HFBoards

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

Is there an equivalent of a "Moneyball" for the NHL?

Reply
 
Thread Tools
Old
08-08-2012, 03:10 PM
  #201
Happy Fan
nifty
 
Happy Fan's Avatar
 
Join Date: Jun 2007
Posts: 3,261
vCash: 500
Quote:
Originally Posted by Champions of Nothing View Post
Read the book. The movie doesn't do justice to the real science behind it.
cool! i'll look for it in the library soon

Happy Fan is offline   Reply With Quote
Old
08-08-2012, 03:51 PM
  #202
Kane One
Registered User
 
Join Date: Feb 2010
Location: Brooklyn, New NY
Country: United States
Posts: 28,805
vCash: 1500
I don't think there is an equivalent. The whole sabremetric thing got them nowhere. The farthest they got was to the first round of the playoffs.

__________________
Kane One is online now   Reply With Quote
Old
08-08-2012, 03:53 PM
  #203
Doctor No
Mod Supervisor
Retired?
 
Doctor No's Avatar
 
Join Date: Sep 2005
Posts: 24,041
vCash: 500
Quote:
Originally Posted by kaneone View Post
I don't think there is an equivalent. The whole sabremetric thing got them nowhere. The farthest they got was to the first round of the playoffs.
Who's the "they" in that sentence?

(And if it's the Athletics, then how does that show that there's not an equivalent in the NHL?)

Doctor No is offline   Reply With Quote
Old
08-08-2012, 04:57 PM
  #204
bucks_oil
Registered User
 
Join Date: Aug 2005
Posts: 1,986
vCash: 500
Quote:
Originally Posted by Roboturner913 View Post
Best example I can think of is Jim Rutherford when last lockout ended. He signed castoffs like Whitney, Stillman, Gerber that nobody else wanted and put them in positions to become stars, let rookies and 2nd year guys take key roles, built a team around depth rather than star power.

Didn't work long-term but that probably had more to do with age, unfortunate injuries and changes in the way the game was being played.
Same can very much be said for Edmonton in 2006. Both of those teams were build around solid two way players (Brind-Amour, Wesley, Horcoff, Pisani, Peca, Stoll) insulating a relatively few offensive guys (Staal, Williams, Hemsky, Samsonov).

Whatever the "measure" is, I have no doubt those guys would have scored off the charts on it.

In my view it boils down fairly closely to puck possession, (which itself is an outcome of a number of skills)... the problem is that it is more difficult to find cheaper guys that are good puck possessors... they tend to be the Lidstrom and Sedins of the league.

bucks_oil is offline   Reply With Quote
Old
08-08-2012, 05:00 PM
  #205
bucks_oil
Registered User
 
Join Date: Aug 2005
Posts: 1,986
vCash: 500
Quote:
Originally Posted by Henrik To Daniel View Post
Speaking of Moneyball is it a good movie? It's the only Jonah Hill movie i haven't watched and he's my favorite actor
If you are a fan of his, you will like it. He does a very good job in a much more serious role than he normally plays.

bucks_oil is offline   Reply With Quote
Old
08-08-2012, 05:08 PM
  #206
Kane One
Registered User
 
Join Date: Feb 2010
Location: Brooklyn, New NY
Country: United States
Posts: 28,805
vCash: 1500
Quote:
Originally Posted by Taco MacArthur View Post
Who's the "they" in that sentence?

(And if it's the Athletics, then how does that show that there's not an equivalent in the NHL?)
The reason the A's use Sabremetrics is because they have a tight budget. They stacked up on players who would be considered scraps on other teams, who had a good fielding percentage and a good OBP.

Since the NHL has a salary cap and a salary floor, teams don't have to resort to this. NHL teams don't have to worry as much about their players signing with a big market team like how the A's worried that their players would get signed by the Yankees and the Red Sox.

Now with the Sabremetrics, baseball has many ways where someone can get on base, advance bases, and score. The most popular way to score is obviously a home run. Hitting a single is just as good as drawing a walk, but the players who walk a lot get overlooked and don't make as much as players who get a lot of hits.

In hockey, there's only one way to score. There aren't really that many stats in hockey like there is in baseball. No statistic in hockey tells the story like they could do in baseball.

Obviously I can't tell anyone for certain that there is no "Moneypuck" in hockey, but I just can't think of anything.

Kane One is online now   Reply With Quote
Old
08-08-2012, 05:10 PM
  #207
Doctor No
Mod Supervisor
Retired?
 
Doctor No's Avatar
 
Join Date: Sep 2005
Posts: 24,041
vCash: 500
Quote:
Originally Posted by kaneone View Post
Now with the Sabremetrics, baseball has many ways where someone can get on base, advance bases, and score. The most popular way to score is obviously a home run. Hitting a single is just as good as drawing a walk, but the players who walk a lot get overlooked and don't make as much as players who get a lot of hits.

In hockey, there's only one way to score. There aren't really that many stats in hockey like there is in baseball. No statistic in hockey tells the story like they could do in baseball.
In both hockey and baseball, there is only one way to score (goals or runs).

And in both hockey and baseball, there are many ways to do it.

Doctor No is offline   Reply With Quote
Old
08-08-2012, 05:20 PM
  #208
Kane One
Registered User
 
Join Date: Feb 2010
Location: Brooklyn, New NY
Country: United States
Posts: 28,805
vCash: 1500
Quote:
Originally Posted by Taco MacArthur View Post
In both hockey and baseball, there is only one way to score (goals or runs).

And in both hockey and baseball, there are many ways to do it.
Come on, you know what I mean.

In baseball, you can draw a walk with the bases loaded and you get a run.
You can get a hit with a man on third and you get a run.
You can hit with a man on second and third and possibly get TWO runs.
How about them grand slams?

You can only score one goal at a time in hockey. In baseball, I guess that's true, too, but you can score more than one run on the same play.

If you don't have too many home run hitters on the team, you have to focus on getting as many men on base as you can, instead of getting one on base and the next batter randomly hits a home run to drive in two runs.

Kane One is online now   Reply With Quote
Old
08-08-2012, 05:41 PM
  #209
stan the caddy
Registered User
 
stan the caddy's Avatar
 
Join Date: Sep 2011
Posts: 1,684
vCash: 500
The whole OBP Sabre stuff is massively overrated. They achieved their success by bringing up three Cy Young contending pitchers and two MVP bats in a short span of time. They were cheap because they were young. Like baseball, rookies are cheap in the NHL. Draft well.

stan the caddy is offline   Reply With Quote
Old
08-08-2012, 05:51 PM
  #210
almostawake
Registered User
 
Join Date: Jan 2006
Posts: 3,265
vCash: 500
Hockey's Moneyball could very well being going on right now for all we know. Until Michael Lewis got involved the vast majority of MLB teams were clueless about what the A's were up to, let alone the media. The very point of a quantitative approach is to uncover some inefficiency, a systematic mis-pricing of players/assets. Once every other team (or a significant fraction of teams) start correctly pricing players your advantage disappears. Thus it's in the best interest of teams to keep their in-house analysis techniques just that, in-house.

As far as what's going on in hockey. GMs aren't staying up all night pouring over Corsi or Fenwick. They're collecting new stats, developing optical recognition algorithms, and doing high level positional/matchup analysis. Basically, the odds that teams are only trying to exploit the statistics that the NHL collects through their RTSS game-by-game server is pretty low. Teams are trying to quantify the game in other ways using the raw game footage.

almostawake is offline   Reply With Quote
Old
08-08-2012, 06:00 PM
  #211
almostawake
Registered User
 
Join Date: Jan 2006
Posts: 3,265
vCash: 500
Quote:
Originally Posted by stan the caddy View Post
The whole OBP Sabre stuff is massively overrated. They achieved their success by bringing up three Cy Young contending pitchers and two MVP bats in a short span of time. They were cheap because they were young. Like baseball, rookies are cheap in the NHL. Draft well.
Err, not sure where you're getting this 'rookies are cheap' idea from. The sign-ability/cost of a player is a major factor in 1st round selections for over 80% of teams in major league baseball.

almostawake is offline   Reply With Quote
Old
08-09-2012, 01:07 PM
  #212
stan the caddy
Registered User
 
stan the caddy's Avatar
 
Join Date: Sep 2011
Posts: 1,684
vCash: 500
Quote:
Originally Posted by almostawake View Post
Err, not sure where you're getting this 'rookies are cheap' idea from. The sign-ability/cost of a player is a major factor in 1st round selections for over 80% of teams in major league baseball.
They're still cheap in the grand scheme of things. At one one point the A's had Giambi, Tejada, Chavez, Zito, Mulder, and Hudson and they could afford them all because they were young and cheap. They were all playing well beyond their value while the A's had them. That's the major fallacy with moneyball, teams didn't "undervalue" Beane's players, Beane just didn't have to pay his star players what they were really worth. It works like that with hockey. You look at a guy like Eberle, he put up a monster season making just over a million dollars.

The book devotes an entire chapter to Chad Bradford and barely mentions what was arguably the best pitching rotation in baseball at the time. As much as the Sabre's want to bash old school scouting, that's what got them there. If you bring up 2 MVP's, 1 Cy Young, and a bunch of other all star caliber players within a span of 4-5 years, you're going to be good unless you really suck at filling out the roster.

stan the caddy is offline   Reply With Quote
Old
11-28-2012, 05:43 PM
  #213
Missionhockey
Registered User
 
Missionhockey's Avatar
 
Join Date: Jul 2003
Location: New Jersey
Country: United States
Posts: 7,132
vCash: 500
Moneyball relation to hockey?

I saw the movie Moneyball recently (fantastic movie if you haven't seen it) and I found it extremely interesting. It took a tremendous amount of ingenuity to rethink how construct a team and I wondered how this concept would be applied to hockey and if there already sabremetrics applied to it? In the movie, the talk about instead of buying wins, they should be buying runs. I would think that in hockey terms you would be buying goals? Few questions I guess would be:

How many goals would it take a reach the post season?

How many goals against could be allowed to reach the post season?

What actual statistics are the best indicators of producing goals?

I'm not much of a math person so forgive me. However I would have to think that puck possession would have the most positive correlation.

Missionhockey is offline   Reply With Quote
Old
11-28-2012, 05:55 PM
  #214
TrillMike
Registered User
 
TrillMike's Avatar
 
Join Date: Feb 2012
Location: Dallas, TX
Posts: 4,639
vCash: 500
I think a lot of people wondered about this as it relates to the NHL more. I really have no idea. I think there are so many variables and outliers in the game to actually "predict" what a team needs to do to make the playoffs. In my opinion, with all the outliers (like forechecking pressure, d play, injury, pims, chemistry...etc) hockey would be significantly harder to predict than baseball. I think chemistry is the biggest difference. I'm not bashing baseball, it's my 2nd favorite sport by the way.

TrillMike is offline   Reply With Quote
Old
11-28-2012, 05:55 PM
  #215
Bergeron47
Registered User
 
Bergeron47's Avatar
 
Join Date: Apr 2004
Location: Australia
Country: Australia
Posts: 7,885
vCash: 500
I don't think you can quite follow a statistical method as you can in baseball. Hockey involves more team chemistry & structured systems.
Baseball players are far more independent & consistent with offensive statistics. Barring RBI, a player's teammates have no impact for their at-bat.

Bergeron47 is online now   Reply With Quote
Old
11-28-2012, 06:07 PM
  #216
DownFromNJ
Registered User
 
Join Date: Mar 2004
Posts: 2,536
vCash: 500
Measuring those concepts with statistics is pretty much impossible with hockey, for reasons stated above. You just don't have the raw data that baseball generates. Every single event is recorded as an isolated data point in baseball. Events just aren't isolated like that in hockey.

More broadly, I think the basic idea of looking at players as net contributors is a useful concept that you can apply to hockey. The Devils just lost Zach Parise, who is probably worth X wins over a replacement-level hockey player. The Devils need to improve their team by X in order to come back to the same level. That can come through offense, defensive, or goaltending.

DownFromNJ is offline   Reply With Quote
Old
11-28-2012, 06:10 PM
  #217
Hammer Time
Registered User
 
Hammer Time's Avatar
 
Join Date: May 2011
Country: Canada
Posts: 3,710
vCash: 500
Quote:
Originally Posted by Missionhockey View Post
I saw the movie Moneyball recently (fantastic movie if you haven't seen it) and I found it extremely interesting. It took a tremendous amount of ingenuity to rethink how construct a team and I wondered how this concept would be applied to hockey and if there already sabremetrics applied to it? In the movie, the talk about instead of buying wins, they should be buying runs. I would think that in hockey terms you would be buying goals? Few questions I guess would be:

How many goals would it take a reach the post season?

How many goals against could be allowed to reach the post season?

What actual statistics are the best indicators of producing goals?

I'm not much of a math person so forgive me. However I would have to think that puck possession would have the most positive correlation.
Indeed, hence the existence of possession stats such as Fenwick.
Fenwick is measured as the number of shots on goal and missed shots taken by a team, divided by the total number of shots on goal and missed shots taken (by both teams). A 100% Fenwick means that the other team didn't get any shots (implying that you pretty much had complete puck possession), and a 0% Fenwick means your team didn't get any shots. Pretty much all NHL teams have Fenwicks between 40% and 60%.

The idea is that increased puck possession leads to quality scoring chances, quality scoring chances lead to goals, goals lead to wins, and wins lead to championships.

Behindthenet is a great statistics website which tracks Fenwick and other "advanced" stats. http://www.behindthenet.ca/

Stats probably aren't as useful in hockey as they are in baseball due to the fluid nature of the game (instead of the one person at a time, one pitch at a time in baseball). However, it isn't coincidence that since Behindthenet began tracking Fenwick in 2007, the top two teams in that stat were the 2008 Red Wings and 2010 Blackhawks.

The main weakness of Fenwick is that it doesn't measure the impact of goalies, who have the ability to neutralize great offences, and the impact of great defencemen who can limit the quality of opponent shots. Fenwick will therefore underestimate the performance of teams with great goalies and great defences, and overestimate the performance of teams with crap goalies and D. (for instance, the Bruins were a 50% Fenwick team in 2011, and the Predators were 2nd-last in Fenwick in 2012. Both these teams were much better than their Fenwick indicates.)

Hammer Time is offline   Reply With Quote
Old
11-28-2012, 06:11 PM
  #218
UnrefinedCrude
Registered User
 
UnrefinedCrude's Avatar
 
Join Date: Jun 2011
Country: Canada
Posts: 2,027
vCash: 833
In hockey it seems goal differential is the king, not just GF or GA.

The thing is there are so many factors (read every factor,) that leads to this.

Moneyball works as it does in baseball because more than any sport Baseball is a collection of individual efforts. Play is also less fluid.

There might be a way to quantify a MoneyPuck system, but I think it would have to be more complex, with more subtle calculations involved.

Advanced stats in hockey are still in their infancy, and the continuous play, body contact and a myriad of other factors not present in baseball make each metric less definitive, and much harder to isolate.

UnrefinedCrude is offline   Reply With Quote
Old
11-28-2012, 06:13 PM
  #219
TrillMike
Registered User
 
TrillMike's Avatar
 
Join Date: Feb 2012
Location: Dallas, TX
Posts: 4,639
vCash: 500
Quote:
Originally Posted by DownFromNJ View Post
Measuring those concepts with statistics is pretty much impossible with hockey, for reasons stated above. You just don't have the raw data that baseball generates. Every single event is recorded as an isolated data point in baseball. Events just aren't isolated like that in hockey.

More broadly, I think the basic idea of looking at players as net contributors is a useful concept that you can apply to hockey. The Devils just lost Zach Parise, who is probably worth X wins over a replacement-level hockey player. The Devils need to improve their team by X in order to come back to the same level. That can come through offense, defensive, or goaltending.
Exactly. Great example. Here's an example with the Stars: Brad Richards leaves in FA. The team signs Michael Ryder, Sheldon Souray, Adam Burish and Vernon Fiddler to replace Richards' offensive production and actually add more depth/grit/D than Richards would have put up. (The team finished about the same without him). There's tons of better examples out there.


Last edited by spiny norman: 11-28-2012 at 08:00 PM. Reason: corrected [/B][/QUOTE]
TrillMike is offline   Reply With Quote
Old
11-28-2012, 10:59 PM
  #220
Missionhockey
Registered User
 
Missionhockey's Avatar
 
Join Date: Jul 2003
Location: New Jersey
Country: United States
Posts: 7,132
vCash: 500
Quote:
Originally Posted by UnrefinedCrude View Post
In hockey it seems goal differential is the king, not just GF or GA.

The thing is there are so many factors (read every factor,) that leads to this.

Moneyball works as it does in baseball because more than any sport Baseball is a collection of individual efforts. Play is also less fluid.

There might be a way to quantify a MoneyPuck system, but I think it would have to be more complex, with more subtle calculations involved.

Advanced stats in hockey are still in their infancy, and the continuous play, body contact and a myriad of other factors not present in baseball make each metric less definitive, and much harder to isolate.
Perhaps, but I would argue that maybe they aren't piggy backing off the right stats. To dismiss the notion that advanced statistics wouldn't help evaluate players better than certain archaic hockey values seems almost naive. After all, many hockey fans and probably GMs and coaches place a ridiculously high value on two way play, but how much more value really does someone like Manny Malhotra have over someone like Michael Ryder? What statistical evidence proves that this player is better than that player?

Basketball employs the same tactics, and I would argue that basketball is also a fluid game.

Missionhockey is offline   Reply With Quote
Old
11-29-2012, 01:09 AM
  #221
Vamos Rafa
ADT = GH
 
Vamos Rafa's Avatar
 
Join Date: Jan 2010
Location: Clay Court
Country: Spain
Posts: 11,132
vCash: 500
Oh God no. As a baseball fan who can't stand Sabremetric fan boys, I see hockey as my asylum.

Vamos Rafa is offline   Reply With Quote
Old
11-29-2012, 09:55 AM
  #222
RedWingsForPresident
Registered User
 
RedWingsForPresident's Avatar
 
Join Date: Nov 2012
Location: Indiana
Posts: 1,324
vCash: 500
If you look at the St Louis Blues from last year, they were somewhat Moneyball themselves. A Cap of 51,180,000. That means they left 19,020,000 in cap space for the season. Just like in the movie, they started the season out rough but then started putting wins together and finished 2nd in the Western Conference and they were contending for the Presidents Trophy in the last week of the season.

I wouldn't say they were buying goals though like the A's were buying runs. The Blues had one of the lowest scoring offenses in the NHL but had maybe the best defense in the league. I guess you could argue they were buying GAA?

That's the closest comparison I can think of in the NHL

RedWingsForPresident is offline   Reply With Quote
Old
11-29-2012, 10:03 AM
  #223
Hynh
Registered User
 
Join Date: Jun 2012
Posts: 1,351
vCash: 500
Wait, moneyball (making the most of limited dollars, something every team in the NHL has to achieve now that there is a salary cap) or mathematical analysis of players? Two different concepts, with one helping the other.

Hynh is offline   Reply With Quote
Old
11-29-2012, 10:10 AM
  #224
P U L L H A R D
 
P U L L H A R D's Avatar
 
Join Date: Jul 2007
Location: Ottawa
Country: Somalia
Posts: 24,413
vCash: 500
If we're ignoring the salary part, I think I'd try to get players that historically have racked up the most takeaways. Seems like that type of effort would lead to more offensive chances. This is just my gut feeling.

P U L L H A R D is offline   Reply With Quote
Old
11-29-2012, 10:18 AM
  #225
JohnnyOnTheSpot
I Believe in G-Sus
 
JohnnyOnTheSpot's Avatar
 
Join Date: Feb 2010
Location: Philadelphia
Country: United States
Posts: 1,904
vCash: 500
Quote:
Originally Posted by Eternal Sunshine View Post
If we're ignoring the salary part, I think I'd try to get players that historically have racked up the most takeaways. Seems like that type of effort would lead to more offensive chances. This is just my gut feeling.
You ever watch Michael Handzus on the PK? As a Flyers fan, I'd cheer because it broke up cross ice passes for a clean breakaway constantly, then cry when I realized it was the mud monster, and he would be caught before he hit the red line. Takeaways aren't always the best hahaha.

As for Moneyball, it'd never work in the NHL, but the closest I think we see to the concept is the Predators. They continuously ice a competitive team every year, on a budget. They get the most out of the players, and when someone prices out, they manufacture a replacement. Baseball is a stat sport because it's X vs. Y, there are so many more variables in hockey and so much left to uncertainty. You smack one out of the park in baseball, it's a homerun. No matter how perfect of a shot you take, goalie can always interject.

JohnnyOnTheSpot 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 01:27 AM.

monitoring_string = "e4251c93e2ba248d29da988d93bf5144"
Contact Us - HFBoards - Archive - Privacy Statement - Terms of Use - Advertise - Top - AdChoices

vBulletin Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
HFBoards.com is a property of CraveOnline Media, LLC, an Evolve Media, LLC company. 2014 All Rights Reserved.