Miscellaneous NHL Talk XXX - 2019 Offseason

Status
Not open for further replies.

prototypical4thliner

Registered User
Jan 12, 2017
3,989
5,971
Theres no way to predict it so I am guessing they are using status quo which is why its heavily flawed and ultimately worthless.

Look at the difference between Calgary in 17-18 and then 18-19 after a coaching change and a few tweaks. Outhouse to Penthouse (though they did **** the bed in the playoffs).
Honestly it’s labor intensive but you would have to run the data for WAR for that year then individually analyze each player to prognosticate. Or develop some sort of algorithm to predict.

Either way not easy. I can see why we are as low as we are.

Next question is what moves WAR and how far? Because you have to think provorov and ghost return to a level closer to 17-18, Patrick improves, sanheim and Myers provide a better performance than hagg, Macdonald and gudas. Also, Hart is the real deal.
 
  • Like
Reactions: Rebels57

Rebels57

Former Flyers fan
Sponsor
Sep 28, 2014
76,383
122,697
Honestly it’s labor intensive but you would have to run the data for WAR for that year then individually analyze each player to prognosticate. Or develop some sort of algorithm to predict.

Either way not easy. I can see why we are as low as we are.

Next question is what moves WAR and how far? Because you have to think provorov and ghost return to a level closer to 17-18, Patrick improves, sanheim and Myers provide a better performance than hagg, Macdonald and gudas. Also, Hart is the real deal.

Yeah there a ton of factors that can add up to the Flyers having a major turn-around this season.
 
  • Like
Reactions: Ghosts Beer

Magua

Doer of Hoffific Things
Apr 25, 2016
37,169
154,076
Huron of the Lakes
My question is it predictive of this current year or do they assume status quo?

In order to make a predictive model, by nature, you’re using past values and assessments, whether on an individual basis or an age curve or whatever. I’m not sure how they weight everything.

The problem is in what you can’t predict because there are no numbers to properly assign. And we are dealing with a Flyers team that has flirted with both sides of mediocrity for most of a decade. People should probably not get crazy up in arms that a mathematical model predicts more mediocrity. This is a team that’s entering a season with much uncertainty.
 
  • Like
Reactions: kudymen

wasup

Registered User
Mar 21, 2018
2,443
2,283
Yeah there a ton of factors that can add up to the Flyers having a major turn-around this season.
Different coach should bring a different mindset to the players as well . There are a bunch of factors for last year but the Flyers were better than last years record .
 
  • Like
Reactions: Rebels57

prototypical4thliner

Registered User
Jan 12, 2017
3,989
5,971
In order to make a predictive model, by nature, you’re using past values and assessments, whether on an individual basis or an age curve or whatever. I’m not sure how they weight everything.

The problem is in what you can’t predict because there are no numbers to properly assign. And we are dealing with a Flyers team that has flirted with both sides of mediocrity for most of a decade. People should probably not get crazy up in arms that a mathematical model predicts more mediocrity. This is a team that’s entering a season with much uncertainty.
Oh I’m not annoyed by any stretch. I’m just going through the exercise of “what if”. If Patrick goes from a 30 point guy to a 50 point guy, what does that do? Et cetera et cetera
 

MacDonald4MVP

Registered User
May 7, 2016
9,931
5,269
Theres no way to predict it so I am guessing they are using status quo which is why its heavily flawed and ultimately worthless.

Look at the difference between Calgary in 17-18 and then 18-19 after a coaching change and a few tweaks. Outhouse to Penthouse (though they did **** the bed in the playoffs).
Teams over perform and underperform on a daily. Many players that underperformed last year wouldn't be rated all that high and vice versa unless you think Folin really was our second best dman and his on ice percentages have nothing to do with it.
Flyers had 5th worst PDO in the league and one might expect upswing, but let's not act as if it wasn't earned. Our goalie carousel was bad management and not bad luck. Our special teams have been slumping for way to long to chalk it up to bad fortune or small sample size either and let's not forget one important fact when it comes to corsi it makes a difference if you have the likes of Simmonds and Filppula get scoring chances or Kessel and Malkin as I'm pretty sure theres a massive difference when it comes to converting on those chances, much less when your whole offense is geared towards shots from the blue line.
 
  • Like
Reactions: Rebels57

Magua

Doer of Hoffific Things
Apr 25, 2016
37,169
154,076
Huron of the Lakes
Oh I’m not annoyed by any stretch. I’m just going through the exercise of “what if”. If Patrick goes from a 30 point guy to a 50 point guy, what does that do? Et cetera et cetera

I won't do this for every player, but take the 3rd line as shown: JVR -- Patrick -- Rubtsov (replacement value)

JVR was worth ~2 wins each of the previous 2 seasons before last. He was worth 0.2 last season. Patrick has been worth ~0.5 wins in his 2 seasons. It looks to me like they just added 0.2 + 0.5 + 0 together to get a line worth 0.7 wins. Well, if JVR bounces back and is at 1.5 wins; and if Patrick takes a step and is worth 1 win; and you substitute Frost/Farabee and they are worth any sort of positive value, you could be talking 2-3 more wins alone.

I look at Konecny or Voracek, who weren't graded kindly, and think there's more room to add. What it really comes down to is their projections shred our defense. 0.5 wins above replacement for 6 players is awful. That's a complicated issue I don't feel like getting into. Again, it's already imperfect and they can't model mystery effects.
 

Striiker

Former Flyers Fan
Jun 2, 2013
89,442
155,161
Pennsylvania
giphy.gif
 
  • Like
Reactions: Don Nachbaur 26

LegionOfDoom91

Registered User
Jan 25, 2013
81,647
138,659
Philadelphia, PA
You’re also getting dinged for what I’d imagine is a below average season statistically last year from Myers but also in a 21 game sample size.

On the flip side of that though I’d imagine you’re getting credit for Hart’s good season albeit in 31 game sample size.

It’s kind pointless looking at that in totality as no one can accurately pinpoint things. I doubt they finish as bad as 84 points pending something catastrophic. But moderately it’s probably fair to say they’re still a bubble team right now knowing what we know. In order to take that next step to become a sure fire playoff team & possibly win a playoff series. Some guys need to bounce back while the younger players either need to maintain &/or get better.
 

TheKingPin

Registered User
Nov 16, 2005
20,494
9,948
Philadelphia, PA
Where did you see this?

Oh...and we could trade for Marner. Look at this trade suggestion on the main boards:

Toronto:
Nolan Patrick
Ivan Provorov
1st 2020
1st 2021
Conditional 1st 2022 or 2nd 2021, depending on the contract

Philadelphia:
Mitch Marner

I forget. A news site. It’s out there so. So is that proposal.
 

TheKingPin

Registered User
Nov 16, 2005
20,494
9,948
Philadelphia, PA
The flyers appear to be very underrated this year. I think that’s a good thing. I think the major thing people aren’t taking into account is the really bad situation of having a horrible coach the firing the GM then the coach and replacing him with an interim level guy + 8 goalies. Those two things will throw literally any team off. Then people tend to think hayes’ contract amount will cause the team to not do well. I think you positive steps forward from Konecny, Hayes, Patrick, Lindblom, Provorov, Sanhiem, Ghost, Myers. Plus Frost. That’s a lot for a media source to assume, but I think we all know that most and more than likely all those players will be better than last year. That’s 8-9 guys.
 
  • Like
Reactions: Rebels57

MacDonald4MVP

Registered User
May 7, 2016
9,931
5,269
I'm deeply curious to know how they've calculated WAR but I also refuse to give them a click for this. Gudas...eh. MacDonald, get out of here with that.
I remember reading on one of those WAR models and it had insane amount of variables like faceoff%, hits blocks, shots, giveaways, takeaways, some relative metrics and even those xGF metrics. From what I remember it pretty much had everything that is tracked sans corsi or fenwick and it also depended on actual on ice scoring.

After looking at average over the whole 11 seasons they have on evolvingwild and with 6000 minutes min toi here are some tidbits:

Ellis rates as best d, maata is at 13, Keith and Doughty didn't even crack top60.

Among forwards it's Datsyuk, McJesus, Stone, Crosby and Hornqvist? as top5.
G is at 40, 4th liner in Marcus Foligno is at 91 ahead of Voracek. Erik Staal is 161 just below Darren Helm.
 

Rebels57

Former Flyers fan
Sponsor
Sep 28, 2014
76,383
122,697
I remember reading on one of those WAR models and it had insane amount of variables like faceoff%, hits blocks, shots, giveaways, takeaways, some relative metrics and even those xGF metrics. From what I remember it pretty much had everything that is tracked sans corsi or fenwick and it also depended on actual on ice scoring.

After looking at average over the whole 11 seasons they have on evolvingwild and with 6000 minutes min toi here are some tidbits:

Ellis rates as best d, maata is at 13, Keith and Doughty didn't even crack top60.

Among forwards it's Datsyuk, McJesus, Stone, Crosby and Hornqvist? as top5.
G is at 40, 4th liner in Marcus Foligno is at 91 ahead of Voracek. Erik Staal is 161 just below Darren Helm.

So pretty much mostly worthless.
 

FlyerNutter

In the forest, a man learns what it means to live
Jun 22, 2018
12,366
28,238
Winnipeg
The flyers appear to be very underrated this year. I think that’s a good thing. I think the major thing people aren’t taking into account is the really bad situation of having a horrible coach the firing the GM then the coach and replacing him with an interim level guy + 8 goalies. Those two things will throw literally any team off. Then people tend to think hayes’ contract amount will cause the team to not do well. I think you positive steps forward from Konecny, Hayes, Patrick, Lindblom, Provorov, Sanhiem, Ghost, Myers. Plus Frost. That’s a lot for a media source to assume, but I think we all know that most and more than likely all those players will be better than last year. That’s 8-9 guys.

I’m as negative about this franchise as it gets, but I’ve got hope about this year. I think the kids will rebound or grow, and the vets - while another year older will be fired up to actually be on a club that has a chance to compete.

I won’t pretend to know the mindset of a pro athlete, but I don’t think it was easy to get motivated the last few years.

I’m looking forward to an exciting and competitive club. It’s been a while. Let’s get it on.
 
  • Like
Reactions: TheKingPin

Beef Invictus

Revolutionary Positivity
Dec 21, 2009
127,494
164,354
Armored Train
I remember reading on one of those WAR models and it had insane amount of variables like faceoff%, hits blocks, shots, giveaways, takeaways, some relative metrics and even those xGF metrics. From what I remember it pretty much had everything that is tracked sans corsi or fenwick and it also depended on actual on ice scoring.

After looking at average over the whole 11 seasons they have on evolvingwild and with 6000 minutes min toi here are some tidbits:

Ellis rates as best d, maata is at 13, Keith and Doughty didn't even crack top60.

Among forwards it's Datsyuk, McJesus, Stone, Crosby and Hornqvist? as top5.
G is at 40, 4th liner in Marcus Foligno is at 91 ahead of Voracek. Erik Staal is 161 just below Darren Helm.

That's a sample size that becomes so large it's useless for the present, too. Someone who is on the decline would get floated higher than they should be by earlier seasons.
 

Ghosts Beer

I saw Goody Fletcher with the Devil!
Feb 10, 2014
22,590
16,402
I remember reading on one of those WAR models and it had insane amount of variables like faceoff%, hits blocks, shots, giveaways, takeaways, some relative metrics and even those xGF metrics. From what I remember it pretty much had everything that is tracked sans corsi or fenwick and it also depended on actual on ice scoring.

After looking at average over the whole 11 seasons they have on evolvingwild and with 6000 minutes min toi here are some tidbits:

Ellis rates as best d, maata is at 13, Keith and Doughty didn't even crack top60.

Among forwards it's Datsyuk, McJesus, Stone, Crosby and Hornqvist? as top5.
G is at 40, 4th liner in Marcus Foligno is at 91 ahead of Voracek. Erik Staal is 161 just below Darren Helm.
I recall reading a description of hockey WAR previously & shot differentials were a big part of it, so....Corsi is a major element. Unsurprisingly, since it’s so beloved among the advanced stat crowd.
 
Status
Not open for further replies.

Ad

Upcoming events

Ad

Ad

-->