2018-19 stats and underlying metrics thread

[Ukkosenjumala]

Okay, so which players have taken a performance dip compared to last season? Here is a chart that shows the difference in expected goals plus-minus per 60 minutes (xGPM_60) from last year to this year. This statistic uses a model to adjust for quality of teammate and competition, among other factors.

It's pretty clear that most Jets have worse metrics, but the huge drops in performance are among those who the Jets rely on a lot, including Scheifele, Lowry, Wheeler, Connor and Perreault. Interestingly, the D haven't declined as much, and Myers is actually quite a bit better than last season (the only Jet with a strong improvement). Laine hasn't changed much, but that's because he doesn't perform well in this metric in either season.

Unless the Jets get Scheifele and Wheeler sorted out, along with their secondary forwards, they are going to continue to be under water. I think Maurice really should have tried to shake things up a bit with Scheifele and Wheeler, probably by splitting them up. I think they've become too stale and too confident in their offense as their key weapon.

View attachment 196361

What are their actual xGPM.60 this year? Scheifele has about the same shooting percentage so he's not getting any luckier than last year and he's scored more goals, so is he getting torched by the competition on the defensive end or is he still on the plus side?

 

[Whileee]

What are their actual xGPM.60 this year? Scheifele has about the same shooting percentage so he's not getting any luckier than last year and he's scored more goals, so is he getting torched by the competition on the defensive end or is he still on the plus side?

He's giving up a lot more xGA than last season, when he was about even. He's also dropped in xGF. I've been posting about it since early in the season. The CSW line has been really bad, with soft D zone play, poor zone exits, and getting hemmed. They've also had way too many penalties taken when they are on the ice, often due to a bad shift in their own zone. Their offensive brilliance has masked the problems in their overall play. I'm really surprised Maurice hasn't shaken things up by separating them.

They have very positive results with Ehlers on their wing, but when he was injured they went way back into the red on shot metrics with Connor back on LW.

 

[DashingDane]

Okay, so which players have taken a performance dip compared to last season? Here is a chart that shows the difference in expected goals plus-minus per 60 minutes (xGPM_60) from last year to this year. This statistic uses a model to adjust for quality of teammate and competition, among other factors.

It's pretty clear that most Jets have worse metrics, but the huge drops in performance are among those who the Jets rely on a lot, including Scheifele, Lowry, Wheeler, Connor and Perreault. Interestingly, the D haven't declined as much, and Myers is actually quite a bit better than last season (the only Jet with a strong improvement). Laine hasn't changed much, but that's because he doesn't perform well in this metric in either season.

Unless the Jets get Scheifele and Wheeler sorted out, along with their secondary forwards, they are going to continue to be under water. I think Maurice really should have tried to shake things up a bit with Scheifele and Wheeler, probably by splitting them up. I think they've become too stale and too confident in their offense as their key weapon.

View attachment 196361

A start would be to play the players that aren't slumping more... Copp at 12:07 ATOI is pretty ridiculous with how he has played. Scheifele, Wheeler and Connor with more than 2min on average compared to other forwards just makes no sense to me with how they are playing this season.

 

[DashingDane]

 

[Duke749]

A start would be to play the players that aren't slumping more... Copp at 12:07 ATOI is pretty ridiculous with how he has played. Scheifele, Wheeler and Connor with more than 2min on average compared to other forwards just makes no sense to me with how they are playing this season.

Connor is the real odd one there. I don’t see what he’s done to garner that kind of trust over someone like Ehlers.

 

[DRW204]

Connor is the real odd one there. I don’t see what he’s done to garner that kind of trust over someone like Ehlers.

KC does PK though and sees PP1 time. EV/TOI: KC - 15:11 Ehlers - 14:49 per game. Special teams time: 4:07 per game for KC and 1:21 for Ehlers. Doesn't answer your question but provides a bit of commentary.

 

[Duke749]

KC does PK though and sees PP1 time. EV/TOI: KC - 15:11 Ehlers - 14:49 per game. Doesn't answer your question but provides a bit of commentary

He’s really not adding much to PP1. He’d be much more useful as a trigger man on PP2 or simply rotating with Wheeler for a potential one timer or cycling the puck back up top and have Wheeler slide in front for a quick screen or deflection.

EDIT: He also doesn’t go to the net enough. He can lift the puck from in tight. Try it some more. And he doesn’t give a quick one touch to Scheif in the slot. Of course lord knows Scheif hasn’t been open in the slot in three months.

 

[DashingDane]

Connor is the real odd one there. I don’t see what he’s done to garner that kind of trust over someone like Ehlers.

I guess Paul thinks he is a good defensive player or at least good enough to pk. I don't see it but would also not claim to be a hockey coach lol

 

[DRW204]

He’s really not adding much to PP1. He’d be much more useful as a trigger man on PP2 or simply rotating with Wheeler for a potential one timer or cycling the puck back up top and have Wheeler slide in front for a quick screen or deflection.

agreed. I think Hayes should get a look in KC's spot on PP1 & KC become the focal shooter on PP2.

 

[Duke749]

I guess Paul thinks he is a good defensive player or at least good enough to pk. I don't see it but would also not claim to be a hockey coach lol

PKing isn’t the same as defending 5on5. He makes bad reads all the time 5on5. PK is more about reacting quick and using that quickness. I’m not seeing it either.

 

[ps241]

Okay, so which players have taken a performance dip compared to last season? Here is a chart that shows the difference in expected goals plus-minus per 60 minutes (xGPM_60) from last year to this year. This statistic uses a model to adjust for quality of teammate and competition, among other factors.

It's pretty clear that most Jets have worse metrics, but the huge drops in performance are among those who the Jets rely on a lot, including Scheifele, Lowry, Wheeler, Connor and Perreault. Interestingly, the D haven't declined as much, and Myers is actually quite a bit better than last season (the only Jet with a strong improvement). Laine hasn't changed much, but that's because he doesn't perform well in this metric in either season.

Unless the Jets get Scheifele and Wheeler sorted out, along with their secondary forwards, they are going to continue to be under water. I think Maurice really should have tried to shake things up a bit with Scheifele and Wheeler, probably by splitting them up. I think they've become too stale and too confident in their offense as their key weapon.

View attachment 196361

That is telling and ugly.

 

[Ducky10]

He’s really not adding much to PP1. He’d be much more useful as a trigger man on PP2 or simply rotating with Wheeler for a potential one timer or cycling the puck back up top and have Wheeler slide in front for a quick screen or deflection.

EDIT: He also doesn’t go to the net enough. He can lift the puck from in tight. Try it some more. And he doesn’t give a quick one touch to Scheif in the slot. Of course lord knows Scheif hasn’t been open in the slot in three months.

Every time I see KC on the power play, all I see him do is get rid of the puck before it explodes. I really don't see what his purpose it there, he should be an alternative passer to Scheifele in the bumper spot but he's not. I think they should try Hayes there, good passer and a bigger body in front.

 

[Whileee]

A start would be to play the players that aren't slumping more... Copp at 12:07 ATOI is pretty ridiculous with how he has played. Scheifele, Wheeler and Connor with more than 2min on average compared to other forwards just makes no sense to me with how they are playing this season.

Agree.

 

[Jet]

+/- is next to useless.

No it's not.

It's something the fancystats people want everyone to believe and it's now become a meme.

It's certainly an imperfect measure and shouldn't be taken in isolation, but it has value.

There - I said it.

 

[JetsFan815]

No it's not.

It's something the fancystats people want everyone to believe and it's now become a meme.

It's certainly an imperfect measure and shouldn't be taken in isolation, but it has value.

There - I said it.

What is the advantage of +/- over 5 on 5 goal differential?

 

[Jet]

What is the advantage of +/- over 5 on 5 goal differential?

I am not talking about advantage or perfect statistics (because I don't think they exist).

I think that a player who, year over year, is a plus demonstrates that they are a net positive.

That takes into account some of the quirks of plus minus and also things like team and partner impacts.

Again, I am not saying it's a perfect measure, nor should it be used in isolation to prove anything, but its hardly useless.

 

[Halberdier]

No it's not.

It's something the fancystats people want everyone to believe and it's now become a meme.

It's certainly an imperfect measure and shouldn't be taken in isolation, but it has value.

There - I said it.

Ordinary +/- which counts SH goals as plus and PP goals against as minuses is more than problematic.

Other than that, I agree. Real goals matter.

To check how well some guys have been working alone or together in 5-on-5, you need to check 5-on-5 goal differential, or "5-on-5 +/-", if you wish.

All situations goal differential is also interesting in some sense, but that normal +/- is useful only when comparing guys that have similar usage or ES/PP/PK TOI distribution (and similar performing team & linemates as well, obviously).

The goal of this game is to score more goals than your opponent, not having the best Corsi, sCorsi or xGoals. It's not good than when your best 5-on-5 +/- guy Laine of seasons 16-17 and 17-18 including playoffs all in the sudden is one of the worst if not worst on 18-19. Haven't checked his 5-on-5 goal differential for 18-19, but I guess it's bad.

 

[Tommigun]

Ordinary +/- which counts SH goals as plus and PP goals against as minuses is more than problematic.

Other than that, I agree. Real goals matter.

To check how well some guys have been working alone or together in 5-on-5, you need to check 5-on-5 goal differential, or "5-on-5 +/-", if you wish.

All situations goal differential is also interesting in some sense, but that normal +/- is useful only when comparing guys that have similar usage or ES/PP/PK TOI distribution (and similar performing team & linemates as well, obviously).

The goal of this game is to score more goals than your opponent, not having the best Corsi, sCorsi or xGoals. It's not good than when your best 5-on-5 +/- guy Laine of seasons 16-17 and 17-18 including playoffs all in the sudden is one of the worst if not worst on 18-19. Haven't checked his 5-on-5 goal differential for 18-19, but I guess it's bad.

+/- doesn't count PP goals as a + if your team is on the PP, nor does it count it as - if your team is killing the penalty. If your team lets in a goal while shorthanded you should get a -, and the other team should get a +. Don't see the problem, except it does count empty netters but whatever, they are few.

 

[Halberdier]

I am not talking about advantage or perfect statistics (because I don't think they exist).

I think that a player who, year over year, is a plus demonstrates that they are a net positive.

That takes into account some of the quirks of plus minus and also things like team and partner impacts.

Again, I am not saying it's a perfect measure, nor should it be used in isolation to prove anything, but its hardly useless.

Problem with that is you get minuses from PP you play, and zero compensation from PP goals for or PK you don't play. That's the exact reason why you cannot look +/- alone to decide whether or not some guy is "net positive", as that +/- by design omits important portions of tha "net production".

I tried to implement a synthetic +/- stat that gave you +0.5 for every PP goal for and -2.0 for every SH goals against. (And +2.0 for every SH goals for and -0.5 for every PP goals against to keep things equal). Something like that could decide very roughly whether some guy is winning his game or not.

But there is tremendously better stats for that actually: WAR. And also Corsica rating. And slightly less advanced: GAR.

Corsi or even xGoals in isolation tells you nothing.

 

[YWGinYYZ]

No it's not.

It's something the fancystats people want everyone to believe and it's now become a meme.

It's certainly an imperfect measure and shouldn't be taken in isolation, but it has value.

There - I said it.

Yes it is. It's pretty much random, and has arbitrary rules for assignment of +/-

Have a read: Analysis: Plus/Minus Sucks

 

[Calendal]

No it's not.

It's something the fancystats people want everyone to believe and it's now become a meme.

It's certainly an imperfect measure and shouldn't be taken in isolation, but it has value.

There - I said it.

I think what you describe can be expressed as ”next to useless”. It’s not completely useless but when you have much better measures like 5v5 goal differential why would you ever prefer plus-minus?

 

[mzappa]

What are everyone's favourite analytics sites? I'm familiar with naturalstattrick, but not much else. If someone can recommend sites with fun graphs, that works!

 

[Whileee]

What are everyone's favourite analytics sites? I'm familiar with naturalstattrick, but not much else. If someone can recommend sites with fun graphs, that works!

Evolving Hockey has some great stuff.

 

[scelaton]

I think what you describe can be expressed as ”next to useless”. It’s not completely useless but when you have much better measures like 5v5 goal differential why would you ever prefer plus-minus?

For most people +/- is simply more easily accessible. I wouldn't call it next to useless at all. It's useful, provided its limitations are understood. Other stats, like 5x5 g diff are better, but if very few are accessing or talking about them on a social media platform, they are thus limited too.

 

[YWGinYYZ]

For most people +/- is simply more easily accessible. I wouldn't call it next to useless at all. It's useful, provided its limitations are understood. Other stats, like 5x5 g diff are better, but if very few are accessing or talking about them on a social media platform, they are thus limited too.

Studies have shown it to be close to random.