Winnipeg Jets Top 20 Prospects 2017 - #4

[Daximus]

I wrote a long post last night about this but didn't post it. I would also be interested in seein 5v5 sv% numbers for Comrie, as well as high/low/medium danger SV% + shot attempts, rel. to the rest of the AHL.

There's definitely more to the story than pointing to his save percentage and saying he sucks. But that won't stop people from doing it.

 

[Holden Caulfield]

Raw save percentage will soon become as arcane as raw Corsi, I think.

Already a lot of folks are using "adjusted save percentage".

If shot quality is related to shooting percentage, it should be related to save percentage, right?

If there is a difference in terms of shot quality yielded by a team (there is), then that should affect save percentage. Some analyses have shown that save percentage changes by as much as 1% (e.g. from 0.905 to 0.915) when you adjust for shot quality.

I would add that there are other situational dimensions to goalie performance that I don't think have been adequately reflected in statistical methods.

I have no idea whether Comrie's save percentage is affected by shot quality, but the concept is not absurd.

When you start looking at that though your being in a huge variable that is nearly impossible to ensure lacks any bias. How is it determined what is high danger shot, what is a low danger shot? Who determines that? How is it free of bias? Look at how unreliable hits, giveaways, takeaways are because of the unreliability of the person recording the stat. I don't know how you get around that. Do you know?

There's definitely more to the story than pointing to his save percentage and saying he sucks. But that won't stop people from doing it.

And no one has said that. Literally no one. But he needs to improve alot and fast if he wants to remain in the conversation as a potential NHL goalie.

 

[truck]

There's definitely more to the story than pointing to his save percentage and saying he sucks. But that won't stop people from doing it.

Nobody is saying he sucks, just that his numbers suck. They are far worse than those posted by Hellebuyck and Pav on the same team. He's young and talented so he may rebound, but he has not stopped enough pucks in the AHL. Apprehension is fair.

 

[surixon]

Nobody is saying he sucks, just that his numbers suck. They are far worse than those posted by Hellebuyck and Pav on the same team. He's young and talented so he may rebound, but he has not stopped enough pucks in the AHL. Apprehension is fair.

Pavs sample size was pretty darn small tough. It's not like Comrie didn't match those numbers over similarly long stretches last season.

 

[Whileee]

When you start looking at that though your being in a huge variable that is nearly impossible to ensure lacks any bias. How is it determined what is high danger shot, what is a low danger shot? Who determines that? How is it free of bias? Look at how unreliable hits, giveaways, takeaways are because of the unreliability of the person recording the stat. I don't know how you get around that. Do you know?



And no one has said that. Literally no one. But he needs to improve alot and fast if he wants to remain in the conversation as a potential NHL goalie.

Analysts are using shot locations as a proxy because shot location correlates with shooting percentage. There are other variables (like screen shots, odd man chances), but adjusting for shot location is already an improvement over raw save % it would seem.

 

[truck]

Pavs sample size was pretty darn small tough. It's not like Comrie didn't match those numbers over similarly long stretches last season.

For sure. A full season of a starter isn't enough to get a meaningful read, but... there are almost games of "not Comrie" over the last two seasons, spread between Hellebuyck, Pav and Phillips. I suppose you could and Jusso in there too, but he was bad, bad, bad.

 

[ps241]

There's definitely more to the story than pointing to his save percentage and saying he sucks. But that won't stop people from doing it.

I like Comrie and am hoping for him but at some point he needs to produce better results.

 

[surixon]

I like Comrie and am hoping for him but at some point he needs to produce better results.

I don't think anyone disagrees with this. I still maintain that his numbers where do to fatigue and being asked to do too much by a coach who had no faith in Phillips.

Comrie's low save percentage is predominantly driven by a 7 straight game stretch in early March that saw him post an .831 save percentage over that time frame. Kind of looks out of place doesn't it? So the question then becomes why did Comrie go ice cold for such am extended stretch when up until that point he had a much more even keeled distribution between performances.

Well the 20 day stretch that immediately predated that cold spell had Comrie play 10 games with 4 consecutive B2B situations with a 3 in 4 game stretch at the beginning of the start of the B2B stretch. To me that is an absolutely ridiculous work load and imo directly tied into the cold stretch.

It's completely fine to be apprehensive with Comrie but sometimes I think we need to look beyond the overall number and try to find legitimate reasons for unexpected performances. I'd say smart coaching by Vincent would have lead to Phillips being in for two of those b2b situations which would have allowed Comrie to not get played into the ground. There is zero time to recharge physically and mentally when you play a full game every second day for nearly a month.

Edit: we also see this type of defense with other players like Petan and Dano who's box car stats look terrible but then you look deeper and realize that some of it is due to not being utilized properly or being put in a place to really succeed. As Garret posted in another thread about over training proper recovery time is equally if not more so important and id argue that Vincent didn't place Comrie in place to succeed by giving him the time he needed between starts to properly recover during the stretch I mentioned. At the end of the day Comrie still needs to up his consistency but I maintain his year last year was much better than then the end result indicated.

 

[Hammer Slammer]

What do you hear about him? Second question: can you verify what you've heard?

I'm also waiting on this info.

 

[Mortimer Snerd]

When you start looking at that though your being in a huge variable that is nearly impossible to ensure lacks any bias. How is it determined what is high danger shot, what is a low danger shot? Who determines that? How is it free of bias? Look at how unreliable hits, giveaways, takeaways are because of the unreliability of the person recording the stat. I don't know how you get around that. Do you know?



And no one has said that. Literally no one. But he needs to improve alot and fast if he wants to remain in the conversation as a potential NHL goalie.

^This.

I still have high hopes for Comrie but optimism is fading. Compare his sv% last year to Pav's. Same team, same PK in front of both.

Bottom line is that Comrie has not been good enough. The upcoming season will be crucial for his career progress.

 

[Deif]

I'm also waiting on this info.

I think he's referring to my post the other day where I told him that "Berdin was not at development camp, the Jets did not say why, Garret said he was injured, Berdin was posting pictures on instagram on vacation with his girlfriend+friends as dev camp was going on."

Personally, I'm not concerned about it, just would be interested to know what the injury is as he did not seem injured in any of the pictures he posted (not to say he isn't injured, though).