So yesterday I started a pretty ambitious project: Determining the best defenseman in the NHL.
here's the link if you missed it:
PART ONE
The preamble was basically a discussion about the difficulty and potential ridiculousness of ranking players in the first place. Regardless, we carry on.
What I have done today is compile a list of the top players in several important categories. You will notice that none of these categories are blocked shots, +/- or hits. I feel these stats don't significantly matter enough to include in our decision. It doesn't mean they don't matter (expect +/- which doesn't) but that you couldn't realistically hit enough people or block enough shots to make you one of the best D in the NHL if you aren't already on the other lists.
At the end, I'll use my own feelings about the weight each category has to determine who I think - statistically speaking - are the five best defenseman in the NHL this year. Then I will compare it to my list from yesterday and see what happens.
Here goes:
Unless otherwise noted, all stats are for 5V5 only. The thinking is that everyone does alright on the PP and that 80% of the game is played at even strength. A look at PP scoring also shows no one is dominating to a crazy enough degree that we should adjust for it this year.
Goals
Karlsson 11
Byfuglien 11
Leddy 9
Krug 9
Burns 9
Klingberg 8
Weber 8
Points
Carlson 29
Burns 28
Josi 27
Subban 27
Karlsson 26
Doughty 25
Weber 25
Klingberg 23 (avg 12 less games played)
Points Per 60 Minutes
Klingberg 1.5
Hedman 1.5
Carlsson 1.5
Ellis 1.4
Burns 1.4
CF% (of players who’ve played 58 games +)
Muzzin 57.8
Keith 56.7
Doughty 56.6
Stralman 56.5
Boychuck 56.4
Leddy 56.3
Hamilton 56.1
Quality of Competition: I am going to ignore this on the basis that a) if you play against better players, you probably play with better players b) If you are in the conversation for best D, you most likely play 22 + minutes and in all situations C) sheltering players shouldn’t hurt your evaluations of them, since putting a player where he can succeed means there’s a way to utilize him most effectively. Goals scored by using your players to emphasize their skills still count the same. D) We’ll use zone starts instead, though Reason “C” makes me think we shouldn’t weight them too heavily.
Offensive Zone Starts Relative to Defensive Zone Starts
Best: Ekman-Larsson, Weber, Chara, Brodie, Carlson, Giordano
Worst: Rundblad, Krug, Suter, Yandle, Hedman ,S.Jones
Note: Amusing that you never hear people excuse Suter or Hedman's output, but Yandle's is always hedged against his sheltered minutes.
Power Play Points
Yandle 26
Strait 25
Shattenkirk 24
Karlsson 22
Letang 22
Defensive Coris Against Impact
Best: Giordano, OEL,, Jordie Benn, Brodie, Michalek, Demers, Ekholm, Gardiner
Worst: Daly, Wideman, Josi, Stanton, Weber, Harmonic, Orpik, Kronwell, Girardi, Phaneuf
Defensive Corsi Impact
Best: Hamilton, Giordano, OEL, Keith,Leddy, Jordie Benn, Demers, Michalek
Worst: Daly, Strait, Josi, Weber, Giardi, Kronwell, Suter, Pietrangelo
Ice Time Per Game
Doughty 29:26
Suter
Karlsson
Josi
Weber
Subban
Letang
Keith
Pietrangelo
Brodie 25:14
Total Points
Letang 51
Karlsson 50
Burns 49
Subban 49
Giordano 48
Luck Adjustments
Muzzin has a PDO of 97.2 and Hamilton has a PDO of 98.5
Doughty, Stralman and Boychuck all probably should have more goals b/c they have low Shooting % but, their higher PDOs suggest they are racking up assists bc their on-ice Shooting % is more normalized.
Yandle has the 11th lowest PDO among Defensemen and is the only one of them that is a top pairing player. His shooting percentage (5v5) is under 2% (i.e EXTREME bad luck) he also is almost 50% CF even though he’s played for the league’s second worst team. If he had spent the whole year with NYR and shot a league average S% he’d be near the top of the league in all categories. Again, a look into the stats shows Yandle is shockingly underrated.
Ryan Suter also has a ridiculous (1.2) shooting % and if luckier, would surely have more points. His CF% is high. His PDO is low. Thus, he should have more points. He also finishes near the bottom of a lot of key categories however, and statistically seems to be massively overrated.
Underated Players
Kindle
Ekholm
Hamilton
Ellis
Rundblad
Klingberg
Michalek
Jordie Benn
J.Faulk
Yandle
Gardiner
Demers (leads league in shorthanded CF% )
Conclusion
Now how exactly are you supposed to synthesize all that into deciding who is the best defensemen in the NHL? Even though we all now have the same information, I am willing to bet we all value different things to different degrees and as such very few of us will agree on a top five - and none of us will be "factually" proving anything, although I am willing to say that if you appear on more than a few of these lists, you are "factually"
one of the best.
What I did was I gave a weight to each category and whether at the top or the bottom, I gave a player the same weight/score just for appearing in the top five, but did give out points to lower ranked players.
I'm not going to get into my formula because it will side track the conversation, but I gave the highest weights to the defensive Corsi numbers because they are already complex formulas made in an attempt to answer this very question. I gave the lowest rates to PP points.
I also tried to account for usage, for luck, for playing on a bad team etc. I admit it isn't very scientific, but it's a lot more thought out and based on reality than the list I provided yesterday, which I feel I put together based a lot on just reputation. One example of this is Giordano, who has a CF% of 48.4 which is a far cry from Muzzin's 58%, but all things being equal, I feel confident that if you moved Gio to the Kings and played him under their system, that he would be A LOT closer.
Yesterday I posited this top five based on my perceptions, reputations, skills, how I figured they were doing so far etc. basically it's the normal way:
1. Karlsson
2. Doughty
3. Subban
4.Weber
5. keith/Letang
Based on my scores of all the statistics I compiled my top five now looks like this:
1. Mark Giordano
2. T.J Brodie
3. Oliver Ekman-Larsson
4. Drew Doughty
5. Jake Muzzin / Erik Karlsson
Honorable mention to Jason Demers, Jordie Benn and John Carlsson who are crazy undervalued throughout the league.
Obviously, this is going to ruffle some feathers, but Byfuglien and Subban and especially Suter do not appear to be anywhere close to as good as their reputations suggest.
All in all, this has been an eye-opening exercise for me, and I don't feel like the results are too surprising. I initially left Giordano off my list yesterday because of his injury, but with only 15 games to go (roughly) maybe he should still be getting consideration. He was the best while he played, after-all. If not Gio, the Norris should go to either Brodie or Doughty. I'd like to say OEL, but you can't give a best defensemen trophy to the league's worst team.
Giordano doesn't make a lot of the top five lists, but he's always hovering somewhere just outside when he doesn't. I tried to just look at who averages out to being the most effective. It's problematic, but it should be as there really cannot be a perfect system for ranking something so subjective.
Speaking of OEL, the Coyotes lost a heart-breaker (or what would have been one if I wasn't secretly cheering against them) last night to the Preds, dropping a 2-1 OT decision. Big props to my boy Louis Domingue for his 37 save performance against one of the NHL's top teams.
All Stats: WaronIce.com and NHL.com
Blogger • 
Join the Discussion:
»