NHL Master Thread

[yuneeq]

I would expect comparable players that the one with more ozone face offs would have more points/goals. This is Toews and Kane. Again this is Toews and Kane.
This is what the +/- shows about these two players but you say it is useless. Kane is better offensively but sucks on defensive. Toews is good offensively and great on defense.  The +/- should be higher for Toews and it is.

I’m not saying these numbers contradict what you think about Toews and Kane. What I’m saying is these are far more accurate metrics to see how good or bad each one is on offense/defense respectively. Plus minus does not paint an accurate picture. For Kane - it considers his PP contributions negatively (it only counts shorties against), and also has empty net goals for/against making it even more unreliable for both players. In fact when the BH pull their goalie Toews is more likely to get a minus despite that being the likely outcome and the trusted player in that situation. And Kane might play more often than Toews when BH opponent pulls their goalie, making it seem like he’s better than he really is. In short, plus minus is not a useful stat even if you watch the team, and completely useless if you don’t watch them often.

The spread between Toews and Kane should be higher than what plus minus indicates, and for other players it can be the reverse.

[CountValentine]

I’m not saying these numbers contradict what you think about Toews and Kane. What I’m saying is these are far more accurate metrics to see how good or bad each one is on offense/defense respectively. Plus minus does not paint an accurate picture. For Kane - it considers his PP contributions negatively (it only counts shorties against), and also has empty net goals for/against making it even more unreliable for both players. In fact when the BH pull their goalie Toews is more likely to get a minus despite that being the likely outcome and the trusted player in that situation. And Kane might play more often than Toews when BH opponent pulls their goalie, making it seem like he’s better than he really is. In short, plus minus is not a useful stat even if you watch the team, and completely useless if you don’t watch them often.

The spread between Toews and Kane should be higher than what plus minus indicates, and for other players it can be the reverse.

All I am saying is +/- is good indicator of two players playing on the same team. It is not 100%.

You make assumptions that are not correct.  When opponents pull their goalie both Toews and Kane will be on the ice most of the time. When the BH pull their goalie Toews is on the ice way more than Kane. The reason is simple. Toews and Kane are both offensive players but Kane sucks on defense.

Your comment about PP contributions. +/- does not count those goals?

[yuneeq]

Ok got it. Where do I find this stat. I am guessing Toews will be better than Kane.

Just looked up a few time periods, and at even strength Kane was actually better overall in many important metrics, sometimes very clearly, though he did get a more ozone starts, especially in recent years.

https://www.naturalstattrick.com/playercompare.php?fromseason=20202021&thruseason=20222023&stype=2&sit=ev&score=all&stdoi=oi&rate=n&p1=8474141&p2=8473604&loc=B&gpfilt=none&fd=&td=&tgp=410&lines=single

What do you think about the goals +/- for the team, worthless?

Not worthless, actually it's more predictable about a team's future performance than points or wins.

[yuneeq]

All I am saying is +/- is good indicator of two players playing on the same team. It is not 100%.

You make assumptions that are not correct.  When opponents pull their goalie both Toews and Kane will be on the ice most of the time. When the BH pull their goalie Toews is on the ice way more than Kane. The reason is simple. Toews and Kane are both offensive players but Kane sucks on defense.

You find plus minus useful for these 2 players that you watch all the time, but looking it up they've both been mostly bad with plus minus since 2017. PLus minus also tells you that Ryan Lindgren at +27 is better than Adam Fox who is a Norris winner and perennial top 3 Norris candidate. How is it insightful at all? The more player comparisons you look at the more obvious how flawed it is.

Plus minus is also a cumulative stat. So a hypothetical 2 identical players on a bad team and play identically in all situations, however one plays 20 minutes, the other plays 10 minutes. If the 10 minute player is a minus 10, the 20 minute player will likely be a -20 simply because he played more minutes. The -10 player not one iota better.

Your comment about PP contributions. +/- does not count those goals?

Plus minus does not count PP goals, but it counts shorthanded goals against. And for PK'ers, it doesn't count goals against, but counts shorthanded goals.

So a terrific PP player like Fox will have his PP impact completely ignored, and actually since he leads in TOI, and opponents scored 5 SHG, he's a minus 5 on the powerplay. And Lindgren never plays on the PP, but plays the PK and therefore is a plus 4 for being on the ice for 4 SHG.

[CountValentine]

You find plus minus useful for these 2 players that you watch all the time, but looking it up they've both been mostly bad with plus minus since 2017.

As it should be. So again +/- works. 

Plus minus is also a cumulative stat. So a hypothetical 2 identical players on a bad team and play identically in all situations, however one plays 20 minutes, the other plays 10 minutes. If the 10 minute player is a minus 10, the 20 minute player will likely be a -20 simply because he played more minutes. The -10 player not one iota better.

I have seen this many times. This is the classic playing with numbers. They claim "identical in all situations" and then change the most important variable. Even you can see this bogus argument.

[yuneeq]

As it should be. So again +/- works.  :)I have seen this many times. This is the classic playing with numbers. They claim "identical in all situations" and then change the most important variable. Even you can see this bogus argument.

If anything, the player with more minutes is trusted to play more, yet is looked at more negatively. That is the exact opposite purpose of what you are trying to glean with the stat. Bottom line, no respected hockey analyst relies on +- as an indicator of player ability, and there are many reasons why.

[CountValentine]

If anything, the player with more minutes is trusted to play more, yet is looked at more negatively.

There is many reasons one player plays more. No, it is not looked at negatively. That reasoning shows how numbers are twisted to fit a narrative. On a team with a positive +/- you would expect two identical players that the player with more playing time would have a greater +/-. That is common sense. Any respected hockey analyst understands that and calculates that in.

[CountValentine]

What happens every time he scores? 

Let's see what happens tonight.

[yuneeq]

There is many reasons one player plays more. No, it is not looked at negatively. That reasoning shows how numbers are twisted to fit a narrative. On a team with a positive +/- you would expect two identical players that the player with more playing time would have a greater +/-. That is common sense. Any respected hockey analyst understands that and calculates that in.

You may have a good argument if players were in fact identical (you don't, but not worth elaborating on this scenario) and you looked up TOI as well - however no 2 players are alike. Therefore to use +- you would have to look up other stats to find out exactly the players usage, and at that point you eliminated any purpose of looking at a supposedly all-encompassing +- stat. 5v5 GF% tells you 100x more about a players overall contribution with 100x more accuracy and you know it in a single glance.

And to clarify - earlier I was talking about team goal differential (that you see on the standings page). This metric includes goals scored in all situations, PP, PK, empty net, etc. It is not the same as the plus minus metric used for players.

[CountValentine]

You may have a good argument if players were in fact identical...

Hello, you posted that scenario. 
All I did was show they used numbers for their narrative but change the most important variable.

We all know there are hundreds of advanced stats. What these advance stats do is take a simple metric like +/- and break it down into hundreds of parts. They all have their use.

[yuneeq]

Hello, you posted that scenario. 
All I did was show they used numbers for their narrative but change the most important variable.

Indeed I did, to prove a point that even if they played identically you would find one player looking much better than the other when the truth can be the opposite. Type of usage and luck both play a huge role in +- for identical players, and since players are not identical it further muddies the waters beyond usability.

We all know there are hundreds of advanced stats. What these advance stats do is take a simple metric like +/- and break it down into hundreds of parts. They all have their use.

Not at all. Those metrics stay far away from +-, and show much more accurate ways of assessing a player. +- is for dinosaurs. This isn't even up for debate in the hockey world for the last 10 years.

[CountValentine]

Not at all. Those metrics stay far away from +-, and show much more accurate ways of assessing a player. +- is for dinosaurs. This isn't even up for debate in the hockey world for the last 10 years.

We come from different worlds. 

[CountValentine]

What happens every time he scores gets a point? 

Slight revision. 

[NTorch]

What happens every time he scores? 

That's nice, but we win games even when he does not score.

I must have missed it. Which game was that?

BH last night.

[CountValentine]

BH last night.

See revision.

[NTorch]

See revision.

I'm confused, he did not have any points last night either?

[CountValentine]

I'm confused, he did not have any points last night either?

He had an assist or did they take it away?
ETA: It was on Kreider's first goal, but it looks like they took it away. Anyone know why? @yuneeq

[yuneeq]

He had an assist or did they take it away?
ETA: It was on Kreider's first goal, but it looks like they took it away. Anyone know why? @yuneeq

Didn't notice that it was taken away, but I see that Trocheck deflected Fox's shot before Kreider scored on the rebound. Initially they must have missed the deflection and credited Kane as he passed to Fox.

[CountValentine]

but I see that Trocheck deflected Fox's shot before Kreider scored on the rebound.

That would do it.

[CountValentine]

Smart move to move Kane to a different line. If you want to recreate the Kane/Panarin magic, then you need a Keith on the back end. Something the Rangers do not have.