Post-Game Talk: Ready. Set. Terrible | Jets lose 4-1

[Duke749]

Btw, I haven't read "Moneyball", but would I be right to assume that would take you a lot deeper then SV% and GAA? More into advanced stats really?

 

[Positive]

I blame Pavelec's salary on me. On the TSN play of the year contest I set up a Firefox macro and voted for him like 10000 times. I just left it running. Sorry guys.

 

[theamazingchris]

Btw, I haven't read "Moneyball", but would I be right to assume that would take you a lot deeper then SV% and GAA? More into advanced stats really?

More generally that people are prone to biases, and a healthy dose of statistics can moderate them. I'm not saying it's impossible to diagnose problems from observation alone, but humans are fallible and forgetful, y'know? At least I am

 

[Hank Chinaski]

Btw, I haven't read "Moneyball", but would I be right to assume that would take you a lot deeper then SV% and GAA? More into advanced stats really?

Moneyball is an amazing book, the movie (which is awesome in it's own right) doesn't do it justice.

The thing about those advanced stats you mention, a large amount of the book is devoted to how baseball used to be in the dark ages with regards to stats. Then along came Bill James, who slowly revolutionized the sport's statistical measurements. It isn't all about stats though, it delves into other problems that are inherent in assessing talent and performance in sports, natural human biases chief among them.

Hockey hasn't come nearly as far as baseball in embracing a statistical revolution, but it's getting there. I believe there are still some problems with bad data being used (the old garbage input -> garbage output problem) and some faulty conclusions being drawn. But that's hardly exclusive to hockey stats, heck, you could say that's a problem across many scientific disciplines. I've certainly been warming up to the advanced stats perspective, and it sure helps we've got some people on this board with a great grasp of it.

Anyway, that's my little OT ramble. Back to your regularly scheduled Pavelec debate.

 

[Libertyhaze]

I blame Pavelec's salary on me. On the TSN play of the year contest I set up a Firefox macro and voted for him like 10000 times. I just left it running. Sorry guys.

Ugh, I didn't even think about this... I had my auto clicker going on my laptop all day too. Karma is a *****

 

[truck]

Numbers do lie, Chris Osgood anyone?

Which numbers lied about Osgood?

He was a nothing special goalie on a bumch of great teams.

 

[TroubaFan1]

Which numbers lied about Osgood?

He was a nothing special goalie on a bumch of great teams.

The great teams made his numbers better. The numbers lied about his play, because of the teams that were infront of him.

 

[truck]

The great teams made his numbers better. The numbers lied about his play, because of the teams that were infront of him.

His numbers didn't lie. His sv% numbers were never elite.

 

[garret9]

How come no non-Jets' fans find Pavelc to be an above average goalie...?

How come the only thing you'd have to change is Pavelec's save% and the Jets' would have been (barely) above average in EVERY category, this and last season (with exception of PP for this season)...?

 

[garret9]

Btw, I haven't read "Moneyball", but would I be right to assume that would take you a lot deeper then SV% and GAA? More into advanced stats really?

I'll answer
People have taken a delve into SV% and GAA...
Years of work has been done to investigate shot quality, goaltending and shot volume, by many many highly intelligent people.

In summary, the conclusion has been:

SV%
*players/team/system can push up goalie's SV% by pushing pucks outside of "danger" zone, which tends to be about 10ish meters, but this usually at a maximum improves SV% by 0.008 (see Tippett)
*goalies SV% in small samples vary wildly and can even be poor in bad games and great in poor games, but once you expand the shots to ~3000 every goaltender becomes quite stable and is very indicative of talent
*changes in teams and/or system rarely changes a goalie's SV% outside of natural variance, again the 3000 shot rule seems to show this

SA
*this is the best way to show a strong defensive game... look at last season and you'll see something that matches with the eye test and strong defensive teams for the best SA/60
*this is how a team can best control their PK and GA without changing goaltending
*while it is correct over short periods of time, one shot may be more of a threat than another, it's been shown that shot quality only accounts for a maximum approximately 10% of SH% in micro situations, and once you go to macro (like a full season), it becomes almost nil

GAA
*this stat actually IS what some of the people here who defend Pav's numbers claim SV% is... it is strongly affected by team play, given a sturdy SV%
*one example: Brodeur's SV% didn't change much when they moved away from their super defensive system in the past, but the teams SA/60 and Brodeur's GAA did
*makes sense when you think about it logically... if a goalie's SV% didn't change

 

[truck]

I'll answer
People have taken a delve into SV% and GAA...
Years of work has been done to investigate shot quality, goaltending and shot volume, by many many highly intelligent people.

In summary, the conclusion has been:

SV%
*players/team/system can push up goalie's SV% by pushing pucks outside of "danger" zone, which tends to be about 10ish meters, but this usually at a maximum improves SV% by 0.08 (see Tippett)
*goalies SV% in small samples vary wildly and can even be poor in bad games and great in poor games, but once you expand the shots to ~3000 every goaltender becomes quite stable and is very indicative of talent
*changes in teams and/or system rarely changes a goalie's SV% outside of natural variance, again the 3000 shot rule seems to show this

SA
*this is the best way to show a strong defensive game... look at last season and you'll see something that matches with the eye test and strong defensive teams for the best SA/60
*this is how a team can best control their PK and GA without changing goaltending
*while it is correct over short periods of time, one shot may be more of a threat than another, it's been shown that shot quality only accounts for a maximum approximately 10% of SH% in micro situations, and once you go to macro (like a full season), it becomes almost nil

GAA
*this stat actually IS what some of the people here who defend Pav's numbers claim SV% is... it is strongly affected by team play, given a sturdy SV%
*one example: Brodeur's SV% didn't change much when they moved away from their super defensive system in the past, but the teams SA/60 and Brodeur's GAA did
*makes sense when you think about it logically... if a goalie's SV% didn't change

Only other thing worth noting. Even strength save % is the number most look at to judge a goalie.

PK save % fluctuates more year to year, but over 3 ish seasons even than noise tends to vanish.

There is no number anywhere that suggests Pav is among the top 20 goalies in the NHL outsode of the number of people who are delusional.

 

[garret9]

From this game:

shows you how the matchups went and how they did, using Fenwick (aka shots on goal, goals, and missed attempts)

for players totals and other such things check here:
http://www.arcticicehockey.com/winn...013/3/5/4069828/nightly-fenwick-jets-panthers

 

[cbcwpg]

I don't know what bugs me more about losing to Florida ....

Losing to what is basically an AHL team because of all the injures

Or

Seeing Antropov and. Burmistrov sharing a laugh while being benched in the 3rd period.

The Jets won't give up on Burmistrov because he's young and still has a lot of upside, but take this to the bank.... Antropov is playing his last games as a Jet. I don't care if they can only trade him for a puck bag or just let him walk because he's an UFA.... He's gone after this season.

 

[Hank Chinaski]

New GDT is up, please resume there.