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.

Can the Home-Ice Advantage be Measured Statistically?

Reply
 
Thread Tools
Old
07-30-2012, 08:40 PM
  #1
Blueline Bomber
Expectations - high
 
Blueline Bomber's Avatar
 
Join Date: Oct 2007
Posts: 23,000
vCash: 500
Can the Home-Ice Advantage be Measured Statistically?

First, let me preface this by saying I'm not a stats guy. English was always my subject in school, Math was not. I have no idea if this is relevant or if anything can be done with it, but I thought it was cool.

This is an idea I've been tossing around in my head for a couple years now. I've always been fascinated in the advantage home-ice gives a team, not only by way of the rulebook (faceoffs, shootouts, line changes, etc.), but in the "intangible" effects, such as Detroit's "springy" boards and each arena's individual crowd. Each fanbase is different, and I've always been curious if those differences could effect the way the team plays. Whether attendance has an effect on a team's performance, how "hostile" a crowd can be (either to opposing teams or to the home team), etc.

So I started the year after the lockout (since that seemed like the most logical place to start). At first, the idea was simply to see which team racks up the most wins on home ice and claim they've taken advantage of the "home ice advantage" the most. But it didn't make much sense, since a good team is going to have more wins regardless of whether they're at home or away. And unsurprisingly, the two teams that have the most wins since the lockout (Detroit and San Jose) also have the most wins on home ice (189 and 182 respectively). It seemed disadvantageous to measure those teams against teams that have, historically, been bottom-feeders.

So instead, I decided to do winning percentage. That is, of a team's total wins, how many of them came on home ice? This puts all teams on equal ground, since all teams have 41 home games, and whether they have a really good season or a really poor season often doesn't change the percentage that a team wins at home.

First, the teams in alphabetical order, for those who are just interested in their team's home-ice advantage record:

Code:
Anaheim - 2005-11 - 293 wins, 170 home  (58.02%)
Boston - 2005-11 - 292 wins, 148 home (50.68%)
Buffalo - 2005-11 - 312 wins, 165 home (52.88%)
Calgary - 2005-11 - 295 wins, 174 home (58.98%)
Carolina - 2005-11 - 288 wins, 165 home (57.29%)
Chicago - 2005-11 - 284 wins, 160 home (56.33%)
Colorado - 2005-11 - 277 wins, 154 home (55.59%)
Columbus - 2005-11 - 238 wins, 140 home (58.82%)
Dallas - 2005-11 - 305 wins, 166 home (54.42%)
Detroit - 2005-11 - 352 wins, 189 home (53.69%)
Edmonton - 2005-11 - 236 wins, 130 home (55.08%)
Florida - 2005-11 - 251 wins, 141 home (56.17%)
Los Angeles - 2005-11 - 267 wins, 146 home (54.68%)
Minnesota - 2005-11 - 282 wins, 164 home (58.15%)
Montreal - 2005-11 - 286 wins, 156 home (54.54%)
Nashville - 2005-11 - 320 wins, 181 home (56.56%)
New Jersey - 2005-11 - 326 wins, 178 home (54.60%)
New York (I) - 2005-11 - 235 wins, 134 home (57.02%)
New York (R) - 2005-11 - 304 wins, 162 home (53.28%)
Ottawa - 2005-11 - 296 wins, 160 home (54.05%)
Philadelphia - 2005-11 - 288 wins, 145 home (50.34%)
Phoenix - 2005-11 - 278 wins, 149 home (53.23%)
Pittsburgh - 2005-11 - 308 wins, 168 home (54.54%)
San Jose - 2005-11 - 339 wins, 182 home (53.68%)
St. Louis - 2005-11 - 256 wins, 146 home (57.03%)
Tampa Bay - 2005-11 - 260 wins, 150 home (57.69%)
Toronto - 2005-11 - 253 wins, 135 home (53.35%)
Vancouver - 2005-11 - 329 wins, 180 home (54.71%)
Washington - 2005-11 - 294 wins, 166 home (56.46%)
Winnipeg - 2005-11 - 266 wins, 145 home (54.51%)
And now the teams in order of percentages:

Code:
1. Calgary - 2005-11 - 295 wins, 174 home (58.98%)
2. Columbus - 2005-11 - 238 wins, 140 home (58.82%)
3. Minnesota - 2005-11 - 282 wins, 164 home (58.15%)
4. Anaheim - 2005-11 - 293 wins, 170 home  (58.02%) 
5. Tampa Bay - 2005-11 - 260 wins, 150 home (57.69%)
6. Carolina - 2005-11 - 288 wins, 165 home (57.29%)
7. St. Louis - 2005-11 - 256 wins, 146 home (57.03%)
8. New York (I) - 2005-11 - 235 wins, 134 home (57.02%)
9. Nashville - 2005-11 - 320 wins, 181 home (56.56%)
10. Washington - 2005-11 - 294 wins, 166 home (56.46%)
11. Chicago - 2005-11 - 284 wins, 160 home (56.33%)
12. Florida - 2005-11 - 251 wins, 141 home (56.17%)
13. Colorado - 2005-11 - 277 wins, 154 home (55.59%)
14. Edmonton - 2005-11 - 236 wins, 130 home (55.08%)
15. Vancouver - 2005-11 - 329 wins, 180 home (54.71%)
16. Los Angeles - 2005-11 - 267 wins, 146 home (54.68%)
17. New Jersey - 2005-11 - 326 wins, 178 home (54.60%)
18. Montreal - 2005-11 - 286 wins, 156 home (54.54%)
18. Pittsburgh - 2005-11 - 308 wins, 168 home (54.54%)
20. Winnipeg - 2005-11 - 266 wins, 145 home (54.51%)
21. Dallas - 2005-11 - 305 wins, 166 home (54.42%)
22. Ottawa - 2005-11 - 296 wins, 160 home (54.05%)
23. Detroit - 2005-11 - 352 wins, 189 home (53.69%)
24. San Jose - 2005-11 - 339 wins, 182 home (53.68%)
25. Toronto - 2005-11 - 253 wins, 135 home (53.35%)
26. New York (R) - 2005-11 - 304 wins, 162 home (53.28%)
27. Phoenix - 2005-11 - 278 wins, 149 home (53.23%)
28. Buffalo - 2005-11 - 312 wins, 165 home (52.88%)
29. Boston - 2005-11 - 292 wins, 148 home (50.68%)
30. Philadelphia - 2005-11 - 288 wins, 145 home (50.34%)
I've got each team's record for each individual year, as well as a "5-year mark" (right after the 09 season) to compare which teams change position every half-decade.

But mostly, I can't find any discernible pattern as to why Boston and Philly seem to always struggle to win at home (or have such success away from home, if you look at it another way) and why Calgary and Columbus have led the NHL in home success for the past couple years.

So I've come here to see if any of the HF users can explain why some teams struggle on home ice and some succeed.

Blueline Bomber is offline   Reply With Quote
Old
07-30-2012, 08:47 PM
  #2
Doctor No
Mod Supervisor
Retired?
 
Doctor No's Avatar
 
Join Date: Sep 2005
Posts: 24,308
vCash: 500
Interesting launching point!

One way to go about this would be to hypothesize about what factors could lead to a team having a high/low home ice advantage. Do teams of younger players do better on home ice (since veterans learn how to play on the road?)

I've always expected that Colorado would have a higher home-ice advantage than they do (because of the altitude).

Doctor No is offline   Reply With Quote
Old
07-30-2012, 08:54 PM
  #3
Lonewolfe2015
Registered User
 
Lonewolfe2015's Avatar
 
Join Date: Dec 2007
Country: United States
Posts: 12,062
vCash: 50
Quote:
Originally Posted by Taco MacArthur View Post
I've always expected that Colorado would have a higher home-ice advantage than they do (because of the altitude).
Try comparing our record to non-divisional opponents over the course of the same number of years versus divisional opponents. Either we simply suck against our own division beyond fathomable means or other teams have a little more noticeable difference due to the altitude.

I'd suspect that comparison will highlight your theory here at least a little better anyways, since divisional opponents are a little more accustomed to Colorado's altitude.

Lonewolfe2015 is offline   Reply With Quote
Old
07-30-2012, 09:34 PM
  #4
Playerwinner
Registered User
 
Join Date: Jan 2012
Posts: 792
vCash: 50
Its interesting that a lot of the lower teams have like a big home presence such as Philly, Boston, New York, Toronto, San Jose, with Montreal and Vancouver rounding out the lower half.

You should have Winnipeg, and Atlanta as separate entities as Winnipeg has only been home to the franchise for 1 season.

Playerwinner is offline   Reply With Quote
Old
07-30-2012, 09:36 PM
  #5
Blueline Bomber
Expectations - high
 
Blueline Bomber's Avatar
 
Join Date: Oct 2007
Posts: 23,000
vCash: 500
Quote:
Originally Posted by Taco MacArthur View Post
Interesting launching point!

One way to go about this would be to hypothesize about what factors could lead to a team having a high/low home ice advantage. Do teams of younger players do better on home ice (since veterans learn how to play on the road?)

I've always expected that Colorado would have a higher home-ice advantage than they do (because of the altitude).
I've heard a couple theories. The one I'm going to take a look at next would be powerplay opportunities. I remember reading somewhere that officials are influenced (sometimes consciously, sometimes unconsciously) by the home crowd to give the calls in favor of the home team.

If these teams are getting a much bigger advantage at home compared to away (or vise-versa), it would explain why they're having more/less success at home.

Quote:
Originally Posted by Playerwinner View Post
Its interesting that a lot of the lower teams have like a big home presence such as Philly, Boston, New York, Toronto, San Jose, with Montreal and Vancouver rounding out the lower half.

You should have Winnipeg, and Atlanta as separate entities as Winnipeg has only been home to the franchise for 1 season.
Once Winnipeg establishes some more seasons to give a bigger sample size, I plan on doing just that. Winnipeg's home winning percentage this year was much higher than any of the other Atlanta years I've researched.

Blueline Bomber is offline   Reply With Quote
Old
07-30-2012, 09:38 PM
  #6
Czech Your Math
Registered User
 
Czech Your Math's Avatar
 
Join Date: Jan 2006
Location: bohemia
Country: Czech_ Republic
Posts: 3,715
vCash: 500
Quote:
Originally Posted by Lonewolfe2015 View Post
Try comparing our record to non-divisional opponents over the course of the same number of years versus divisional opponents. Either we simply suck against our own division beyond fathomable means or other teams have a little more noticeable difference due to the altitude.

I'd suspect that comparison will highlight your theory here at least a little better anyways, since divisional opponents are a little more accustomed to Colorado's altitude.
This seems like a good idea.

The way home/road splits have been calculated seems good. The only complicating factor is that since the lockout, there are shootout W/L.

I can think of a few factors that could affect difference in home/road records:

- coaching: coaches that utilize matchups more should have larger H/R splits

- depth of talent: teams with skater talent more evenly distributed should have smaller H/R splits than those which are imbalanced and more dependent on favorable matchups

- goaltending: teams with a very large proportion of team value at goalie should have smaller H/R splits

- travel: teams in cities which are particularly far away from other NHL cities should have larger H/R splits

- team dynamic: teams which are particularly close knit may have smaller H/R splits, as may those that are particularly "unfocused" at home (e.g. we're home... it's party time!)

- team quality/refereeing: bad teams may be favored more by refs at home, while good teams would seem less likely to get (more like need) such advantages to stay competitive

- team style: some teams may play a more entertaining style at home, while playing a more disciplined style on the road, which would decrease their H/R splits


Last edited by Czech Your Math: 07-30-2012 at 09:50 PM.
Czech Your Math is offline   Reply With Quote
Old
07-31-2012, 12:03 AM
  #7
Czech Your Math
Registered User
 
Czech Your Math's Avatar
 
Join Date: Jan 2006
Location: bohemia
Country: Czech_ Republic
Posts: 3,715
vCash: 500
Quote:
Originally Posted by Playerwinner View Post
Its interesting that a lot of the lower teams have like a big home presence such as Philly, Boston, New York, Toronto, San Jose, with Montreal and Vancouver rounding out the lower half.
That's an interesting observation. Perhaps the road teams play better when there's more of a big game atmosphere? It's more difficult for the home team to get pumped for 41 games per season.

Definitely looks like a lot of weaker, smaller market teams at the top and stronger, bigger market teams at the bottom. It seems like refereeing and travel would be other factors that might help explain that.

There also might be a tendency for weaker teams to say "screw it" on the road, once they know they are very unlikely to make the playoffs, but still give it more of a maximum effort at home.

Czech Your Math is offline   Reply With Quote
Old
07-31-2012, 12:44 AM
  #8
Greg02
Registered User
 
Greg02's Avatar
 
Join Date: Jun 2009
Posts: 1,876
vCash: 500
I hypothesized that this might be a result of the matchup advantages given to the home team being more significant for weaker teams. If this was the case, weaker teams would primarily get their wins at home because they couldn't win without things like the faceoff advantage, where the stronger teams were able to win both at home and on the road so it becomes closer to 50-50.

A brief glance at past years' standings seems to disagree with my little theory. Two fairly illustrative examples (which are obviously not proof) are when Philly came in last in the East they won more on the road, and when Columbus made the playoffs they won far more often at home.

Greg02 is offline   Reply With Quote
Old
07-31-2012, 01:14 AM
  #9
Blueline Bomber
Expectations - high
 
Blueline Bomber's Avatar
 
Join Date: Oct 2007
Posts: 23,000
vCash: 500
For those interested, here's how the rankings looked after the 09 season:

Code:
1. Columbus - 2005-09 - 175 wins, 106 home (60.57%)
2. Minnesota - 2005-09 - 208 wins, 125 home (60.09%)
3. Calgary - 2005-09 - 217 wins, 128 home (58.98%)
4. New York (I) - 2005-09 - 171 wins, 100 home (58.47%)
5. Anaheim - 2005-09 - 212 wins, 123 home (58.01%)
6. Nashville - 2005-09 - 228 wins, 131 home (57.45%)
7. Carolina - 2005-09 - 215 wins, 123 home (57.20%)
8. Florida - 2005-09 - 183 wins, 104 home (56.83%)
9. Tampa Bay - 2005-09 - 176 wins, 100 home (56.81%)
10. Washington - 2005-09 - 204 wins, 115 home (56.37%)
11. Colorado - 2005-09 - 206 wins, 116 home (56.31%)
12. Vancouver - 2005-09 - 224 wins, 126 home (56.25%)
13. Chicago - 2005-09 - 195 wins, 109 home (55.89%)
14. Ottawa - 2005-09 - 223 wins, 124 home (55.60%)
15. Dallas - 2005-09 - 221 wins, 122 home (55.20%)
16. New York (R) - 2005-09 - 209 wins, 115 home (55.02%)
17. New Jersey - 2005-09 - 240 wins, 132 home (55.00%)
18. Montreal - 2005-09 - 211 wins, 116 home (54.97%)
19. Phoenix - 2005-09 - 193 wins, 106 home (54.92%)
20. Pittsburgh - 2005-09 - 208 wins, 114 home (54.80%)
21. Edmonton - 2005-09 - 179 wins, 98 home (54.74%)
22. Los Angeles - 2005-09 - 181 wins, 99 home (54.69%)	
22. Toronto - 2005-09 - 181 wins, 99 home (54.69%)
24. Atlanta - 2005-09 - 195 wins, 105 home (53.84%)
24. St. Louis - 2005-09 - 169 wins, 91 home (53.84%)
26. Buffalo - 2005-09 - 230 wins, 123 home (53.47%)
27. Detroit - 2005-09 - 257 wins, 137 home (53.30%)
28. San Jose - 2005-09 - 248 wins, 131 home (52.82%)
29. Philadelphia - 2005-09 - 194 wins, 101 home (52.06%)
30. Boston - 2005-09 - 197 wins, 102 home (51.77%)
Of note:

- The percentage, for the vast majority of the teams, has dropped since then. Probably has more to do with the sample size increasing more than anything though.
- Ottawa and New York appear to have dropped the most in terms of placement and percentage.
- St. Louis has had the biggest increase, around 3%, which is ridiculous. It does lend credence to the idea that the coaching and the system may have a big effect on a H/R split.

Blueline Bomber is offline   Reply With Quote
Old
07-31-2012, 04:19 AM
  #10
Jevo
Registered User
 
Jevo's Avatar
 
Join Date: Oct 2010
Country: Denmark
Posts: 2,517
vCash: 500
Quote:
Originally Posted by Blueline Bomber View Post
I've heard a couple theories. The one I'm going to take a look at next would be powerplay opportunities. I remember reading somewhere that officials are influenced (sometimes consciously, sometimes unconsciously) by the home crowd to give the calls in favor of the home team.

If these teams are getting a much bigger advantage at home compared to away (or vise-versa), it would explain why they're having more/less success at home.
The book Scorecasting, which is a very interesting read, concludes that the biggest part of home-advantage is ref-bias, because the refs subconsciously are less likely to make a judgement call against the home team because they don't want to feel the wrath of the home crowd. This was consistent between several sports, I remember them analyzing soccer, baseball, basketball and hockey in the book. There are of course also other factors, especially in hockey where there are rules that favour the home team, and the fact that you usually play many games in many cities in a short amount of time during a road trip.
A very interesting fact that they pointed out in the book, is that in situations where it's all up to the players and the refs have no influence, like free throws in basketball and penalty shots in hockey. The players perform equally well at home and away, indicating that players don't perform differently, but there are other factors like the ones mentioned above that drives home-advantage.

Jevo is offline   Reply With Quote
Old
07-31-2012, 09:18 AM
  #11
Fourier
Registered User
 
Fourier's Avatar
 
Join Date: Dec 2006
Location: Waterloo Ontario
Country: Canada
Posts: 8,857
vCash: 500
Quote:
Originally Posted by Czech Your Math View Post
This seems like a good idea.

The way home/road splits have been calculated seems good. The only complicating factor is that since the lockout, there are shootout W/L.

I can think of a few factors that could affect difference in home/road records:

- coaching: coaches that utilize matchups more should have larger H/R splits

- depth of talent: teams with skater talent more evenly distributed should have smaller H/R splits than those which are imbalanced and more dependent on favorable matchups

- goaltending: teams with a very large proportion of team value at goalie should have smaller H/R splits

- travel: teams in cities which are particularly far away from other NHL cities should have larger H/R splits

- team dynamic: teams which are particularly close knit may have smaller H/R splits, as may those that are particularly "unfocused" at home (e.g. we're home... it's party time!)

- team quality/refereeing: bad teams may be favored more by refs at home, while good teams would seem less likely to get (more like need) such advantages to stay competitive

- team style: some teams may play a more entertaining style at home, while playing a more disciplined style on the road, which would decrease their H/R splits
It would be interesting to understand the travel dynamic on actual performance.

Fourier is offline   Reply With Quote
Old
07-31-2012, 11:15 AM
  #12
seventieslord
Moderator
 
seventieslord's Avatar
 
Join Date: Mar 2006
Location: Regina, SK
Country: Canada
Posts: 24,953
vCash: 500
Quote:
Originally Posted by Jevo View Post
The book Scorecasting, which is a very interesting read, concludes that the biggest part of home-advantage is ref-bias, because the refs subconsciously are less likely to make a judgement call against the home team because they don't want to feel the wrath of the home crowd. This was consistent between several sports, I remember them analyzing soccer, baseball, basketball and hockey in the book. There are of course also other factors, especially in hockey where there are rules that favour the home team, and the fact that you usually play many games in many cities in a short amount of time during a road trip.
A very interesting fact that they pointed out in the book, is that in situations where it's all up to the players and the refs have no influence, like free throws in basketball and penalty shots in hockey. The players perform equally well at home and away, indicating that players don't perform differently, but there are other factors like the ones mentioned above that drives home-advantage.
That is a good point. I've always suspected this myself.

There is more to it though. Being able to get the last line change, the slight faceoff advantage, and travel fatigue are three factors (maybe not after just one night of traveling, but most road games tend to be part of a "road trip" so it adds up)

And still, there is the "energy of the home crowd" which is at least something. Maybe it doesn't show up in "free throw" or "penalty shot" situations, but in situations where that added burst of intensity might make a difference, I think it does.

Those four things probably account for 80% of that apparent 5% advantage teams tend to have at home compared to being on the road.

seventieslord is offline   Reply With Quote
Old
07-31-2012, 12:06 PM
  #13
KINGS17
Smartest in the Room
 
KINGS17's Avatar
 
Join Date: Apr 2006
Country: United States
Posts: 15,975
vCash: 500
Kings playoff opponents in 2012...

What home ice advantage?

KINGS17 is offline   Reply With Quote
Old
07-31-2012, 12:18 PM
  #14
Cliffy1814
Registered User
 
Join Date: Nov 2011
Posts: 652
vCash: 500
This appears to measure teams home success vs. their road success. That means the percentages are somewhat skewed based on their road performance i.e. Columbus.

I'd love to look home and away sperately and also look at it from a "points earned" and "points relinquished" standpoint since the shootout and three point game have become a big part of the equation.

For instance: Team "A" earned 47 points at home and visiting team gained 40 points (+7). On the road, Team "A" earned 38 points and gave up 40 (-2). They are a +9 at home.

Not hating at all. I appreciate the poster who did the work getting this up here. I just dont think this tells the whole story.

Cliffy1814 is offline   Reply With Quote
Old
07-31-2012, 12:25 PM
  #15
seventieslord
Moderator
 
seventieslord's Avatar
 
Join Date: Mar 2006
Location: Regina, SK
Country: Canada
Posts: 24,953
vCash: 500
Quote:
Originally Posted by KINGS17 View Post
Kings playoff opponents in 2012...

What home ice advantage?
Blizzard in winter... What global warming?

seventieslord is offline   Reply With Quote
Old
07-31-2012, 01:10 PM
  #16
WorkingOvertime
Registered User
 
WorkingOvertime's Avatar
 
Join Date: Apr 2011
Posts: 419
vCash: 500
To study home ice advantage, we need to have a starting point for the analysis.

First, what is it that we are trying to measure? Are we solely measuring the cumulative effect of travel, first change, sleeping at home, etc., or do we want deconstruct the cumulative effect?

If we are deconstructing the cumulative effect, it may be worthwhile to look at some of the 'abnormal' cases first to test various hypotheses. This would include one-game road trips, teams within a close distance of each other (day-trips), teams at higher altitudes, etc. to see how these situations compare to the 'normal' situations.

From thinking on this for just a minute, it seems most effective to focus on removing travel distance, trip length, time between games, etc. from the data to isolate a better measure of home-ice advantage.

As a starting point, it would be interesting if the data could be separated by trip length, position on trip, etc.

WorkingOvertime is offline   Reply With Quote
Old
07-31-2012, 01:22 PM
  #17
Guest
Registered User
 
Join Date: Feb 2003
Posts: 5,266
vCash: 500
I'm not sure comparing home ice to road games is quite what I would do. The big problem I see with it is that it's comparing home and road games at the basis of being equal and that a teams home ice advantage would explain any gap. As others have suggested some teams tend to play better on the road, so that would artificially lower their ranking on home ice for being a good road team. I think a more fair comparison, while still flaws would be comparing the total regulation/OT winning percentage for the season (home+road) and giving the variance for just the home winning percentage.

For example a team that has a great record overall like San Jose or Detroit during that sample would likely have a great home record. It would suggest that home ice advantage has little to do with their success. However say a team like Columbus did poorly overall in the season but had a better winning percentage compared at home, that would show a more significant home ice advantage although it's still a bit skewed.

I would also love to see a breakdown of various stats on home ice. Goals for, goals against, PP & PK efficiency, faceoffs %, etc. Those stats are likely available out there and may put some more emphasis on why teams do better/worse at home.

Guest is offline   Reply With Quote
Old
07-31-2012, 01:35 PM
  #18
overpass
Registered User
 
Join Date: Jun 2007
Posts: 3,618
vCash: 500
Quote:
Originally Posted by Jevo View Post
The book Scorecasting, which is a very interesting read, concludes that the biggest part of home-advantage is ref-bias, because the refs subconsciously are less likely to make a judgement call against the home team because they don't want to feel the wrath of the home crowd. This was consistent between several sports, I remember them analyzing soccer, baseball, basketball and hockey in the book. There are of course also other factors, especially in hockey where there are rules that favour the home team, and the fact that you usually play many games in many cities in a short amount of time during a road trip.
A very interesting fact that they pointed out in the book, is that in situations where it's all up to the players and the refs have no influence, like free throws in basketball and penalty shots in hockey. The players perform equally well at home and away, indicating that players don't perform differently, but there are other factors like the ones mentioned above that drives home-advantage.
Phil Birnbaum's take on Scorecasting's argument:
Quote:
Now, let's look at number 18, the hockey case. The authors argue that HFA is caused almost entirely by penalties. If that's the case, then you'd expect home and visiting teams to have similar numbers at equal strength.

They do not. The NHL.com website has home/road goal breakdowns. Here they are for the 2008-09 season, averaged by team:

Even strength... 124-110 (home advantage 12.5%)
Power play...... 35-30 (home advantage 15.1%)
Shorthanded..... 4-4 (home advantage 1.0%)

There's almost as large an advantage at even strength as there is on the power play. Admittedly, the extra power play boost is probably caused by more penalties, as the authors say, but the overall contribution of the extra penalties seems to be pretty small.

Just to make sure it wasn't a fluke, I ran the same numbers for 2009-10:

Even strength... 121-106 (home advantage, 13.9%)
Power play...... 30-25 (home advantage, 21.0%)
Shorthanded..... 4-3 (home advantage, 32.9%)

A bit more extreme in favor of power play. But how do you explain the sizeable advantage for home teams at even strength? One possible explanation is that visiting teams have to play an overcautious game, to avoid being penalized by biased referees. But for a 13.9% disadvantage, that caution would have to be way out of line, wouldn't it?


------

Both of these examples -- and, by the way, they're the only two I checked -- cast doubt on the authors' hypothesis that HFA is almost all refereeing. I have never disagreed that *some* of it might be refereeing, but there's obviously a lot more going on.

And I have to say that the authors have indeed provided a blueprint for how this kind of research should go -- try to break down performance into its constituent parts, and check those.

If there's no home advantage in foul shooting, why not? If there's no HFA in hockey shootouts, why not? If we get a list of areas with high HFA, and a list of areas with low HFA, we can maybe start narrowing down what the causes might be.

overpass is offline   Reply With Quote
Old
07-31-2012, 06:12 PM
  #19
Jevo
Registered User
 
Jevo's Avatar
 
Join Date: Oct 2010
Country: Denmark
Posts: 2,517
vCash: 500
Quote:
Originally Posted by overpass View Post
I hadn't seen that before. Very interesting, I would have thought there would still be a slight advantage at even-strength, especially in hockey where certain rules gives the home team an advantage, but I didn't think it would be as big as it was in those two seasons.

Jevo is offline   Reply With Quote
Old
08-01-2012, 01:48 AM
  #20
Master_Of_Districts
Registered User
 
Master_Of_Districts's Avatar
 
Join Date: Apr 2007
Location: Black Ruthenia
Country: Belarus
Posts: 1,745
vCash: 500
Occam's razor provides that the source of the team-to-team variation is overwhelmingly randomness.

I predict that every team would obtain roughly 55% of its wins at home over a sufficiently large sample. Perhaps counterintuitively, seven NHL regular seasons is not a sufficiently large sample. Who knew?

I created a fake league on microsoft excel in which the home ice advantage for every team in every game is exactly the same at 5%. I then simulated a whole whack of seasons. In looking at seven season increments, the average standard deviation among the teams in % of wins obtained at home is 0.0205.

The actual standard deviation in the same statistic, in the seven season stretch from 2005-06 to 2011-12, is 0.217.

In other words, my fake league, where no team is uniquely advantaged at home, and where differences between teams in % of wins obtained at home are entirely due to random variation, looks an awful lot like the real NHL.

Master_Of_Districts is offline   Reply With Quote
Old
08-01-2012, 11:28 AM
  #21
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 2,795
vCash: 500
Quote:
Originally Posted by Master_Of_Districts View Post
In other words, my fake league, where no team is uniquely advantaged at home, and where differences between teams in % of wins obtained at home are entirely due to random variation, looks an awful lot like the real NHL.
The last thing I ever published at my old site was about this sort of thing. It showed that even if the results of games are completely random, some teams can appear to be much better than others. The article builds narratives out of randomness.

Which is not to say that NHL results are completely random; far from it. Just that the importance of variance is so often understated, or ignored completely in favour of one team "wanting it more."

Iain Fyffe is offline   Reply With Quote
Old
08-01-2012, 10:44 PM
  #22
Hammer Time
Registered User
 
Hammer Time's Avatar
 
Join Date: May 2011
Country: Canada
Posts: 3,736
vCash: 500
Quote:
Originally Posted by Jevo View Post
The book Scorecasting, which is a very interesting read, concludes that the biggest part of home-advantage is ref-bias, because the refs subconsciously are less likely to make a judgement call against the home team because they don't want to feel the wrath of the home crowd. This was consistent between several sports, I remember them analyzing soccer, baseball, basketball and hockey in the book. There are of course also other factors, especially in hockey where there are rules that favour the home team, and the fact that you usually play many games in many cities in a short amount of time during a road trip.
A very interesting fact that they pointed out in the book, is that in situations where it's all up to the players and the refs have no influence, like free throws in basketball and penalty shots in hockey. The players perform equally well at home and away, indicating that players don't perform differently, but there are other factors like the ones mentioned above that drives home-advantage.
The ref influence is eliminated, but so is the coaching influence. Since hockey has a structural advantage given to the home team in terms of matchups, it makes sense that home-ice advantage would present itself in regular play when both teams are lining up for a faceoff, but not on penalty shots. That isn't exactly evidence of refereeing bias.

Refereeing bias most likely does exist, but I don't think it explains all of the home-ice advantage effect.

Hammer Time is offline   Reply With Quote
Old
08-02-2012, 06:53 AM
  #23
Czech Your Math
Registered User
 
Czech Your Math's Avatar
 
Join Date: Jan 2006
Location: bohemia
Country: Czech_ Republic
Posts: 3,715
vCash: 500
Quote:
Originally Posted by Meteor View Post
The ref influence is eliminated, but so is the coaching influence. Since hockey has a structural advantage given to the home team in terms of matchups, it makes sense that home-ice advantage would present itself in regular play when both teams are lining up for a faceoff, but not on penalty shots. That isn't exactly evidence of refereeing bias.

Refereeing bias most likely does exist, but I don't think it explains all of the home-ice advantage effect.
Good point. I did quick calculations based on the data posted for '09 & '10. Using the league gpg and PP efficiency, would guess that ~1/3 of the home ice advantage is due to increased PP (ref bias).

It might be interesting to look at each of the best & worst 5-10 teams each season and see what each group's home ice advantage is.

Czech Your Math is offline   Reply With Quote
Old
08-02-2012, 09:08 AM
  #24
stayinalive
Registered User
 
stayinalive's Avatar
 
Join Date: Jul 2012
Location: Bielefeld
Country: Germany
Posts: 1,474
vCash: 500
In my opinion this is all about succes.

Teams like CBJ, NYI, MIN, TBL who where bad for a lot of years, have problems to win on the road, while good teams like Boston, Philly, Detroit, SJS, NYR know how to win on the road

stayinalive is offline   Reply With Quote
Old
08-02-2012, 09:38 AM
  #25
Blueline Bomber
Expectations - high
 
Blueline Bomber's Avatar
 
Join Date: Oct 2007
Posts: 23,000
vCash: 500
Quote:
Originally Posted by stayinalive View Post
In my opinion this is all about succes.

Teams like CBJ, NYI, MIN, TBL who where bad for a lot of years, have problems to win on the road, while good teams like Boston, Philly, Detroit, SJS, NYR know how to win on the road
But then there are outliers among those teams:

Phoenix and Toronto really haven't been on the same level as Boston, Philly, Detroit, etc. in terms of success, yet have about the same percentage.

Nashville, St. Louis and Calgary really haven't been nearly as bad as CBJ or the Islanders, but yet have really good home records.

Blueline Bomber 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 10:06 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.