Who deserves the Selke strictly based on numbers?

[Love]

Just looking at advanced metrics who has been the NHL’s best defensive forward this season?

And what stats should we primarily be looking at in order to determine who is the best defensively?

Click to read more...

 

[TomasHertlsRooster]

Using Evolving Hockey's GAR and xGAR models, I added EVD GAR and SHD GAR for all forwards to calculate "Defensive GAR", then I added EVD xGAR and SHD xGAR for all forwards to calculate "Defensive xGAR", then took the average of those two metrics. Here are the forwards who rank highest:

As is often the case with defensive metrics for forwards, the guys who sit at the top of the charts for suppression metrics are often not the names that we would expect. Even though GAR is derived from RAPM which uses teammates, competition, usage, etc as predictor variables to isolate a player's impact from his context, it's still much easier to play shutdown, low-event hockey in a 4th line role against other 4th liners than it is to do so playing top competition. Normally I would say just take a look at a list like this, and hand the Selke to a guy who doesn't play super soft minutes, but almost all of the guys near the top play super soft minutes. So I think my vote might still go to Nichushkin, since his results are just off the charts.

 

[De Montreal]

Using Evolving Hockey's GAR and xGAR models, I added EVD GAR and SHD GAR for all forwards to calculate "Defensive GAR", then I added EVD xGAR and SHD xGAR for all forwards to calculate "Defensive xGAR", then took the average of those two metrics. Here are the forwards who rank highest:

View attachment 329873

As is often the case with defensive metrics for forwards, the guys who sit at the top of the charts for suppression metrics are often not the names that we would expect. Even though GAR is derived from RAPM which uses teammates, competition, usage, etc as predictor variables to isolate a player's impact from his context, it's still much easier to play shutdown, low-event hockey in a 4th line role against other 4th liners than it is to do so playing top competition. Normally I would say just take a look at a list like this, and hand the Selke to a guy who doesn't play super soft minutes, but almost all of the guys near the top play super soft minutes. So I think my vote might still go to Nichushkin, since his results are just off the charts.

Interesting. With his offensive number, Blake Coleman could certainly gain some votes for the real thing

 

[GlitchMarner]

Using Evolving Hockey's GAR and xGAR models, I added EVD GAR and SHD GAR for all forwards to calculate "Defensive GAR", then I added EVD xGAR and SHD xGAR for all forwards to calculate "Defensive xGAR", then took the average of those two metrics. Here are the forwards who rank highest:

View attachment 329873

As is often the case with defensive metrics for forwards, the guys who sit at the top of the charts for suppression metrics are often not the names that we would expect. Even though GAR is derived from RAPM which uses teammates, competition, usage, etc as predictor variables to isolate a player's impact from his context, it's still much easier to play shutdown, low-event hockey in a 4th line role against other 4th liners than it is to do so playing top competition. Normally I would say just take a look at a list like this, and hand the Selke to a guy who doesn't play super soft minutes, but almost all of the guys near the top play super soft minutes. So I think my vote might still go to Nichushkin, since his results are just off the charts.

No offense since you're approaching this analytically and not subjectively... but..

How are Bozak, Kovalchuk and JVR in the top 40?

Maybe all of them suddenly improved immensely on defense or maybe the fact that they're all there is an indication that one should be skeptical of the numbers in the last column.

James van Riemsdyk could score goals in TOR, but he was anything but good defensively. Seeing him on the chart is basically the equivalent of seeing Kessel on it in terms of shock value.

 

[TomasHertlsRooster]

No offense since you're approaching this analytically and not subjectively... but..

How are Bozak, Kovalchuk and JVR in the top 40?

Maybe all of them suddenly improved immensely on defense or maybe the fact that they're all there is an indication that one should be skeptical of the numbers in the last column.

James van Riemsdyk could score goals in TOR, but he was anything but good defensively. Seeing him on the chart is basically the equivalent of seeing Kessel on it in terms of shock value.

Yeah, it is pretty strange to see them all that high. However, all 3 of those players have terrible defensive metrics at the career level:

It's actually striking how similar they are in EVD GAR at the career level; all atrocious. In addition, Kovalchuk's EV RAPM xGA/60 in 2008-2009 was literally the worst of all time. Nearly 5 standard deviations above average!

So, this data isn't telling us these guys were always great defensively. It's actually supporting the conclusion that all of them sucked for most of their careers. It's just also telling us that over the course of this one season, they've been pretty good.

 

[Bank Shot]

It's just also telling us that over the course of this one season, they've been pretty good.

Or its just a list that shows how coaching affects some guys metrics more than talent actually does.

 

[SympathyForTheDevils]

Or its just a list that shows how coaching affects some guys metrics more than talent actually does.

Which would be in line with the idea that coaching affects players' defensive performance more than talent does. Which it likely does.

 

[alko]

We had some discussion about this trophy last year and the conclusion was basically that it is a imaginary reputation trophy.

 

[Love]

Which would be in line with the idea that coaching affects players' defensive performance more than talent does. Which it likely does.

So would it be accurate to say that one way to identify a good defensive forward would be to find a player on a bad defensive team (aka poorly coached team) who still has impressive defensive GAR? Because then we would know it’s about the player and not the coach/system.

 

[Chili]

Trophy is voted on by the media.

Would be interested to see both a player's and coaches vote.

 

[Trap Jesus]

Once you start getting into individual advanced analytics, Bergeron stands out again.

 

[TeddyBare]

From the discussion on the athletic

"Based on the criteria, Anthony Cirelli should be the front-runner as I noted at the half-season mark. He plays some of the toughest minutes in the league and is the primary reason the Lightning have elevated their defensive game this season. Of the top 10 candidates, he has the best goal suppression and the second-best expected goal suppression, and that goes the same by his regularized adjusted plus-minus in the same categories. He’s also one of two players listed to play more than half his team’s penalty kill minutes, something only 15 other forwards (including Philip Danault in sixth) in the league can boast."

Nichuskin has some nice metrics, but he is 11th among forwards on just the colorado avalanche in 5v5 TOI% vs elite competition at 29.3%
Anthony cirelli is 2nd among forwards on the lightning in 5v5 TOI% vs elite competition.

 

[Scandale du Jour]

Interesting to see that our entire 4th line (ZAR 3rd, Tanev 10th, Blueger 22nd) is in the top 30.

 

[tarheelhockey]

I know this isn't the right way to think about it in terms of advanced metrics, but in terms of how voters tend to think of this sort of thing:

Sebastian Aho:
- 6th in goals, top-20 in points
- Skated a very regular shift on the PK
- T-1st in SHG
- On the low end in PIM

Kind of a Bergeron-lite on defense, a Bergeron-plus on offense. That will likely draw votes.

 

[mynamejeff420]

I know this isn't the right way to think about it in terms of advanced metrics, but in terms of how voters tend to think of this sort of thing:

Sebastian Aho:
- 6th in goals, top-20 in points
- Skated a very regular shift on the PK
- T-1st in SHG
- On the low end in PIM

Kind of a Bergeron-lite on defense, a Bergeron-plus on offense. That will likely draw votes.

The issue with Aho is he is actually quite terrible defensively, so he wouldn't be a great pick for the Selke. It would be like giving the Selke to someone like Kessel or Ovechkin.

 

[tarheelhockey]

The issue with Aho is he is actually quite terrible defensively, so he wouldn't be a great pick for the Selke. It would be like giving the Selke to someone like Kessel or Ovechkin.

You know we’re talking about the Carolina one, right?

 

[mynamejeff420]

You know we’re talking about the Carolina one, right?

Don't get me wrong, I love Aho and he's one of the most talented (offensively) players in the league, and still one of the most valuable players in the league too. But defense is not one of his strong suits. As much as I hate to say it, it's one of the worst in the league. Doesn't mean he's bad, just means that he shouldn't be anywhere near the terms "Selke candidate" or "Bergeron-lite"

 

[tarheelhockey]

Don't get me wrong, I love Aho and he's one of the most talented (offensively) players in the league, and still one of the most valuable players in the league too. But defense is not one of his strong suits. As much as I hate to say it, it's one of the worst in the league. Doesn't mean he's bad, just means that he shouldn't be anywhere near the terms "Selke candidate" or "Bergeron-lite"

"one of the worst in the league. Doesn't mean he's bad"

I don't even know what to say to that.

I'll leave this up to the jury of our peers to actually watch this player and decide whether he is in fact one of the worst defensive players in the league.

 

[danyhabsfan]

Once you start getting into individual advanced analytics, Bergeron stands out again.

Seems way better than the other one mentionned previously.

Bergeron, Barkov, Stone, Danault.

 

[Chrisinroch]

You really think Eriksson Ek plays soft minutes? It seems he plays a pretty traditional “shut down” role against 1 and 2.

 

[TomasHertlsRooster]

"one of the worst in the league. Doesn't mean he's bad"

I don't even know what to say to that.

I'll leave this up to the jury of our peers to actually watch this player and decide whether he is in fact one of the worst defensive players in the league.

The poster was saying he is one of the worst defensive forwards in the league, which doesn't mean he's a bad player because his offensive impact out-weighs his defensive flaws. You're cherry-picking a portion of his statement to make it seem hypocritical and nonsensical but it makes perfect sense when you don't ignore the context and it's pretty clear what he is saying.

Also, this thread literally asks who deserves the Selke strictly based on numbers. Aho's even-strength RAPM xGA/60 (isolated impact on expected goals against) is 7th worst out of the 103 forwards who've played at least 900 minutes this year, his 5v5 xGA/60 is 3rd worst among 10 Canes forwards with at least 500 minutes played, behind rookie Martin Necas and noted defensive drag Ryan Dzingel. If all of that is still too fancy for you, the Hurricanes allow 3 goals per hour at 5v5 when Aho is on the ice and 2.36 goals per hour when he is on the bench.

The vapid "actually watch this player" doesn't really ever hold much merit unless you can explain what you're seeing the player do well that differs from their results, but in this thread with a title that specifies analyzing things "strictly based on numbers", the guy whose team allows 5v5 goals at at a 27% higher rate when he's on the ice doesn't belong.

 

[tarheelhockey]

The poster was saying he is one of the worst defensive forwards in the league, which doesn't mean he's a bad player because his offensive impact out-weighs his defensive flaws. You're cherry-picking a portion of his statement to make it seem hypocritical and nonsensical but it makes perfect sense when you don't ignore the context and it's pretty clear what he is saying.

Also, this thread literally asks who deserves the Selke strictly based on numbers. Aho's even-strength RAPM xGA/60 (isolated impact on expected goals against) is 7th worst out of the 103 forwards who've played at least 900 minutes this year, his 5v5 xGA/60 is 3rd worst among 10 Canes forwards with at least 500 minutes played, behind rookie Martin Necas and noted defensive drag Ryan Dzingel. If all of that is still too fancy for you, the Hurricanes allow 3 goals per hour at 5v5 when Aho is on the ice and 2.36 goals per hour when he is on the bench.

The vapid "actually watch this player" doesn't really ever hold much merit unless you can explain what you're seeing the player do well that differs from their results, but in this thread with a title that specifies analyzing things "strictly based on numbers", the guy whose team allows 5v5 goals at at a 27% higher rate when he's on the ice doesn't belong.

I never argued that he deserved the award based purely on advanced metrics.

In fact, I led my original post by saying "this isn't how an advanced analysis would look, but these are the numbers that Selke voters often look at". My point was that Aho wouldn't necessarily be a good choice, but he's a likely choice based on how the vote is typically framed around certain commonly-accessible and easily-comprehensible numbers.

Nevertheless, it's complete nonsense to suggest that Aho's a bad defensive player. "Actually watch this player" is not actually a vapid response to someone suggesting that he's worse defensively than e.g. Erik Haula or Nino Niederreiter. You need to understand that this is an eyebrow-raisingly strange claim to make to someone who actually follows the team closely.

 

[mynamejeff420]

I never argued that he deserved the award based purely on advanced metrics.

In fact, I led my original post by saying "this isn't how an advanced analysis would look, but these are the numbers that Selke voters often look at". My point was that Aho wouldn't necessarily be a good choice, but he's a likely choice based on how the vote is typically framed around certain commonly-accessible and easily-comprehensible numbers.

Nevertheless, it's complete nonsense to suggest that Aho's a bad defensive player. "Actually watch this player" is not actually a vapid response to someone suggesting that he's worse defensively than e.g. Erik Haula or Nino Niederreiter. You need to understand that this is an eyebrow-raisingly strange claim to make to someone who actually follows the team closely.

I didn't disagree with the initial premise of your post, the way the Selke is voted on is silly, it should be going to the best defensive player, not the "good defensive player who puts up strong offensive totals". The apprehension only comes from calling Aho even a decent defensive player. Frankly, he's just flat out bad defensively. He allows goals, shots, and expected goals at a much higher rate than the rest of his team, and as someone who watches a fair amount of Hurricanes games, I didn't see anything that would contradict those stats aside from "he plays on the penalty kill so he must be good defensively".

The point of defense is to prevent goals/shots/expected goals, and Aho does a terrible job at that. So if he is in fact "good defensively", he hasn't been able to accomplish the one purpose of playing defense, which is preventing offense. So either he's good defensively but he can't translate that in to results (which seems pretty contradictory imo) or he just isn't good defensively.

 

[tarheelhockey]

I didn't disagree with the initial premise of your post, the way the Selke is voted on is silly, it should be going to the best defensive player, not the "good defensive player who puts up strong offensive totals". The apprehension only comes from calling Aho even a decent defensive player. Frankly, he's just flat out bad defensively. He allows goals, shots, and expected goals at a much higher rate than the rest of his team, and as someone who watches a fair amount of Hurricanes games, I didn't see anything that would contradict those stats aside from "he plays on the penalty kill so he must be good defensively".

The point of defense is to prevent goals/shots/expected goals, and Aho does a terrible job at that. So if he is in fact "good defensively", he hasn't been able to accomplish the one purpose of playing defense, which is preventing offense. So either he's good defensively but he can't translate that in to results (which seems pretty contradictory imo) or he just isn't good defensively.

Aho is a responsible defensive player. If you want to say that Jordan Staal or Teuvo Teravainen are better defensively, I'll agree. If you want to make that argument about the likes of Brock McGinn and Jordan Martinook, I'll urge you to consider deployment and matchups, and for that matter roles in general. If you want to say he's worse defensively than Erik Haula or Andrei Svechnikov, at that point I'm just unable to respond with anything better than "you need better metrics because these aren't working".

 

[mynamejeff420]

Aho is a responsible defensive player. If you want to say that Jordan Staal or Teuvo Teravainen are better defensively, I'll agree. If you want to make that argument about the likes of Brock McGinn and Jordan Martinook, I'll urge you to consider deployment and matchups, and for that matter roles in general. If you want to say he's worse defensively than Erik Haula or Andrei Svechnikov, at that point I'm just unable to respond with anything better than "you need better metrics because these aren't working".

Better metrics than "goals" ? Isn't a mark of a good defensive player "preventing goals"?