
Trash it!
0
trashes
+1
0
props

It won't be too hard to pull the road numbers, but for a simpler first pass at correcting for home scorer bias, I just applied a correction factor to each player's numbers of:

(team-road-BS * 2) / ((team-road-BS + team-home-BS) * 1.0164)

1.0164 is the league-average difference between home blocks and road blocks, so this should be a reasonable approximation of the number of blocks a player would have if he played all of his games with a league-average scorer.

This moves Devils players way up the list (NJ had 322 home blocks and 622 road blocks, giving them a 1.285 correction factor). Volchenkov and Greene lead the new list at .1659 and .1601 respectively, and I'm left wondering what the numbers for other NJ players would be (I just used the ones you listed, not pulling data myself).

The rest of the list doesn't change much, as other teams' factors are all between .9 and 1.1, but the Devils seem to get uniquely hosed on the blocked shot counting.

Avatar
#5 Jay
May 14 2012, 06:53AM
Trash it!
0
trashes
+1
0
props

What I find interesting with blocked shots is that Don Cherry loves them and his philosophy is that if you're team is not blocking shots then you will lose. But it is often the losing teams that block more shots because of scoring effects. I often see this on CanucksArmy.com

I took a look near the end of the regular season of teams points vs blocked shots and came up with this graph: http://i.imgur.com/wYsMo.png

There's not a strong correlation there. But what is interesting is that NY Rangers are one of the highest blocking shots in the league, and they were at the top of the regular season standings and still the top team in the playoffs. But that is a lot of punishment on the players. Makes me wonder if they can sustain it all the way to the end. They do rely on Lundquivst a lot.

Avatar
#6 John
May 14 2012, 10:10AM
Trash it!
0
trashes
+1
0
props

Great stuff.

Do have a feeling stuff like this would be more informative if the sample used was more than one year, as it just seems like there are soo many variables that go into a number like this. As a ranger fan i sometimes see with a real good shot blocker like callahan even when a D-man has time and space he will opt against shooting on him cause he just doesn't think he can get it through, instead possibly passing D to D and trying getting it through the other side. Obviously this never happens with someone like Gaborik.

May be over thinking this, but have a feeling a bigger sample size could be beneficial here.

Avatar
#7 Tach
May 14 2012, 11:13AM
Trash it!
0
trashes
+1
0
props

If we were trying to determine whether a guy was adding value by shot blocking it would be interesting to put these numbers in a table along with Corsi rates (and QualComp and ZoneStart for context maybe). I think we would also need a league average or replacement number. For argument's sake for D-men I will call replacement level .08.

For example, (just to pull a guy at random) Roman Hamrlik appears to have been (I am just using behindthenet.ca GF/GA, SF/SA and BF/BA numbers but I can't make them add up to the overall rate listed at +0.26/60, so if someone can let me know what I am missing - cheers) was 57.55 corsi for/60, 55.82 corsi against/60. At his block rate (.1218) he prevented 6.80 of those attempted shots from getting to the net.

At a replacement rate of .08 he would have only prevented 4.47 of those shot attempts getting through, or his net shot attempt suppression was 2.33 Corsi/60.

As attempted shots convert to goals at a rate of 4.5% (as per this article and comment chain http://flamesnation.ca/2012/4/30/a-look-back-at-even-strength-play), Hamrlik's shot blocking saved .10 Goals/60 minutes of EV icetime. At ~17 EV minutes per game, he prevented about 1 goal every 4 games. Sounds about right to me in terms of shot blockings contribution to winning games.

Avatar
#8 RhysJ
May 15 2012, 02:13PM
Trash it!
0
trashes
+1
0
props

First off, I propose calling it "Block Rate." Easier to look at and more straight forward than "ESBS Ratio."

Also, you have MA Gragnani listed as a forward when he's really just a sparsely used d-man.

A criticism I have of this is that it's tough to draw any sort of conclusions from this analysis without knowing what the league mean and standard deviation is for this measure. Since there are no mean and stdev numbers, it's simply impossible to tell if the difference between let's say Hendricks and Kaleta is statistically significant or just attributable to random chance.

Comments are closed for this article.