Player Discussion: Connor Hellebuyck

[AlphaLackey]

Helly still at .909, and when does he get good again?

Starting December 1st, when he's gone .919 in his last ten starts. And since the Rule of Brossoit says that "last ten starts is a sufficient sample size", therefore he's good again.

 

[AlphaLackey]

Last night I actually said Helley was a lone bright spot for a quality start. But if you take those extra 5 or 6 goals that a 915 or so save % gives you and apply it to a couple of his bad nights, we may be 1st in league.

Even if you had the power to add those 5.2 extra saves anywhere you wanted, we would not even be close to first in the league. Each goal has a maximum impact of 1 standings point, right? One goal turns a lead into a tie (1 point difference) or a tie into a deficit (1 point difference). At most, you would get, well, 5 extra points out of 5 exactly placed extra saves.

In reality, you cannot declare where those goals go, goalies don't have the ability to decide when variance will go their way exactly as a poker player doesn't have the ability to decide when they will fill in a flush draw. So, those 5.2 extra saves are worth less than 2 points in the standings. Call it 2 and we move up from 4th to 3rd; call it 1 and we're still in 4th on tiebreak.

 

[White Out 902]

Even if you had the power to add those 5.2 extra saves anywhere you wanted, we would not even be close to first in the league. Each goal has a maximum impact of 1 standings point, right? One goal turns a lead into a tie (1 point difference) or a tie into a deficit (1 point difference). At most, you would get, well, 5 extra points out of 5 exactly placed extra saves.

In reality, you cannot declare where those goals go, goalies don't have the ability to decide when variance will go their way exactly as a poker player doesn't have the ability to decide when they will fill in a flush draw. So, those 5.2 extra saves are worth less than 2 points in the standings. Call it 2 and we move up from 4th to 3rd; call it 1 and we're still in 4th on tiebreak.

This is such a flimsy argument. You're setting arbitrary rules on how we can magically assign saves into the past. You do know how ridiculous that sounds right?

 

[tbcwpg]

This is such a flimsy argument. You're setting arbitrary rules on how we can magically assign saves into the past. You do know how ridiculous that sounds right?

So your argument that "If Hellebuyck allowed 5 or 6 less goals in his bad starts", where you've assigned the goals to specific games is sound logic, but his argument is silly for setting the same arbitrary rules on deciding where those extra five or six saves go?

 

[KingBogo]

Even if you had the power to add those 5.2 extra saves anywhere you wanted, we would not even be close to first in the league. Each goal has a maximum impact of 1 standings point, right? One goal turns a lead into a tie (1 point difference) or a tie into a deficit (1 point difference). At most, you would get, well, 5 extra points out of 5 exactly placed extra saves.

In reality, you cannot declare where those goals go, goalies don't have the ability to decide when variance will go their way exactly as a poker player doesn't have the ability to decide when they will fill in a flush draw. So, those 5.2 extra saves are worth less than 2 points in the standings. Call it 2 and we move up from 4th to 3rd; call it 1 and we're still in 4th on tiebreak.

Interesting post Alpha. A few times now you have put this lens on goaltending. Are you applying your own knowledge of randomness/variance/probability to goaltending or is there work out there on this?

 

[White Out 902]

So your argument that "If Hellebuyck allowed 5 or 6 less goals in his bad starts", where you've assigned the goals to specific games is sound logic, but his argument is silly for setting the same arbitrary rules on deciding where those extra five or six saves go?

I'm saying I prefer my logic. He's saying that his is the only way it could possibly happen. Keep up.

 

[tbcwpg]

I'm saying I prefer my logic. He's saying that his is the only way it could possibly happen. Keep up.

You're not saying you prefer your logic, you're saying that his is only possible in some magical fantasy land. You've also used the word "ridiculous".

Either youre backtracking or prone to extreme hyperbole. Let me phrase the question this way: If Hellebuyck had made those 6 saves, and the team had the exact same record, would you be pleased with him?

 

[White Out 902]

You're not saying you prefer your logic, you're saying that his is only possible in some magical fantasy land. You've also used the word "ridiculous".

Either youre backtracking or prone to extreme hyperbole. Let me phrase the question this way: If Hellebuyck had made those 6 saves, and the team had the exact same record, would you be pleased with him?

Sure. Why not? Odds are that the 6 saves, if truly randomly applied, would likely give us as many as 6 to 8 points, perhaps as few none or 2. But the point is that he wouldn't be a glaring question mark would he?

 

[tbcwpg]

Sure. Why not? Odds are that the 6 saves, if truly randomly applied, would likely give us as many as 6 to 8 points, perhaps as few none or 2. But the point is that he wouldn't be a glaring question mark would he?

I'm sure we'd be pointing to other things as to why they aren't first in the league. I also don't think he's the glaring question mark you're saying he is. But then again, you've been on this teams goaltending for years so I can appreciate your hesitation with accepting that he's improving.

 

[White Out 902]

I'm sure we'd be pointing to other things as to why they aren't first in the league. I also don't think he's the glaring question mark you're saying he is. But then again, you've been on this teams goaltending for years so I can appreciate your hesitation with accepting that he's improving.

I'm happy with where we are in the standings. If we were 2nd place with Helley producing 915 save % I'd be perfectly content. My goal is to win a Stanley Cup.

 

[KingBogo]

Sure. Why not? Odds are that the 6 saves, if truly randomly applied, would likely give us as many as 6 to 8 points, perhaps as few none or 2. But the point is that he wouldn't be a glaring question mark would he?

Actually Alphalackey gave you the likely impact on our points total if the goals were randomly distributed you just didn't like the answer.

 

[White Out 902]

Actually Alphalackey gave you the likely impact on our points total if the goals were randomly distributed you just didn't like the answer.

His answer was wrong. The truth is we have no idea where those goals would go, if randomly placed.
edit: What is almost surely true though, is his assertion the saves would be evenly spread out is wrong. Random is never perfectly spread out like that, so my answer is closer to a "truth" than his. Someone just posted a page or two ago what a random chart looks like vs one to made look random... see that for reference.

 

[KingBogo]

His answer was wrong. The truth is we have no idea where those goals would go, if randomly placed.
edit: What is almost surely true though, is his assertion the saves would be evenly spread out is wrong. Random is never perfectly spread out like that, so my answer is closer to a "truth" than his. Someone just posted a page or two ago what a random chart looks like vs one to made look random... see that for reference.

You right in that random outcomes could have happened in any game. Alpha was just trying to put the question at hand under the lens of probability theory, which makes it an interesting discussion rather than just choosing the extreme end that you want to believe and say that is the answer. Taking a look back he was probably being generous in his argument as all games that Helly already won it wouldn't have matter if he made one or more of the extra 5.2 saves.

 

[JetsUK]

Helle is 6th in average goalie salary rankings for 2018, according to Spotrac (caveat: not sure how reliable they are as a source). So he's getting paid like an elite goaltender while delivering league-average goaltending -- so far. I think it's fair to argue that he has delivered fewer of the standout performances we got used to last season, and more of the deflating goals we saw in 2016-17. What the impact of his play so far this season is on the team and its record, and whether it represents an outlier or even standard variation in a generally solid-to-sparkling career isn't yet clear.

As someone who does not grasp the technical intricacies of NHL goaltending, I'd be very interested to hear from those who do on Helle's play so far this season: what is he doing differently? Have the new equipment rules forced him to adapt his play significantly? Is the team playing differently behind him? Have opposing teams "figured him out? Is there a correlation between "soft goals" and the tendency to overplay or alter play in order to prevent more of them? Are there notable differences (excluding level of opposition) between Croissant's play and Helle's that might explain their differing results? I feel like I learned a lot about goaltending here from members who were/are goalies at a high level, but haven't seen much of that sort of discussion recently.

 

[tbcwpg]

His answer was wrong. The truth is we have no idea where those goals would go, if randomly placed.
edit: What is almost surely true though, is his assertion the saves would be evenly spread out is wrong. Random is never perfectly spread out like that, so my answer is closer to a "truth" than his. Someone just posted a page or two ago what a random chart looks like vs one to made look random... see that for reference.

Yours isn't closer to truth. You're positing that it's more likely that those 6 saves would be in one goal games in which they lost points instead of games where that save would've made a 4-1 loss into a 3-1 loss, or a 6-4 win into a 6-2 win. Those odds are lesser than those goals coming in games where an extra save wouldn't have mattered.

 

[White Out 902]

Yours isn't closer to truth. You're positing that it's more likely that those 6 saves would be in one goal games in which they lost points instead of games where that save would've made a 4-1 loss into a 3-1 loss, or a 6-4 win into a 6-2 win. Those odds are lesser than those goals coming in games where an extra save wouldn't have mattered.

I didn't take that position. I posited no such thing.

 

[AlphaLackey]

This is such a flimsy argument. You're setting arbitrary rules on how we can magically assign saves into the past. You do know how ridiculous that sounds right?

Isn’t that exactly what’s being done by your attempt to set arbitrarily rules on how to assign saves in the past? My response is “I could easily have the same power you claim for yourself, and there is as yet no sound reason why you should have the power and I don’t.”

The other part (the maximum possible gain from five goals is five standings points) isn’t an arbitrary rule, it’s an explanation of what is or isn’t possible.

 

[AlphaLackey]

Interesting post Alpha. A few times now you have put this lens on goaltending. Are you applying your own knowledge of randomness/variance/probability to goaltending or is there work out there on this?

The prevailing work is well established but can be verified by plotting each team’s goal differential (for versus against) against each team’s point total. The line of best fit shows that one extra goal correlates to about one third of a standings point.

My own work (as seen in the 2017 Hockey Abstract) showed that only about half of goals change the game state; further work would be needed to see how often that work isn’t undone by a later goal but my best guesses suggest this would be a strong corroborating method.

But the linear regression is rock solid and holds year in and year out so it’s solid. Bucky has probably cost us 2-3 standings points and Brossoit has won them back.

 

[tbcwpg]

I didn't take that position. I posited no such thing.

Let's summarize:

-You suggest the team would be, or at least could be, first in the league if Hellebuyck made about 6 extra saves.
-Poster says that you can't actually suggest that, because it's just as likely, if not more likely, that those saves wouldn't result in anything more than maybe a one point difference.
-You suggest that argument is flimsy and ridiculous, and further suggest that it's "likely" worth 6 to 8 points, and "perhaps" worth 0-2.
-I apply this logic to your scenario, in that for the team to actually get those extra points, those saves would have to come in one goal games to influence the result.
-You say you've suggested no such thing.

You actually have suggested that, not in those words, but by your theory. You have to be, otherwise your posts make no sense.

 

[AlphaLackey]

His answer was wrong. The truth is we have no idea where those goals would go, if randomly placed.
edit: What is almost surely true though, is his assertion the saves would be evenly spread out is wrong. Random is never perfectly spread out like that, so my answer is closer to a "truth" than his. Someone just posted a page or two ago what a random chart looks like vs one to made look random... see that for reference.

I didn’t say they’d be evenly spread out. I said they’d be randomly placed. And if something that has a calculable value is randomly placed, we can calculate the average value.

If I roll a thousand dice and add up the totals, and then you say you want to roll five more dice, you’re right, I don’t know what those five extra dice will be. I know the average value will be 17.5;

The average value of a six sided dice is 3.5 and the average value of an extra save is 1/3rd if a standings point; the average value of five six sided dice is 17.5 and the average value of five extra saves is 1 and 2/3rds of a standings point. Not five and certainly not 12.

 

[KingBogo]

The prevailing work is well established but can be verified by plotting each team’s goal differential (for versus against) against each team’s point total. The line of best fit shows that one extra goal correlates to about one third of a standings point.

My own work (as seen in the 2017 Hockey Abstract) showed that only about half of goals change the game state; further work would be needed to see how often that work isn’t undone by a later goal but my best guesses suggest this would be a strong corroborating method.

But the linear regression is rock solid and holds year in and year out so it’s solid. Bucky has probably cost us 2-3 standings points and Brossoit has won them back.

Interesting stuff. I hope you keep posting on this topic.

 

[White Out 902]

His 906 tonight keeps Helley at 908 for the year. Woof.

 

[Adam da bomb]

His 906 tonight keeps Helley at 908 for the year. Woof.

That's unfair Jets were horribly defensively but sure blame the goalie.

 

[Flair Hay]

He was great tonight I dont care if he had a .906. Like, is the idea here that he sucks? I wouldn't blame him for one of the goals that went in tonight...

 

[White Out 902]

He was great tonight I dont care if he had a .906. Like, is the idea here that he sucks? I wouldn't blame him for one of the goals that went in tonight...

The idea is that he has collapsed in save % from 924 to 908. That's all.