HFBoards

Go Back   HFBoards > General Hockey Discussion > By The Numbers
Mobile Hockey's Future Become a Sponsor Site Rules Support Forum vBookie Page 2
Notices

By The Numbers Hockey Analytics... the Final Frontier. Explore strange new worlds, to seek out new algorithms, to boldly go where no one has gone before.

Fenwick Tied Rankings

Reply
 
Thread Tools
Old
11-20-2011, 03:05 PM
  #51
Robo-Pope
Registered User
 
Join Date: Feb 2007
Posts: 1,076
vCash: 500
Do you have an opponent-adjusted Fenwick Tied? Seems like teams wouldn't tend to be tied for equal amounts of time against teams of all skill levels.

Robo-Pope is offline   Reply With Quote
Old
11-20-2011, 03:08 PM
  #52
EVBetting Site
Registered User
 
Join Date: Jun 2011
Location: Edmonton
Country: Canada
Posts: 348
vCash: 500
Quote:
Originally Posted by bluesfan94 View Post
Yeah, I understand variance will never be gone, but if an 82 game sample size isn't enough for the convergence to the average, then the whole idea of correlation is a nonfactor in hockey, with teams changing on a yearly basis. True, variance will likely have less of an e effect over 82 games, but if teams like STL and NJD last year were out of the playoffs despite statistics suggesting they should be in, then correlation shouldn't be regarded as a predictor.
Sure, and it seems completely stupid when a team like Washington dominates by every metric and then gets bounced early in the playoffs, but thats why we watch hockey/sports and why life is interesting in general.

Nothing can be used to predict outcomes/results of games correctly. If you could, you would be extremely wealthy through sports betting. But there are definitely things/stats/metrics/etc that allow you to predict frequencies that are closer to the elusive true average for any particular game. Possession metrics as well as a relatively precise range of goaltender ability (goalies are good for this because they face heavy volume of shots and have a single stat that encompasses almost the whole of their role) allow you to come up with decent frequency predictions though and are, in my experience with sports betting, the most valuable/relevant statistics out there.

Hockey is a relatively new public sport, still arguably not there yet, so Im interested to see how long it takes for more relevant and important stats to make their way onto broadcasts, much like baseball. For the longest time, the only stats that would get shown on TV were average, HR, RBI, ERA, etc.. which are all nearly meaningless. Similarly in hockey, G-A-Points and especially PIM/hits/blocked shots are (almost) meaningless as well, but it will take at least 5+ years for anything relevant to make the mainstream.

EVBetting Site is offline   Reply With Quote
Old
11-20-2011, 03:11 PM
  #53
Jason MacIsaac
MARS Officer
 
Jason MacIsaac's Avatar
 
Join Date: Jan 2004
Location: Halifax, NS
Country: Canada
Posts: 17,104
vCash: 500
Send a message via MSN to Jason MacIsaac Send a message via Yahoo to Jason MacIsaac
Quote:
Originally Posted by Robo-Pope View Post
Do you have an opponent-adjusted Fenwick Tied? Seems like teams wouldn't tend to be tied for equal amounts of time against teams of all skill levels.
No but that would be an interesting dynamic to this ranking. Find what the combined Fenwick Tied rating is for your opposition. It would be very helpful after 10 20 or 30 game intervals.

Jason MacIsaac is offline   Reply With Quote
Old
11-20-2011, 03:13 PM
  #54
iamitter
Thornton's Hen
 
iamitter's Avatar
 
Join Date: May 2011
Location: NYC
Country: United States
Posts: 3,511
vCash: 500
Quote:
Originally Posted by Les Wynan View Post
With the 0.928 even-strength SV% and 8% even-strength SH% the Rangers managed last season, they would currently be being outscored 2.24-1.90 per sixty minutes of even-strength hockey instead of outscoring teams 2.5-2.0 as they are right now.
This is going to be my last post to you because you continue to cherry pick what you like.

Yet again, you pick our shot percentage from last year and our shots/game from this year. At least be friggen consistent. Why not take our shot percentage from this year and our shots/game from last year? We'd be a team full of Crosbys.

iamitter is offline   Reply With Quote
Old
11-20-2011, 03:17 PM
  #55
Jason MacIsaac
MARS Officer
 
Jason MacIsaac's Avatar
 
Join Date: Jan 2004
Location: Halifax, NS
Country: Canada
Posts: 17,104
vCash: 500
Send a message via MSN to Jason MacIsaac Send a message via Yahoo to Jason MacIsaac
Quote:
Originally Posted by iamitter View Post
This is going to be my last post to you because you continue to cherry pick what you like.

Yet again, you pick our shot percentage from last year and our shots/game from this year. At least be friggen consistent. Why not take our shot percentage from this year and our shots/game from last year? We'd be a team full of Crosbys.
I suggest you read the blog article I posted back on the second page. I know 20 games probably is too early to tell if the Rangers will change but it's still interesting to read.

Jason MacIsaac is offline   Reply With Quote
Old
11-20-2011, 03:19 PM
  #56
Les Wynan*
 
Join Date: Nov 2011
Posts: 1,745
vCash: 500
Quote:
Originally Posted by squidz View Post
If that's the case, he can feel free to show the stats on it. It's not like it's a new concern, but none of the random blogs that the "hockey Sabremetrics" types like to use as "proof" ever seem to post any of the important parts of the statistics.

The larger point that's being missed is that he's gone off the deep end in regards to conclusions based upon a single, low correlation stat which hasn't even been established to fall within an extremely generous α value.
No one's claiming possession metrics can perfectly predict the future or are even 100% accurate measures of a team's talent. Just that it's been shown that shooting percentage (which turns a shot into a goal) is not something that remains consistent from season to season among teams, indicating that it isn't an actual talent or even something teams are capable of significantly influencing. Shot differential is the complete opposite. So why judge a team based on goal-based metrics (including standings points) over a small sample size when they've been proven to be driven primarily by luck?

Les Wynan* is offline   Reply With Quote
Old
11-20-2011, 03:28 PM
  #57
Les Wynan*
 
Join Date: Nov 2011
Posts: 1,745
vCash: 500
Quote:
Originally Posted by iamitter View Post
This is going to be my last post to you because you continue to cherry pick what you like.

Yet again, you pick our shot percentage from last year and our shots/game from this year. At least be friggen consistent. Why not take our shot percentage from this year and our shots/game from last year? We'd be a team full of Crosbys.
I'm not cherry-picking. A 20-game sample size is far too small to conclude that the Rangers' shooting or save percentage in that span is a factor of their true talent. So is an 82-game sample size but it at least provides a better approximation of what New York will likely end up with at the end of the year than their SV% and SH% now.

Like I said, they're likely to improve their shot differential based on past seasons' performances but it's far more likely they stay at their current SF/60 and SA/60 rates than them sustaining their current shooting and save percentages based on the fact that NHL teams have been as poor territorially in the past as the Rangers are so far this season but no team in the past four years has had even close to the PDO rating NYR currently benefits from.

Les Wynan* is offline   Reply With Quote
Old
11-20-2011, 03:29 PM
  #58
squidz*
 
Join Date: Jun 2011
Location: South of the Border
Country: United States
Posts: 11,897
vCash: 500
Quote:
Originally Posted by Les Wynan View Post
No one's claiming possession metrics can perfectly predict the future or are even 100% accurate measures of a team's talent. Just that it's been shown that shooting percentage (which turns a shot into a goal) is not something that remains consistent from season to season among teams, indicating that it isn't an actual talent or even something teams are capable of significantly influencing. Shot differential is the complete opposite. So why judge a team based on goal-based metrics (including standings points) over a small sample size when they've been proven to be driven primarily by luck?
Except (as was repeatedly pointed out the last time this whole discussion hashed up) shooting % has a significant correlation (almost perfectly correcting for Corsi's inaccuracies) with winning %.

Regardless, you're making a pretty painful false dichotomy. I never brought up shooting %. If you're really stretching things, you could argue that I was relying upon shooting % against, but you haven't done anything to debunk that stat. The point is, you picked one stat that you determined was the be all end all of stats. One team under Payne for a single season was "good" at that stat, so obviously he's the best coach in the history of the NHL right?

squidz* is offline   Reply With Quote
Old
11-20-2011, 03:31 PM
  #59
squidz*
 
Join Date: Jun 2011
Location: South of the Border
Country: United States
Posts: 11,897
vCash: 500
Quote:
Originally Posted by Les Wynan View Post
I'm not cherry-picking. A 20-game sample size is far too small to conclude that the Rangers' shooting or save percentage in that span is a factor of their true talent. So is an 82-game sample size but it at least provides a better approximation of what New York will likely end up with at the end of the year than their SV% and SH% now.
Oh but I thought previous year SH% wasn't predictive of current year SH%.

squidz* is offline   Reply With Quote
Old
11-20-2011, 03:31 PM
  #60
iamitter
Thornton's Hen
 
iamitter's Avatar
 
Join Date: May 2011
Location: NYC
Country: United States
Posts: 3,511
vCash: 500
Quote:
Originally Posted by Jason MacIsaac View Post
I suggest you read the blog article I posted back on the second page. I know 20 games probably is too early to tell if the Rangers will change but it's still interesting to read.
I already did. I'm not blind to what the stats say, I'm a senior applied math major ffs. Sports is one of the worst things you can apply statistics like these two because of the ridiculously low sample sizes (82 games isn't worth anything statistically speaking). When you get to the playoffs, you face off in a best of seven series. There's going to be games during the regular season where teams play better than they are and worse than they are. How often exactly do you see teams scoring at their average, shooting at their averages, making saves at their averages? Winning in sports means a few things clicking at the right time, the wind blowing your way.

This is why claims about the Rangers shot percentage really tick me off. This year has been so frustrating in that we're missing the net way more than we did last year. They're not necessarily high scoring chance shots, but they're shots nonetheless. It's these things (in a 17 game sample size) that make us seem we're shooting more accurately, but less. We're not. A few of those shots hit a goalie's pads and our shot percentage is where it should be and the same with shots/game.

iamitter is offline   Reply With Quote
Old
11-20-2011, 03:42 PM
  #61
Les Wynan*
 
Join Date: Nov 2011
Posts: 1,745
vCash: 500
Quote:
Originally Posted by squidz View Post
Except (as was repeatedly pointed out the last time this whole discussion hashed up) shooting % has a significant correlation (almost perfectly correcting for Corsi's inaccuracies) with winning %.

Regardless, you're making a pretty painful false dichotomy. I never brought up shooting %. If you're really stretching things, you could argue that I was relying upon shooting % against, but you haven't done anything to debunk that stat. The point is, you picked one stat that you determined was the be all end all of stats. One team under Payne for a single season was "good" at that stat, so obviously he's the best coach in the history of the NHL right?
Of course SH% has a high correlation with winning. You win games by scoring goals. But as has been pointed out in the blog post I linked you to as well as in many other places, SH% isn't something teams can significantly control whereas shot differential absolutely is. There has been a year-to-year correlation of basically 0 for team shooting percentage over the last four years, but a year-to-year correlation of 0.498 for shot differential. Which one do you think is truly indicative of a team's ability?

Cool hyperbole bro. Never said Payne was the greatest coach in history but it's obvious that he was sacrificed as a result of a below-average ES SV% which, as I've linked you to articles that show, wasn't possibly in his control. You didn't bring up shooting percentage, you brought up standings points. Standings points are achieved through goal differential. Goal differential is a factor of shot differential and shooting percentage.

Quote:
Originally Posted by squidz View Post
Oh but I thought previous year SH% wasn't predictive of current year SH%.
It isn't and I never said it was but seeing as the majority of teams converge to a very slim range in PDO, it's a better look at around where the Rangers will finish with respect to the percentages than where they are in that regard now.

Quote:
Originally Posted by iamitter View Post
This is why claims about the Rangers shot percentage really tick me off. This year has been so frustrating in that we're missing the net way more than we did last year. They're not necessarily high scoring chance shots, but they're shots nonetheless. It's these things (in a 17 game sample size) that make us seem we're shooting more accurately, but less. We're not. A few of those shots hit a goalie's pads and our shot percentage is where it should be and the same with shots/game.
Fenwick% includes missed shots.

Les Wynan* is offline   Reply With Quote
Old
11-20-2011, 03:43 PM
  #62
Jason MacIsaac
MARS Officer
 
Jason MacIsaac's Avatar
 
Join Date: Jan 2004
Location: Halifax, NS
Country: Canada
Posts: 17,104
vCash: 500
Send a message via MSN to Jason MacIsaac Send a message via Yahoo to Jason MacIsaac
Quote:
Originally Posted by iamitter View Post
When you get to the playoffs, you face off in a best of seven series. There's going to be games during the regular season where teams play better than they are and worse than they are. How often exactly do you see teams scoring at their average, shooting at their averages, making saves at their averages? Winning in sports means a few things clicking at the right time, the wind blowing your way.
I agree, which is why I say playoffs are luck. You are relying on an unpredictable stat (sh% and sv%) to determine the series.

Jason MacIsaac is offline   Reply With Quote
Old
11-20-2011, 03:44 PM
  #63
iamitter
Thornton's Hen
 
iamitter's Avatar
 
Join Date: May 2011
Location: NYC
Country: United States
Posts: 3,511
vCash: 500
Quote:
Originally Posted by Les Wynan View Post
Fenwick% includes missed shots.
Yes, but you keep saying we're bottom of the league in terms of shots/game. That doesn't include missed shots.

Whatever, I'm done here.

iamitter is offline   Reply With Quote
Old
11-20-2011, 03:47 PM
  #64
Les Wynan*
 
Join Date: Nov 2011
Posts: 1,745
vCash: 500
Quote:
Originally Posted by iamitter View Post
Yes, but you keep saying we're bottom of the league in terms of shots/game. That doesn't include missed shots.

Whatever, I'm done here.
Well they're in the bottom five in Fenwick as well so it's not a huge difference.

Les Wynan* is offline   Reply With Quote
Old
11-20-2011, 03:48 PM
  #65
EVBetting Site
Registered User
 
Join Date: Jun 2011
Location: Edmonton
Country: Canada
Posts: 348
vCash: 500
Quote:
Originally Posted by iamitter View Post
Yes, but you keep saying we're bottom of the league in terms of shots/game. That doesn't include missed shots.

Whatever, I'm done here.
Ya all thats saying, then, is that you guys are not actually missing so many shots (compared to league average).

EVBetting Site is offline   Reply With Quote
Old
11-20-2011, 03:48 PM
  #66
squidz*
 
Join Date: Jun 2011
Location: South of the Border
Country: United States
Posts: 11,897
vCash: 500
Quote:
Originally Posted by Les Wynan View Post
Of course SH% has a high correlation with winning. You win games by scoring goals. But as has been pointed out in the blog post I linked you to as well as in many other places, SH% isn't something teams can significantly control whereas shot differential absolutely is. There has been a year-to-year correlation of basically 0 for team shooting percentage over the last four years, but a year-to-year correlation of 0.498 for shot differential. Which one do you think is truly indicative of a team's ability?

Cool hyperbole bro. Never said Payne was the greatest coach in history but it's obvious that he was sacrificed as a result of a below-average ES SV% which, as I've linked you to articles that show, wasn't possibly in his control. You didn't bring up shooting percentage, you brought up standings points. Standings points are achieved through goal differential. Goal differential is a factor of shot differential and shooting percentage.
You still haven't addressed anything.

You've provided absolutely no evidence that Payne's dismissal was in any way related to, well, anything. You've now finally made a completely unsupported claim that it was a result of a poor ES SV% (which we've already established is often significantly linked to coaching), but prior to this last post your argument rested solely upon the claim that Payne's team (for an insignificant sample of 82 games) had a good Fenwick. That's not an argument, and it's an insult to everyone on these boards that you think you can try get away with that.

squidz* is offline   Reply With Quote
Old
11-20-2011, 03:55 PM
  #67
Les Wynan*
 
Join Date: Nov 2011
Posts: 1,745
vCash: 500
Quote:
Originally Posted by squidz View Post
You still haven't addressed anything.

You've provided absolutely no evidence that Payne's dismissal was in any way related to, well, anything. You've now finally made a completely unsupported claim that it was a result of a poor ES SV% (which we've already established is often significantly linked to coaching), but prior to this last post your argument rested solely upon the claim that Payne's team (for an insignificant sample of 82 games) had a good Fenwick. That's not an argument, and it's an insult to everyone on these boards that you think you can try get away with that.
Lol who's established that ES SV% is significantly linked to coaching? I don't think pulling an unsubstantiated "fact" out of your butt counts as establishing anything.

You've yet to address the fact that there is ZERO year-to-year correlation in shooting percentage at the team level but a nearly 0.5 correlation in shot differential. That would seem to support the fact that an 82-game sample size is sufficient for determining a team's competence based on Fenwick but absolutely not sufficient for accepting shooting percentage over that span as anything other than random variance. Quit trollin'.

Les Wynan* is offline   Reply With Quote
Old
11-20-2011, 04:07 PM
  #68
Jason MacIsaac
MARS Officer
 
Jason MacIsaac's Avatar
 
Join Date: Jan 2004
Location: Halifax, NS
Country: Canada
Posts: 17,104
vCash: 500
Send a message via MSN to Jason MacIsaac Send a message via Yahoo to Jason MacIsaac
Quote:
Originally Posted by squidz View Post
You still haven't addressed anything.

You've provided absolutely no evidence that Payne's dismissal was in any way related to, well, anything. You've now finally made a completely unsupported claim that it was a result of a poor ES SV% (which we've already established is often significantly linked to coaching), but prior to this last post your argument rested solely upon the claim that Payne's team (for an insignificant sample of 82 games) had a good Fenwick. That's not an argument, and it's an insult to everyone on these boards that you think you can try get away with that.
82 games isn't sufficient sample size to correlate Fenwick to goal scoring, well duh. It is impossible to predict scoring rates seeing that sh% is completely unpredictable. Once your sample size is 1000 games or so sh% is taken out of the equation Fenwick correlates positively with goal scoring.

What does this tell us? Fenwick is still the most effective measuring stick we have available and over the course of a year, sh% tend to even out.

Jason MacIsaac is offline   Reply With Quote
Old
11-20-2011, 04:51 PM
  #69
Fire AV
#FreeMcIlrath
 
Join Date: Mar 2011
Location: NYC
Country: Austria
Posts: 19,971
vCash: 420
Quote:
Originally Posted by Les Wynan View Post
1.So I present you with a numerical, reality-based look at the fact that the Rangers do not get more "high quality" chances than the other team and you refute it...based on zero evidence.
2.Yes, they're winning the games. Because their shooting percentage is currently at a ridiculous rate that no team has sustained over the course of a season the last four years
3.and Lundqvist is stopping pucks at a rate that no non-Thomas goalie has come close to in that span.
.
Let me break down all the things wrong in your argument.

1. You were talking about scoring chances, not high quality scoring chances. Huge difference. The other teams get more scoring chances, but we move the puck efficiently, to get mostly high quality scoring chances.
2. As I said in my previous statement. We're taking less useless shots, and preparing high quality chances. In that case, it's not so outlandish.
3. Yet he's only 7th, with almost the same percentage as Thomas and Rinne, while having 3 starters ahead of him. What makes you think that only Tim Thomas can get such numbers?

Fire AV is offline   Reply With Quote
Old
11-20-2011, 04:57 PM
  #70
RogerRoeper*
 
Join Date: Jun 2009
Posts: 21,850
vCash: 500
Quote:
Originally Posted by Les Wynan View Post
Those underlying numbers are terrific and they were phenomenal last season at even strength as well (10th in the league in score-tied Fenwick). A lot of that can presumably be attributed to Payne since they were a fairly mediocre possession team in 09-10. I understand they've had their troubles with special teams but the biggest reason the Blues failed to qualify for the postseason last year, and were out of the playoff picture when Payne was fired this year, was bad luck. They received awful goaltending last season and were a 993 PDO team. Variance working against them to that degree doesn't tend to persist over a larger sample size and we're seeing that now with the Blues winning games on a regular basis. Payne was essentially sacrificed due to a low even-strength SV% that he had no control over despite the fact that team in reality was playing extremely well.
It's not translating to success for the Blues at all.

RogerRoeper* is offline   Reply With Quote
Old
11-20-2011, 05:11 PM
  #71
EVBetting Site
Registered User
 
Join Date: Jun 2011
Location: Edmonton
Country: Canada
Posts: 348
vCash: 500
Quote:
Originally Posted by RogerRoeper View Post
It's not translating to success for the Blues at all.
How has it not? They are a top 10 team leaguewide this year by just about any metric

EVBetting Site is offline   Reply With Quote
Old
11-20-2011, 05:18 PM
  #72
Les Wynan*
 
Join Date: Nov 2011
Posts: 1,745
vCash: 500
Quote:
Originally Posted by nils2317 View Post
Let me break down all the things wrong in your argument.

1. You were talking about scoring chances, not high quality scoring chances. Huge difference. The other teams get more scoring chances, but we move the puck efficiently, to get mostly high quality scoring chances.
2. As I said in my previous statement. We're taking less useless shots, and preparing high quality chances. In that case, it's not so outlandish.
3. Yet he's only 7th, with almost the same percentage as Thomas and Rinne, while having 3 starters ahead of him. What makes you think that only Tim Thomas can get such numbers?
No team in the last four years has sustained a shooting percentage over the course of a season even close to the 10.2% ES rate the Rangers are currently scoring at. You really expect me to believe the Rangers somehow unlocked the secret to taking "higher quality" chances at a rate that blows away anything the rest of the league has been able to accomplish? Also that blog limits its definition of scoring chances tracked to only include shots attempted from an area that has been statistically shown to include the highest-percentage chances. Refuting it with no evidence of your own is just asinine.

It's even-strength SV% I've been referring to the whole time and only Minnesota has a higher one than the Rangers. They're an even worse team than New York and are being masked by an anomalous ES SV% to an even greater degree. Obviously that's going to regress to the mean eventually as well.

Les Wynan* is offline   Reply With Quote
Reply

Forum Jump


Bookmarks

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off



All times are GMT -5. The time now is 12:10 AM.

monitoring_string = "e4251c93e2ba248d29da988d93bf5144"

vBulletin Copyright ©2000 - 2016, Jelsoft Enterprises Ltd.
HFBoards.com is a property of CraveOnline Media, LLC, an Evolve Media, LLC company. 2016 All Rights Reserved.