A look back at Sidney Crosby's defense over the past decade

[nowhereman]

I imagine your bedroom like something out of a Beautiful Mind, adorned with hundreds of Sidney Crosby posters with cigarette burns where the eyes used to be and statistical clippings charting all his secondary assists.

 

[Three On Zero]

This might actually mean something if hockey was a 1 vs 1 sport

 

[Kairi Zaide]

I imagine your bedroom like something out of a Beautiful Mind, adorned with hundreds of Sidney Crosby posters with cigarette burns where the eyes used to be and statistical clippings charting all his secondary assists.

which is even funnier considering Crosby has higher primary points/60 than OV career wise

 

[JackSlater]

Somehow, some way, I knew who the OP was without even looking. Now I also know that Crosby isn't as good defensively as Bergeron is - who knew!

 

[Empoleon8771]

Word of advice: if you think that looking at goals against alone is an accurate representation of defensive ability, you really shouldn't be making threads evaluating how good players are defensively.

I'm not even going to touch on powerplay goals against, because that's asinine to a whole different level.

 

[Hockey Outsider]

I considered that but having looked at on-ice goal tending save percentages on Natural Stat trick for Sid and Ovie and Bergeron, nothing really jumped out. Bergeron has the best on-ice save percentage of the three spanning two main goalies. IMO the data suggests that Thomas and Rask had excellent save percentages because of Bergeron (and other skaters), not despite them.

Also, Marc-Andre Fleury is currently being debated as a hall of famer:

Is Marc-Andre Fleury a Hall-of-Famer?

Playing in front of MAF is not some significant disadvantage relative to the other players in the OP. MAF's save percentages over the years are generally average to above average. Yes, his stats look bad over a very small sample of playoff games over four seasons (one of which Crosby didn't even play), but if you recall, the Pens playoff style from the time - particularly the playoff series against the Flyers in 2012 which was completely out of control - there wasn't much defense being played at either end of the ice by the skaters, and definitely not by Sid (who somehow managed to score 5 even strength points and still wind up at plus minus negative 3 for the 6 game series). Sid's offensive numbers greatly benefitted from the firewagon style that his team deployed at times, this is merely the other side of that coin.

I'm skeptical of trying to evaluate goal tending based on save percentage and concluding the pattern is unrelated to individual skaters even in a very large sample size (such as what is in the OP).

It strikes me as obvious that the quality of the goaltending potentially affects the number of goals against a player is on the ice for. Attributing the differences in a goalie's save percentage to a single forward (who's on the ice for maybe 22 minutes a game, along with four other teammates) simply doesn't make sense. For example, Patrik Elias was on the ice for a minuscule number of goals against relative to his ice time. Yes, he was a great two-way forward, but clearly some of that was due to him playing in front of Martin Brodeur (and generally getting a lot of ice time with Scott Stevens and/or Niedermayer).

All that being said - I'm not saying that Crosby is at a disadvantage relative to any of the other players listed above. Just that it's a potentially important factor that hasn't been considered. Maybe he played in a more favourable environment than some of these players?

I realize some people are debating it, but I think it's laughable that Fleury is even being talked about as a potential Hall of Famer. He's spent his entire career stopping the puck at essentially the league average rate, and was never a Vezina trophy finalist in his long, healthy career.

Going back to the main point of the topic - it would be interesting to combine the goals against with goals for data. For example, the data shows that Bergeron and Datsyuk are pretty clearly better defensively than Crosby and Ovechkin (which should surprise nobody). But are Crosby and Ovechkin more effective players at even-strength, taking into account their superior offense? I might work on this in the new year, time permitting.

 

[Tad Mikowsky]

The biggest Crosby hater makes a post to discredit Crosby.

This is as bad as the guy who props Crosby’s grace period.

 

[Gurglesons]

Glad someone finally ran these numbers as an unbiased party. Been waiting for this type of in depth analysis since we proved goals are worth 2 st dev more than an assist and Ovechkin is the king of all.

 

[Jacob]

I imagine your bedroom like something out of a Beautiful Mind, adorned with hundreds of Sidney Crosby posters with cigarette burns where the eyes used to be and statistical clippings charting all his secondary assists.

Ray Finkle’s bedroom in Ace Ventura.

 

[Kairi Zaide]

It strikes me as obvious that the quality of the goaltending potentially affects the number of goals against a player is on the ice for. Attributing the differences in a goalie's save percentage to a single forward (who's on the ice for maybe 22 minutes a game, along with four other teammates) simply doesn't make sense. For example, Patrik Elias was on the ice for a minuscule number of goals against relative to his ice time. Yes, he was a great two-way forward, but clearly some of that was due to him playing in front of Martin Brodeur (and generally getting a lot of ice time with Scott Stevens and/or Niedermayer).

All that being said - I'm not saying that Crosby is at a disadvantage relative to any of the other players listed above. Just that it's a potentially important factor that hasn't been considered. Maybe he played in a more favourable environment than some of these players?

I realize some people are debating it, but I think it's laughable that Fleury is even being talked about as a potential Hall of Famer. He's spent his entire career stopping the puck at essentially the league average rate, and was never a Vezina trophy finalist in his long, healthy career.

Going back to the main point of the topic - it would be interesting to combine the goals against with goals for data. For example, the data shows that Bergeron and Datsyuk are pretty clearly better defensively than Crosby and Ovechkin (which should surprise nobody). But are Crosby and Ovechkin more effective players at even-strength, taking into account their superior offense? I might work on this in the new year, time permitting.

A lot of the work has already been done.

5v5

All situations (warning, as this inherently puts players who play significant time on the PK at a disadvantage. Also, RelTM stats are not available on E-H combining all situations)

You can play with the data here.

 

[wetcoast]

There is another 5 minutes wasted.

 

[Hockey Outsider]

A lot of the work has already been done.

Great, thanks for sharing.

The other thing that would be interesting to see (and maybe this is in your links, I haven't dug into them yet) - who has the biggest difference between the ratio when they're on vs off the ice? If, for example, Bergeron is on the ice for 1.2 ES GF for every 1 ES GA (a ratio of 1.20), but the Bruins as a whole have a ratio of 1.15, that's not necessarily more impressive than, say, Jamie Benn having a ratio of 1.00 if the Stars are 0.80 without him. (To be clear, those numbers are completely fabricated to illustrate the point - but I think this is the best way of figuring out who's the top ES performer over the past decade).

 

[Midnight Judges]

I'm not even going to touch on powerplay goals against, because that's asinine to a whole different level.

Is preventing power play goals against somehow not an objective - in your opinion?

 

[Midnight Judges]

Word of advice: if you think that looking at goals against alone is an accurate representation of defensive ability...

Do you think it is pure coincidence that Bergeron, Kopitar, and Datsyuk do well by all of these metrics?

 

[Midnight Judges]

It strikes me as obvious that the quality of the goaltending potentially affects the number of goals against a player is on the ice for. Attributing the differences in a goalie's save percentage to a single forward (who's on the ice for maybe 22 minutes a game, along with four other teammates) simply doesn't make sense.

I agree. When I typed the sentence I was still referring to a forward's on-ice SV %. Sure, 1 player is going to have limited control over that. I would predict that if they are a legitimate Selke caliber player, that control is going to be statistically significant one way or another, especially over a very large sample size of games such as the 10 year's worth of data in the OP.

 

[Beauner]

Do you think it is pure coincidence that Bergeron, Kopitar, and Datsyuk do well by all of these metrics?

do YOU think it's pure coincidence that 95% of posts in threads you create are snide or sarcastic remarks/comments with barely any credence given to ANY of your claims?

 

[GreatGonzo]

I’d be more curious to see how Crosby matches up defensively against guys like Backstrom, Giroux, Marchand, and even Barkov. I do believe Crosby’s defensive game is greatly over exaggerated and overrated in general, but Using the best players in terms of defensive play to compare only proves what Crosby lacks unfairly.

 

[3074326]

Imagine being obsessed with someone you hate. The misery of OP makes sense when you think about it.

I feel bad for you, Midnight Judges. You're a sad case.

 

[authentic]

It strikes me as obvious that the quality of the goaltending potentially affects the number of goals against a player is on the ice for. Attributing the differences in a goalie's save percentage to a single forward (who's on the ice for maybe 22 minutes a game, along with four other teammates) simply doesn't make sense. For example, Patrik Elias was on the ice for a minuscule number of goals against relative to his ice time. Yes, he was a great two-way forward, but clearly some of that was due to him playing in front of Martin Brodeur (and generally getting a lot of ice time with Scott Stevens and/or Niedermayer).

All that being said - I'm not saying that Crosby is at a disadvantage relative to any of the other players listed above. Just that it's a potentially important factor that hasn't been considered. Maybe he played in a more favourable environment than some of these players?

I realize some people are debating it, but I think it's laughable that Fleury is even being talked about as a potential Hall of Famer. He's spent his entire career stopping the puck at essentially the league average rate, and was never a Vezina trophy finalist in his long, healthy career.

Going back to the main point of the topic - it would be interesting to combine the goals against with goals for data. For example, the data shows that Bergeron and Datsyuk are pretty clearly better defensively than Crosby and Ovechkin (which should surprise nobody). But are Crosby and Ovechkin more effective players at even-strength, taking into account their superior offense? I might work on this in the new year, time permitting.

I think it's possible that Datsyuk was the overall most effective player at this peak, in other words technically the best, but also how do we account for playing with Lidstrom? We do know he was still dominant player anyway when he finished 6th in points per game with over 60% corsi at 36 with Abdelkader and Helm as his ES linemates and Lidstrom retired.

 

[authentic]

I’d be more curious to see how Crosby matches up defensively against guys like Backstrom, Giroux, Marchand, and even Barkov. I do believe Crosby’s defensive game is greatly over exaggerated and overrated in general, but Using the best players in terms of defensive play to compare only proves what Crosby lacks unfairly.

For most of his career he was probably a bit worse defensively than Backstrom, but that's who Ovechkin plays with though so I'm not sure how that helps anyone believe Ovechkin is better defensively than Crosby. Barkov, no, Marchand probably close and Giroux is probably close too.

 

[GreatGonzo]

For most of his career he was probably a bit worse defensively than Backstrom, but that's who Ovechkin plays with though so I'm not sure how that helps anyone believe Ovechkin is better defensively than Crosby. Barkov, no, Marchand probably close and Giroux is probably close too.

I don’t see how anyone can suggest Ovechkin being better than Crosby, and on the flip side I don’t think it’s strong at all to build an argument about Crosby being strong defensively using Ovechkin as comparison.

I think Giroux and Marchand are close. I think Backstrom gets underrated defensively. Still better players to match against than the ones used

 

[Kairi Zaide]

Great, thanks for sharing.

The other thing that would be interesting to see (and maybe this is in your links, I haven't dug into them yet) - who has the biggest difference between the ratio when they're on vs off the ice? If, for example, Bergeron is on the ice for 1.2 ES GF for every 1 ES GA (a ratio of 1.20), but the Bruins as a whole have a ratio of 1.15, that's not necessarily more impressive than, say, Jamie Benn having a ratio of 1.00 if the Stars are 0.80 without him. (To be clear, those numbers are completely fabricated to illustrate the point - but I think this is the best way of figuring out who's the top ES performer over the past decade).

This is somewhat what the "Rel TM" does, except it's not "relative to team" (aka when the player is on vs off the ice, which is what you are talking about), but "relative to teammates". I suggest reading this well written article to understand the difference : Revisiting Relative Shot Metrics – Part 1

Basically, looking at GA (you can replace GA with any other metric), the formula given by Evolving-Hockey in their glossary is :
player on-ice GA/60 - weighted average of all teammates' on-ice GA/60 (weighted by player TOI% with teammate)
I am not sure if he means "weighted average [...] GA/60 without player" as the article I linked above (different writer) suggests, although I would assume so.

The article discusses issues with this method and proposes adjustments, although I'm not sure if they're used on Evolving-Hockey. One notable issue discussed is the "team effect", where a good player on an other really bad team will fare better in relative metrics than a player who's technically just as good but playing on a much better team. For instance, just continuing with the fictitious example you provide, while a 1.2:1.0 ratio versus a 1.15:1.0 ratio (Bruins with Bergeron vs Bruins without Bergeron) might not appear as good as a 1.0:1.0 ration versus a 0.8:1.0 ratio (Stars with Benn vs Stars without Benn), it might in reality be harder to achieve.

 

[Empoleon8771]

A lot of the work has already been done.

5v5

All situations (warning, as this inherently puts players who play significant time on the PK at a disadvantage. Also, RelTM stats are not available on E-H combining all situations)

You can play with the data here.

I think the best chart to look at from that data is the relTM xGA/60:

Ovechkin: +0.18
Kopitar: -0.02
Toews: +0.08
Bergeron: -0.16
Datsyuk: -0.23
Crosby: +0.06

I think this gives a reasonable description of Crosby's defensive ability. Datsyuk and Bergeron were amazing defensively and way better than Crosby, but I think everyone knew that, and there isn't a massive difference between Crosby and guys like Toews, Kopitar, Zetterberg and Kesler (last 2 were names I looked up, Kesler was at +0.06 and Zetterberg was at +0.03). With that, I feel like Crosby's reputation of being good defensively, just not on par with Bergeron and Datsyuk, is totally fair. Based on the fact that Toews, Kesler and Kopitar combined for 4 Selkes, Crosby probably should have had more Selke votes in the last decade.

 

[Gurglesons]

I’d be more curious to see how Crosby matches up defensively against guys like Backstrom, Giroux, Marchand, and even Barkov. I do believe Crosby’s defensive game is greatly over exaggerated and overrated in general, but Using the best players in terms of defensive play to compare only proves what Crosby lacks unfairly.

Interesting how none of those players have an Art Ross or a PPG close to Crosby which is why their moderately better defensive skills mean Jack shit.

 

[TheAngryHank]

I give Pens fans a hall pass in this department .
They have never seen a two way forward before.
Who in their history is the best they have had? Billy Guaren? I honestly can't remember one ,Stall?