GBU #5 Sabres-Caps: Concepts and Principles

[Snippit]

I mean, there’s no such thing as a guarantee in statistics, that’s why we produce confidence intervals and look at values like r-squared in reference to linear regressions. So yeah, of course you could have a 30 game sample that doesn’t resemble the population, but the odds of this happening are excessively low and it kind of just seems like you’re arguing in bad faith here

Maybe I’m jumping into this prematurely but are the odds of an unrepresentative 30 game sample really as excessively low as you are making it out to be? Skinner last year for example?

This is hockey where players go through streaks and slumps, not rolling dice.

 

[valet]

Internet rhetorical memes aside, citing stat 102 terms doesn’t make you right, op. You’ve got things jumbled

Tell me how I have things jumbled then. You haven’t made an actual argument yet

 

[enthusiast]

I’m in a nasty mood and that’s not conducive to discussion, apologies

e: I have, you’ve seen it and responded to it. If you want to treat this as concession go ahead

 

[valet]

Maybe I’m jumping into this prematurely but are the odds of an unrepresentative 30 game sample really as excessively low as you are making it out to be? Skinner last year for example?

This is hockey where players go through streaks and slumps, not rolling dice.

Skinner last year is a bad example, really. When you consider his usage his numbers weren’t exactly unrepresentative. Near career lows in shots and shot attempts due to injury and deployment, and yeah, career low shooting %, all leading to less overall goal production, but the fancy stats had him near tops in the league in certain categories (5v5 xGF/60). To answer the question, his low shooting % definitely falls in the range of excessively unlikely

 

[Snippit]

Skinner last year is a bad example, really. When you consider his usage his numbers weren’t exactly unrepresentative. Near career lows in shots and shot attempts due to injury and deployment, and yeah, career low shooting %, all leading to less overall goal production, but the fancy stats had him near tops in the league in certain categories (5v5 xGF/60). To answer the question, his low shooting % definitely falls in the range of excessively unlikely

Except Skinner has crazy up and down swings every other season.

 

[valet]

Except Skinner has crazy up and down swings every other season.

The only thing that could be said to fluctuate is skinners sh%. last season it was unusually low, even within the realm of his so-called ‘crazy swings’. What I’m saying is that the fluctuations in skinners production aren’t as ‘fluctuation-y’ as people think, that this is an overstated narrative that relies on cherry picking stats born out of completely different contexts

 

[SundherDome]

I’m in a nasty mood and that’s not conducive to discussion, apologies

e: I have, you’ve seen it and responded to it. If you want to treat this as concession go ahead

 

[buffalowing88]

considering on paper we have the best team with where our players are development wise easily in over a decade and have played two cup contenders and it will remain that way through six games and only these two cup contenders, yet somehow managed to be in every single game until the final minute, I don’t get this attitude...yet. Talk to me in a few weeks. Don’t give up already

I appreciate your optimism. And willingness to present a reasonable case.

But I'm done. I am going full NBA at this point. Rather root for a team that travels without being called than a team that can't even get the puck across mid-ice. We're talented, sure, but we're also as inept as ever before. The tank will continue until Buffalo brings a McDermott-type in. Our coach isn't that guy.

 

[jc17]

Why are we stuck on the sample size being number of games?

The number of shots are what matters. And there's no number that guarantees regression to the mean, but the larger the sample size the greater the chance it will be around the mean.

Eichel could shoot at 4% on the season, the math doesn't say it's impossible, just very unlikely.

 

[joshjull]

Yeah for sure, just that 55 games or whatever is more than enough to have an “unlucky” streak

Definitely. Plenty of players “underperform” or “outperform” their underlying numbers over a 82 games season.

But players like Skinner, Jack and Hall never scoring 5v5 over 55gms would go way beyond being “unlucky” into absurdity.

 

[enthusiast]

Definitely. Plenty of players “underperform” or “outperform” their underlying numbers over a 82 games season.

But players like Skinner, Jack and Hall never scoring 5v5 over 55gms would go way beyond being “unlucky” into absurdity.

I agree that a zero goal season would be a bit of an eye opener



with respect to yesterday’s disagreement it’s difficult to speak on stats even when one has hours and hours of study. I assume that for valet’s word choice he’s got that familiarity but his frame of reference isn’t appropriate to the issue at hand and I’m not smart enough to explain it such that it’s evident despite having performed well in that area for a while at a graduate level

statistical education at most accessible is as much about instilling a proper intuition into someone as anything, but that doesn’t mean one can really be literate enough to dissect, word choice aside

here, fundamentally, it’s inappropriate to leverage past data as predictive in the way being asserted given the myriad assumptions one has to make, which is exacerbated by the truncated season

I realize that might not be satisfying as an explanation for any peanut gallery but we all have our limitations

 

[joshjull]

I agree that a zero goal season would be a bit of an eye opener



with respect to yesterday’s disagreement it’s difficult to speak on stats even when one has hours and hours of study. I assume that for valet’s word choice he’s got that familiarity but his frame of reference isn’t appropriate to the issue at hand and I’m not smart enough to explain it such that it’s evident despite having performed well in that area for a while at a graduate level

statistical education at most accessible is as much about instilling a proper intuition into someone as anything, but that doesn’t mean one can really be literate enough to dissect, word choice aside

here, fundamentally, it’s inappropriate to leverage past data as predictive in the way being asserted given the myriad assumptions one has to make, which is exacerbated by the truncated season

I realize that might not be satisfying as an explanation for any peanut gallery but we all have our limitations

I wasn’t thinking much about statistical concepts like regression when I posted that tweet. I just thought it was a succinct way of saying if we keep playing well systemically the goals should come. Especially since the guys not scoring are our best players. Then the conversation went into a heavy focus on regression itself.


I have to blame myself for even posting that tweet.

 

[enthusiast]

I wasn’t thinking much about statistical concepts like regression when I posted that tweet. I just thought it was a succinct way of saying if we keep playing well systemically the goals should come. Especially since the guys not scoring are our best players. Then the conversation went into a heavy focus on regression itself.


I have to blame myself for even posting that tweet.

Yeah I’m just trying to get out what I couldn’t yesterday

online arguments never die...

 

[Washed Up 29YearOld]

accidental post carry on

 

[Tage2Tuch]

Gladly

Game one with 1:03 left on the clock they were down by two. (Not in it)
Game two with 1:00 left on the clock they were down by one (in it)
Game three with 1:11 left on the clock they were up by 5 (in it)
Game four with 1:42 left on the clock, they were down by 3 (not in it)
Game five with 1:00 left on the clock they were tied (in it)

So by the fact based assessments stated above, the Sabres, were in fact, NOT in every game in the last minute.

Pretty sad that you have to take what I said literal meaning 60 seconds to the point.

For instance in your first example you say they weren’t in the game because with 63 seconds left the other team got an empty netter. First of all being down by two is still somewhat in it, this is why teams pull their goalies almost always when down by two with a minute left but my point was around the minute mark (this could be with one second to three mins left) as the two endpoints ...I didn’t mean with 60 seconds left exactly, come on now...So your first example is off by three seconds even if you thought I meant one minute exactly? That’s reaching man.


My point was we could of won any of the games we’ve played this year and the only game we played bad was opening night. Which after 300 days off and no preseason while the other team played just a few months ago playoff games is understandable.

We are the only team in the league who’s had to play ONLY top two division teams from last year through six games yet sit 2-3-1 and like I said have been technically in all of them.

 

[Tage2Tuch]

I appreciate your optimism. And willingness to present a reasonable case.

But I'm done. I am going full NBA at this point. Rather root for a team that travels without being called than a team that can't even get the puck across mid-ice. We're talented, sure, but we're also as inept as ever before. The tank will continue until Buffalo brings a McDermott-type in. Our coach isn't that guy.

Thanks and Cool, I guess you won’t be posting much this season if you’re done, is that what you’re saying? Can’t you give it until we play a team that didn’t finish in the top two teams in their division last year? It’s a little early.

 

[SundherDome]

Pretty sad that you have to take what I said literal meaning 60 seconds to the point.

For instance in your first example you say they weren’t in the game because with 63 seconds left the other team got an empty netter. First of all being down by two is still somewhat in it, this is why teams pull their goalies almost always when down by two with a minute left but my point was around the minute mark (this could be with one second to three mins left) as the two endpoints ...I didn’t mean with 60 seconds left exactly, come on now...So your first example is off by three seconds even if you thought I meant one minute exactly? That’s reaching man.


My point was we could of won any of the games we’ve played this year and the only game we played bad was opening night. Which after 300 days off and no preseason while the other team played just a few months ago playoff games is understandable.

We are the only team in the league who’s had to play ONLY top two division teams from last year through six games yet sit 2-3-1 and like I said have been technically in all of them.

I don't know why people get pissy when they are fact checked about what they said. What you said was wrong, I called you on it, you then continued blindly to challenge my assertion and then you were proven wrong again and then you felt the need to puff up your chest like it matters.

 

[enthusiast]

knock it off

 

[MtlsabresFAN]

First period they sucked and then they got better.. can't complain they won against a very good team.. and wow Mitts lost a lot of face weight

 

[BananaSquad]

Cozens better play next game. I don’t want this keep the same lineup BS cuz they won.

 

[MarkusKetterer]

Cozens better play next game. I don’t want this keep the same lineup BS cuz they won.

Why must he “better play”?

 

[Buffaloed]

If you want a GBU for game #6 make one.