Post-Game Talk: #39 | Penguins 5 at FLYERS 1 | Tue., Jan. 2, 2018, 7:00 pm ET

[Delete99991]

Oh of course it's flawed. It's hockey. All stats are flawed. One could make the same argument about Corsi (to far greater effect, in my opinion).

But what we know for sure is that removing special teams, if everyone on a team finishes +12 for the year, the team will have a positive goal differential of something like +36 to +48. No room for argument.

Further, if we grant that goal differential is a "bottom line stat" in terms of Wins and Losses, we also know that Hagg is either a) better than the median of his competiton, b) lucky, or c) both, given that he's got the highest +/- on the team and has played mainly with the human timebomb known as A-Mac.

So it's not nothing. That smug HF hockey fan thing of "pfft, +/-" (not that you've taken that attitude in this discussion FlyGuy - you haven't - but others do) is silly when those same people are so often devotees of Corsi and the like.

 

[JojoTheWhale]

Oh of course it's flawed. It's hockey. All stats are flawed. One could make the same argument about Corsi (to far greater effect, in my opinion).

But what we know for sure is that removing special teams, if everyone on a team finishes +12 for the year, the team will have a positive goal differential of something like +36 to +48. No room for argument.

But we know that extremely high/low shooting percentages for/against aren’t sustainable over the long haul, right? No one reasonable is saying those events did not occur. The issue is repeatability.

I understand that you think Corsi is more flawed and that’s certainly your right, but when it comes to repeatability and predictive quantification of play, it’s not an opinion as to whether sample sizes of ~63 per (NHL average shot total this year and an admitted oversimplification) are more instructive than those of ~6 (average Goals) per over an 82 game schedule.

Of course raw Corsi is not the best way we have to measure these things anymore, either. We’re only recently getting to the point where we can account for things like shot quality, game situations, and even venue adjustments. Those are all huge steps that needed to be taken. As we refine models for Expected Goals and the like, we’ll be getting closer to what some other sports have.

 

[Striiker]

+/- is useless, eh.

Literally saying that goals for and goals against are useless information.

I love this place

+/- is a worthless stat for individuals.

If someone is on the ice and does EVERYTHING perfect (not possible, but imagine) they can still get a minus because of teammates failures. So using it for individuals obviously makes no sense.

 

[Delete99991]

+/- is a worthless stat for individuals.

If someone is on the ice and does EVERYTHING perfect (not possible, but imagine) they can still get a minus because of teammates failures. So using it for individuals obviously makes no sense.

By that logic, if a guy's a +12 on a garbage team he must be a stud.

Listen I get it, it's one metric among many and there's a big heat map around it. Fine.

But it is *highly* unlikely that a guy who is a +12 on a bad team through 30-odd games is a bum.

 

[Striiker]

By that logic, if a guy's a +12 on a garbage team he must be a stud.

Listen I get it, it's one metric among many and there's a big heat map around it. Fine.

But it is *highly* unlikely that a guy who is a +12 on a bad team through 30-odd games is a bum.

No, by that logic the +12 still means absolutely nothing, just like a -12 wouldn't.

For all we know, the +12 guy could have been absolutely HORRIBLE and still got a plus because his teammates scored a goal. Even if he did absolutely nothing to positively contribute to that goal, he still gets that plus, which is why it's a useless stat for judging individuals.

 

[Delete99991]

No, by that logic the +12 still means absolutely nothing, just like a -12 wouldn't.

For all we know, the +12 guy could have been absolutely HORRIBLE and still got a plus because his teammates scored a goal. Even if he did absolutely nothing to positively contribute to that goal, he still gets that plus, which is why it's a useless stat for judging individuals.

On a bad team, said player would have to be ridiculously lucky for his teammates/linemates to score with no input whatsoever from him, AND to prevent goals on the other end with no input whatsoever from him.

Frankly, if there's that much luck involved we're all wasting our time watching the sport.

 

[Striiker]

On a bad team, said player would have to be ridiculously lucky for his teammates/linemates to score with no input whatsoever from him, AND to prevent goals on the other end with no input whatsoever from him.

Frankly, if there's that much luck involved we're all wasting our time watching the sport.

The fact that Hagg is currently +12 proves it's possible.

Also, do you believe MacDonald and Hagg are better than Provorov and Ghost? Because that's what +/- tells us.

If your answer is "yes" then you're either out of your mind insane or just trolling me, either way I'm done.

 

[Rebels57]

I wish they would just get rid of +/- altogether.

Worthless stat that leads to poor player evaluation.

 

[SolidSnakeUS]

I wish they would just get rid of +/- altogether.

Worthless stat that leads to poor player evaluation.

We're finally gonna get rid of it in my main fantasy league.

 

[Magua]

Hagg's #8 predecessor led the Flyers in +/- in 2015 when we finished 7th from the bottom.

 

[Curufinwe]

Gross.