News Article: The Athletic on Detroit's "Contract Efficiency"

[InjuredChoker]

Where does he rank if you eliminate goalies that played under 40 games?

Howard himself didn't play 40 games in 15-16 or 16-17.

 

[InjuredChoker]

from 14 to 19, howie's sv% rank among goalies.

tied for 37th with 3 other goalies of 55. minimum 100 games played.

tied for 32 with 2 other goalies out of 41. minimum 150 games played. funny enough, those two goalies are bernier and mrazek.

and tied with those two for 25th out of 31 goalies if the minimum gp cutoff is 200.

 

[ricky0034]

If Wings were on a rocket ship to compete for a Stanley Cup in the next 3 or 4 years, then yeah those contracts would really sting.

Since we're rebuilding and need to lose for Hughes Lafreniere, those contracts don't sting. They help.

help? only under the assumption that the GM is so incompetent that he'll inevitably use the money to do something else stupid

there's no other way that they could "help"

 

[WingedWheel1987]

Not surprising. The Gator contract alone would put Detroit in the bottom third of the rankings. It really is that bad.

 

[Syckle78]

Sure during that time frame was he EVER awarded the end of the year All Star team? First or second team?

So only two good goalies a year. Lol

 

[Hockeyville19Calumet]

So only two good goalies a year. Lol

Dude I didn't bring up the All Star stuff.

 

[kliq]

Howard himself didn't play 40 games in 15-16 or 16-17.

Ya I know, its tough to even count those seasons, good or bad.

 

[lidstromiscool]

Glendening at 1.8 is a great deal. It's almost like stats don't always tell the whole story

 

[Henkka]

Glendening at 1.8 is a great deal. It's almost like stats don't always tell the whole story

I would many many cases against that statistical methodology. It's not a God's word.

It just one Dom Luszczyszyn's method with it's own flaws. Especially it won't work with defencemen. Maybe a good method of analyzing forwards.

 

[Bench]

Glendening at 1.8 is a great deal. It's almost like stats don't always tell the whole story

Trying to moneyball hockey has always been difficult. The inability to isolate players and the rapid shift from offense to defense means there's always going to be holes in your data.

Brendan Smith was an advanced stats darling that Mike Babcock was holding back from greatness. Or maybe ol' Mike saw what the fancy stats weren't showing. Important lesson when it comes to how much emphasis we put on the raw numbers.

 

[BinCookin]

Not big names but we might not have traded Tatar and or Nyquist if not for the Abby/Helm/Franz signings. Yes the Tatar deal was great for us but I'd like to think if this team subtracted the 3 players above and added a real 2C to replace Franz along with keeping Tatar and Nyquist that this team could have been a playoff team. Then again, the defense would prob still been in shambles so prob not lol.

Not true. Tatar and Nyquist were traded because we were rebuilding, and they were our most valuable assets in regards to playoff teams (27-28 year olds in their prime).

Average save % in the NHL by year:

14-15 - .915%
15-16 - .915%
16-17 - .913%
17-18 - .912%
18-19 - .910%

So that’s 4 of the last 5 years he was below average in save percentage. He plays in front of a trash defense, but I don’t see him holding up well in a stats-based argument using the last 5 years.

To me Save Pct is the true goalie stat. GAA is more of a team stat. IMO

I infact think ALL goalie stats are team based.
I bet there would be a very strong correlation towards the order teams finished and those goaltending stats. As we know Detroit has been one of the worst teams in the league for 3 years running. Howard having numbers below average isn't just expected, its almost garaunteed. Only Luongo turned in amazing stats on a junker team. I mean its a really short list of goalies who can do that (im guessing). In either case Howard is the source of our problems. Anyone think we have a good team with Bobrovsky in net right now?

Trying to moneyball hockey has always been difficult. The inability to isolate players and the rapid shift from offense to defense means there's always going to be holes in your data.

Brendan Smith was an advanced stats darling that Mike Babcock was holding back from greatness. Or maybe ol' Mike saw what the fancy stats weren't showing. Important lesson when it comes to how much emphasis we put on the raw numbers.

You guys remember my Defensive analysis Eye-ball test scoring system 3 years or so ago?
This was around the time Smith's "advanced stats" were just glowing. And yet I rated him as our worst defenseman by far.
It made absolutely no sense as I was counting good and bad plays by defensemen, and he was always skewed towards the bottom of all our D men. (And its not like our D core is strong).

I mean I really have come to despise the CF% etc stats, as I do not think they measure what they claim to measure. Sure possession is probably correlated with how many shots you take.... But technically you are not measuring possession, you are only measuring shots. Its just an "advanced shot" statistic. Evander Kane should be one of the highest possession players in the league based on CF% because he shoots a lot! But all that stat tells me is... he shoots alot (I already know this from the SOG stat).

Hockey is a complicated game in regards to stats vs baseball.

But for the most part how do you determine what is a "high danger shot" or a "good chance of scoring shot" vs a "garbage perimeter shot". You can take garbage shots from the slot. Its just very hard to count the things that you really want to.. or at least no public site I know of is publishing these counts. Zone entries and exits would be a good one for D men.

 

[Hammettf2b]

Not true. Tatar and Nyquist were traded because we were rebuilding, and they were our most valuable assets in regards to playoff teams (27-28 year olds in their prime).

I don't think you understood what I was trying to say. My post was just a theory if Helm, Abby, Nielsen weren't signed and replaced with actual good players keeping us playoff contenders. At that point Nyquist and Tatar prob don't get traded for rebuilding pieces. Again, everything I said is hypothetical.

 

[BinCookin]

^^ Ya i wasn't harshing, just analyzing.

But I do think we were going to be bad no matter whether we had those deals or not. Bottom line is its just really hard to gain top end talent drafting so high forever. I infact think we have done a solid drafting job for all the time everyone says we have not. The problem is, we always have lower picks than other teams. Maybe my analysis was flawed based on position. But in a few years, ill have some more data with us near the bottom. And then I can really see if the system works well.

 

[DatsyukToZetterberg]

Average save % in the NHL by year:

14-15 - .915%
15-16 - .915%
16-17 - .913%
17-18 - .912%
18-19 - .910%

So that’s 4 of the last 5 years he was below average in save percentage. He plays in front of a trash defense, but I don’t see him holding up well in a stats-based argument using the last 5 years.

I don't think using sv% is the best way to evaluate a goalie and it can paint a picture that isn't totally accurate. Howard's GSAA, Goals Saved Above Average, has been anywhere from below average to above average over that time period.

14-15 - -8.13
15-16 - -3.64
16-17 - 6.9
17-18 - 2.9
18-19 - 0.8

I think at this point Howard is just your typical average starting goalie. He'll give you 50-60 starts a year in which some nights he'll steal the show and others he won't be able to stop a beach ball. I will say that evolving-hockey's WAR model thinks he was an average or above average goalie the past 2 seasons.

 

[NotLeddy]

I don't think using sv% is the best way to evaluate a goalie and it can paint a picture that isn't totally accurate. Howard's GSAA, Goals Saved Above Average, has been anywhere from below average to above average over that time period.

14-15 - -8.13
15-16 - -3.64
16-17 - 6.9
17-18 - 2.9
18-19 - 0.8

I think at this point Howard is just your typical average starting goalie. He'll give you 50-60 starts a year in which some nights he'll steal the show and others he won't be able to stop a beach ball. I will say that evolving-hockey's WAR model thinks he was an average or above average goalie the past 2 seasons.

Evolving-Hockey model's isn't at 5v5 is it? That's very interesting. Corsica has Howard's GSSA as 7.89 this season and -2.65 in 2017-18. NaturalStatTrick has 7.09 and -1.58 the last two seasons, 2018-19 and 2017-18 respectively. If I recall though, Corsica has had some criticism for undervaluing the slot and overvaluing the crease in their model.

Also interesting is that the majority of the season, on NST's model, Howard consistently had the 3rd best GSAA in the league (up until the stretch from February 16th-26th, where he had by far the worst stretch of his season).

I'm very surprised their model says Howard was worse last season than the one before.

 

[MBH]

Trying to moneyball hockey has always been difficult. The inability to isolate players and the rapid shift from offense to defense means there's always going to be holes in your data.

Brendan Smith was an advanced stats darling that Mike Babcock was holding back from greatness. Or maybe ol' Mike saw what the fancy stats weren't showing. Important lesson when it comes to how much emphasis we put on the raw numbers.

Not that Brendan Smith ever did much to prove himself, but it's pretty hard to defend the way Babcock utilized him.
Smith's TOI
PP: 148
SH: 359
Not how I'd develop a swift-skating, puck moving, offensive defenseman.

As for Jimmy Howard. He's a fine goalie in October and November.
But without fail, he gets worse as the season goes on.
That might be why fans overrate him. He starts strong early and is often the only thing keeping Detroit from falling into the cellar.
But after about eight weeks, he starts sliding a bit.
Then he wheels come off in February and March. And he ends up with the same average to mediocre stats he posts every season.

Perhaps Yzerman would do well to shop him in November instead of waiting until February.

 

[Lazlo Hollyfeld]

Trying to moneyball hockey has always been difficult. The inability to isolate players and the rapid shift from offense to defense means there's always going to be holes in your data.

Brendan Smith was an advanced stats darling that Mike Babcock was holding back from greatness. Or maybe ol' Mike saw what the fancy stats weren't showing. Important lesson when it comes to how much emphasis we put on the raw numbers.

I swear the Ranges must've based Smith's contract a couple years ago solely on fancy stats.

My other working theory is that they were drunk.

 

[Sadekuuro]

I swear the Ranges must've based Smith's contract a couple years ago solely on fancy stats.

My other working theory is that they were drunk.

My brother is a Rangers fan and he said their fanbase also definitely thought they were getting something they weren't, not just the front office.

I laughed hysterically at that deal and told him what to expect though

 

[avssuc]

One of the worst managed teams in the league for some time. Glad to have Holland out in that regard, that’s for sure.

I think the DeKeyser deal (or Nielsen, whichever came last) was the last bad one to pin on Holland. He did some decent work trying to clean up his mess. And it's been stated many times that his direction was probably dictated from Mr. I... so let's be fair.

As it looks right now, he robbed Calgary in the Lucic deal.

 

[Winger98]

I swear the Ranges must've based Smith's contract a couple years ago solely on fancy stats.

My other working theory is that they were drunk.

He looked really good after the trade in their playoff run. Just looking at his numbers, and I wonder if NYR tried to use him differently the following seasons, and it just hasn't worked at all.

 

[Lazlo Hollyfeld]

He looked really good after the trade in their playoff run. Just looking at his numbers, and I wonder if NYR tried to use him differently the following seasons, and it just hasn't worked at all.

I remember he had a really good stint with them at the end of the season. But it's seriously like they did no other research on him or something.

 

[ShelbyZ]

As for Jimmy Howard. He's a fine goalie in October and November.
But without fail, he gets worse as the season goes on.
That might be why fans overrate him. He starts strong early and is often the only thing keeping Detroit from falling into the cellar.
But after about eight weeks, he starts sliding a bit.
Then he wheels come off in February and March. And he ends up with the same average to mediocre stats he posts every season.

Perhaps Yzerman would do well to shop him in November instead of waiting until February.

IMO, for the past 4 years Howards seasons have consistently been more of a backwards "J" shape... He starts red hot for the first 7 to 10 or so weeks then either gets hurt and struggles to get going or struggles mightily without being injured, then he somehow picks it up enough to be "average to good" (somewhere right in the middle between his red hot start and the worst of his mid season slump like where the shorter part of a J is...) from around the trade deadline to the end of the season:

15-16: Starts great, then puts up two stinkers in a row around Christmas. Goes on a lengthy losing streak (getting only the 2nd on back to back games). Then around the TDL when Mrazek starts shitting the bed hard, Jimmy puts up that huge game against the Rangers and finishes the season 7-5-1 and .91 from 2/21 on.

16-17: Is tremendously good (.934), until Nick Jensen falls on him in a game right before Christmas, causing him to miss almost 3 months. Comes back and platoons with Mrazek for the last month of the year and is 5-4-0 and .916.

17-18: Starts out hot, but then really takes a tumble from late November to early Dec. Puts up a good 3-4 weeks in December/January, then tumbles again until around that odd 2/21 mark and finishes 6-8-3 and .909.

18-19: Started hot again, enough so to get named to the all start game, then starts to tumble again in late November. From then on he's up and down until that 8-1 embarasment against the Habs. After that game (on 2/26), he goes 6-5-0 and .913 for the rest of the season.

You also have to keep in mind that when he started to rebound a bit the past 3 seasons, he's playing behind a team that's shutting good players down and cycling guys up from GR and in some aspects just mailing it in.



As far as his contract went, he was probably going to get that deal in the summer of 2013 from a lot of different teams. In 2 seasons (well really 1.5ish) he was at or above .92, had 11 shutouts and had carried the team on his back in 2013 when Babcock had no faith in Glasstavsson. He started 13 games in 27 days and went 7-3-3 and .942 and then was excellent in that almost upset of eventual Cup winner Chicago. He definitely earned the contract, he just failed to live up to it. In some ways he's basically Osgood 2.0, except the team trended opposite directions during their respective prime.

 

[Retire91]

I remember that smith deal I think every redwings fan did a happy dance that day

 

[Hockeyville19Calumet]

Evolving-Hockey model's isn't at 5v5 is it? That's very interesting. Corsica has Howard's GSSA as 7.89 this season and -2.65 in 2017-18. NaturalStatTrick has 7.09 and -1.58 the last two seasons, 2018-19 and 2017-18 respectively. If I recall though, Corsica has had some criticism for undervaluing the slot and overvaluing the crease in their model.

Also interesting is that the majority of the season, on NST's model, Howard consistently had the 3rd best GSAA in the league (up until the stretch from February 16th-26th, where he had by far the worst stretch of his season).

I'm very surprised their model says Howard was worse last season than the one before.

Howard has been the modern day Cheveldae.

 

[Hockeyville19Calumet]

IMO, for the past 4 years Howards seasons have consistently been more of a backwards "J" shape... He starts red hot for the first 7 to 10 or so weeks then either gets hurt and struggles to get going or struggles mightily without being injured, then he somehow picks it up enough to be "average to good" (somewhere right in the middle between his red hot start and the worst of his mid season slump like where the shorter part of a J is...) from around the trade deadline to the end of the season:

15-16: Starts great, then puts up two stinkers in a row around Christmas. Goes on a lengthy losing streak (getting only the 2nd on back to back games). Then around the TDL when Mrazek starts ****ting the bed hard, Jimmy puts up that huge game against the Rangers and finishes the season 7-5-1 and .91 from 2/21 on.

16-17: Is tremendously good (.934), until Nick Jensen falls on him in a game right before Christmas, causing him to miss almost 3 months. Comes back and platoons with Mrazek for the last month of the year and is 5-4-0 and .916.

17-18: Starts out hot, but then really takes a tumble from late November to early Dec. Puts up a good 3-4 weeks in December/January, then tumbles again until around that odd 2/21 mark and finishes 6-8-3 and .909.

18-19: Started hot again, enough so to get named to the all start game, then starts to tumble again in late November. From then on he's up and down until that 8-1 embarasment against the Habs. After that game (on 2/26), he goes 6-5-0 and .913 for the rest of the season.

You also have to keep in mind that when he started to rebound a bit the past 3 seasons, he's playing behind a team that's shutting good players down and cycling guys up from GR and in some aspects just mailing it in.



As far as his contract went, he was probably going to get that deal in the summer of 2013 from a lot of different teams. In 2 seasons (well really 1.5ish) he was at or above .92, had 11 shutouts and had carried the team on his back in 2013 when Babcock had no faith in Glasstavsson. He started 13 games in 27 days and went 7-3-3 and .942 and then was excellent in that almost upset of eventual Cup winner Chicago. He definitely earned the contract, he just failed to live up to it. In some ways he's basically Osgood 2.0, except the team trended opposite directions during their respective prime.

You just described a very inconsistent player. Also Osgood was NEVER this inconsistent or this injured/often.