Andrew Berkshire: Ranking the Top 20 Defencemen in the NHL, by the Numbers

[fiveonfive]

But that's precisely the point. If you know anything about him, it's that he's a numbers first, 'what's hockey' second.

He's a context-less **** who needs public shaming at very opportunity.

Any little acquiescence of his fact is fuel for his ******** machine...

I agree with you on principles, but listen to him in the various podcast he's in. This is not a man who deal in anything but spreadsheets...

And he deserves scorn for it.

This seams like it's more about you and your hate for numbers, than it is about the article.

 

[KaseMeOutside]

vatanen 15 lindholm 10
good list imo

 

[KrisLetAngry]

I understand what he is trying to do. However he will have to fine tune the numbers to make a more appropriate list. I think this is the worse list out of them aLl

Like how does Karlsson not have the best offensive score

 

[Micklebot]

Personally I've always found that defencemen are far more difficult to quantify compared to forwards

and yet you consistently reduce it down to goals against as though that tells the whole story.

While I agree the effort by Birkshire came up with some dubious results, I admire the effort. He attempted to come to the conclusion based on the process not the end result. The process (the various types of defensive plays recorded such as loose puck recoveries, defensive plays: hits, stickchecks, pass blocks, and shot blocks) is more of a measurement of the individual than the end result (GA).

There are some clear problems though, with his methodology.

 

[Rorschach]

did vlasic die?

Highly suspect list, and this is coming from somebody who loves advanced stats. This is overdoing it.

rip...

 

[fiveonfive]

A lot of the stats seem to be done by occurrences per 20 mins. So defensive stats can only happen when the opposition has the puck, so obviously players on the best possession team in the league will have their defensive play occurrences artificially lower than teams always defending.

This makes it make far more sense why Giordano, who's team hasn't got the puck much, and gets defensive deployment, is ranked high while Doughty who's team is more often on attack is ranked lower; you can't register loose puck recoveries, defensive plays (hits, stickchecks, pass blocks, and shot blocks) when you have the puck.

Yup, that was exactly my point. That’s why I think it’s a bit absurd that people still want to say “This guy’s defensive game is so incredible, he adds so much defensive value to his teamâ€, when that player has to play a materially lesser amount of defending in his own zone or fighting for pucks at all. It’s a contradiction.

 

[Machinehead]

For those wondering about Weber:

Red flag right here. Weber plays with Josi who... Well, black hole of shot suppression would be an understatement.

Don't think Berkshire is weighing quality of teammates enough.

That's the problem with this study; he's weighing a variety of factors according to his own subjectivity.

 

[Kranix]

The flaw in his methodology is that it doesn't completely flatter my player.

 

[Pinto Bean]

Yea some of these rankings are terrible but one thing stands, Erik Karlsson passes both the eye test and the stats test!

 

[MXD]

Andrew Jerkshire strikes again.

 

[Sens Rule]

I understand what he is trying to do. However he will have to fine tune the numbers to make a more appropriate list. I think this is the worse list out of them aLl

Like how does Karlsson not have the best offensive score

Simple. All his offensive stats are per 60 minutes. Karlsson averaged about 28 minutes per game the last 3 years. Even though he outscored all D in each year, per 60 minutes he didn't. Klingberg for instace has basically one year with limited ice time per game and is used in offensive situations far more often then not. Karlsson is used in all situations.

That is a huge flaw in his ratings. Offensively Karlsson is clearly the best over 3 seasons. Burns might be statistically if you solely looked at last year. But over 3 years it has to be no contest.

His ratings on D do not give higher rankings to D that get the most icetime, comparing 25-28 minute guys equal to 20-22 minute guys. I think icetime is a huge factor in how you should look at D-Men.

Doughty is ranked lower then he should be because he plays 27 mins a game for 3 years and so many of his ratings are based on /60 minutes.

 

[Deaner]

Red flag right here. Weber plays with Josi who... Well, black hole of shot suppression would be an understatement.

Don't think Berkshire is weighing quality of teammates enough.

That's the problem with this study; he's weighing a variety of factors according to his own objectivity.

I agree, and the D list is where that will have the greatest impact. I think you mean subjectivity though

 

[KrisLetAngry]

Simple. All his offensive stats are per 60 minutes. Karlsson averaged about 28 minutes per game the last 3 years. Even though he outscored all D in each year, per 60 minutes he didn't. Klingberg for instace has basically one year with limited ice time per game and is used in offensive situations far more often then not. Karlsson is used in all situations.

That is a huge flaw in his ratings. Offensively Karlsson is clearly the best over 3 seasons. Burns might be statistically if you solely looked at last year. But over 3 years it has to be no contest.

His ratings on D do not give higher rankings to D that get the most icetime, comparing 25-28 minute guys equal to 20-22 minute guys. I think icetime is a huge factor in how you should look at D-Men.

Doughty is ranked lower then he should be because he plays 27 mins a game for 3 years and so many of his ratings are based on /60 minutes.

Okay. I did know why. But why!!!!

As in when you make this list up and the best defense offensively isn't 1st wtf!!! If I make a list and I see thee results I change criteria

 

[Michael Farkas]

Red flag right here. Weber plays with Josi who... Well, black hole of shot suppression would be an understatement.

Don't think Berkshire is weighing quality of teammates enough.

That's the problem with this study; he's weighing a variety of factors according to his own subjectivity.

Aren't you discounting it to whatever degree because it doesn't agree - subjectively - with the things that you find important though...? As if the term "shot suppression" isn't subjective and biased in its own right...just because it's in numerical form doesn't mean it's untouchable. It's just that you choose to believe certain numbers versus others for various reasons, many of which may be perfectly legitimate. I think you're just about the loudest and most "rigid" in terms of these shot counting tactics for player evaluation, so surely you recognize how limited those figures are at this point, right?

 

[Erik Alfredsson]

and yet you consistently reduce it down to goals against as though that tells the whole story.

While I agree the effort by Birkshire came up with some dubious results, I admire the effort. He attempted to come to the conclusion based on the process not the end result. The process (the various types of defensive plays recorded such as loose puck recoveries, defensive plays: hits, stickchecks, pass blocks, and shot blocks) is more of a measurement of the individual than the end result (GA).

There are some clear problems though, with his methodology.

Once again Mickle says all that needs to be said.

 

[HarrisonFord]

and yet you consistently reduce it down to goals against as though that tells the whole story.

While I agree the effort by Birkshire came up with some dubious results, I admire the effort. He attempted to come to the conclusion based on the process not the end result. The process (the various types of defensive plays recorded such as loose puck recoveries, defensive plays: hits, stickchecks, pass blocks, and shot blocks) is more of a measurement of the individual than the end result (GA).

There are some clear problems though, with his methodology.

I don't think that, I just don't think it's something to be ignored

 

[fiveonfive]

I don't think that, I just don't think it's something to be ignored

Did you read the article? It does consider GA/60rel as part of the defensive score. It’s not ignored. In addition, most of the actions that sustain that number for a guy like Doughty (like his absurdly high CF %) are reflected in his incredible Transition Game score as well. What exactly is your issue?

 

[Kairi Zaide]

Simple. All his offensive stats are per 60 minutes. Karlsson averaged about 28 minutes per game the last 3 years. Even though he outscored all D in each year, per 60 minutes he didn't. Klingberg for instace has basically one year with limited ice time per game and is used in offensive situations far more often then not. Karlsson is used in all situations.

That is a huge flaw in his ratings. Offensively Karlsson is clearly the best over 3 seasons. Burns might be statistically if you solely looked at last year. But over 3 years it has to be no contest.

His ratings on D do not give higher rankings to D that get the most icetime, comparing 25-28 minute guys equal to 20-22 minute guys. I think icetime is a huge factor in how you should look at D-Men.

Doughty is ranked lower then he should be because he plays 27 mins a game for 3 years and so many of his ratings are based on /60 minutes.

Your post makes me realize something that I think could be very important.

Everywhere, you see people using /60 stats. But, has anyone actually proved that most of the statistics are linear in terms of minutes played?

Because quite frankly, my theory would be that most of them are not and start to "plateau" the more a player plays per game. For instance, let's use Player A's pts/60 of 1.5, ie that player put up a point on average every 40 minutes. This assumes that if the player played 10 minutes per game, he would put up 20.5 points, then 41 points for 20 minutes per game, and 61.5 points for 30 minutes per game.

While it mathematically makes sense, I just don't think it realistically makes sense. Would be nice if someone found a way to confirm (or not) whether the use of /60 stats is justified, and if a non-linear model for stats adjustment in matter of time played is more adequate than a linear model.

 

[Rorschach]

If you don't factor teammates, strength of opposition or you choose to weigh stats wrong (in other words you misunderstand how hockey is played), then you get this sort of thing where the results completely fail the eye test.

 

[triggrman]

Red flag right here. Weber plays with Josi who... Well, black hole of shot suppression would be an understatement.

Don't think Berkshire is weighing quality of teammates enough.

That's the problem with this study; he's weighing a variety of factors according to his own subjectivity.

Weber is also hurt because his 5:5 offensive numbers are 4th on the team for defensemen last season, while Josi's on the otherhand were very good (5th in the league), Weber also stuggles at transitioning the puck, which again, Josi excels at.

 

[fiveonfive]

If you don't factor teammates, strength of opposition or you choose to weigh stats wrong (in other words you misunderstand how hockey is played), then you get this sort of thing where the results completely fail the eye test.

Teammates and strength of opposition are factored in though. Also, strength of opposition among top pairing dmen is usually extremely similar.

What do you think is weighted wrongly in this article? I personally think the ranking isn't that far off from the eye test at all, definitely not as far as so many people keep stating they are.

 

[ManyIdeas]

We made that scrub Pietrangelo captain instead of Shattenkirk?

What a huge mistake!

 

[Machinehead]

I agree, and the D list is where that will have the greatest impact. I think you mean subjectivity though

Yes. Brainfart

Aren't you discounting it to whatever degree because it doesn't agree - subjectively - with the things that you find important though...? As if the term "shot suppression" isn't subjective and biased in its own right...just because it's in numerical form doesn't mean it's untouchable. It's just that you choose to believe certain numbers versus others for various reasons, many of which may be perfectly legitimate. I think you're just about the loudest and most "rigid" in terms of these shot counting tactics for player evaluation, so surely you recognize how limited those figures are at this point, right?

It's all limited though. If we had a definitive stat for player quality there would be nothing to discuss and HF would be boring.

That's why I find it hard to talk baseball anymore even though I love baseball.

"What's his WAR? Oh ok. Guess we're done here."

 

[perronist]

I'm so mad abt this list w. it's 'statistics' and 'empirical evidence' fyad Andrew derpshire give me my eye test anyday

 

[JohnLennon]

Are Giordano and Brian Campbell really the two best in the league in the defense category?