« I loved this question, and answer | Main | The 100 best jazz albums? »
Home court advantage in basketball
We all feel the Celtic ouch and perhaps some of us delight in it. Matt writes:
Kevin Drum notes two smart responses to the question of why home court advantage is so big, with one hypothesis pointing to the refs and another pointing to the idea that there are actually lots of differences from arena-to-arena.
Of course if the arena is the difference you would expect shooting guards, who need a good feel for the lights and angles of the basket, to have a bigger relative advantage at home than do the dunking big men. That should be easy enough to test. And maybe a look at Lakers-Clippers or Nets-Knicks history can clear up the importance of arena by holding geographic area constant.
I wonder if a third component of home court advantage has to do with sleep. People sleep better at home, if only because they don't have to go to such great lengths to get sex. I recall reading that Larry Bird became a truly great player only once he...um...calmed down a bit.
Posted by Tyler Cowen on May 13, 2008 at 01:50 PM in Sports | Permalink
Comments
Hmm.. maybe you should allow spouses/girlfriends to accompany the team?
Posted by: PLW at May 13, 2008 2:31:36 PM
In a sport where effort can make such a big difference (just how hard are they running to get to their defensive or offensive positions?) the emotion and encouragement of the crowd might also have an effect.
Posted by: RMH at May 13, 2008 2:31:36 PM
Vanderbilt's Memorial Gym is an interesting case for this. As far as I know, it's the only major basketball arena currently in use that puts the coaching benches at the ends of the court rather than the sides. It's a clever way to give the team a home court advantage: they split their season 50/50 at Memorial and on normal courts, but visiting teams are only exposed to the weird arrangement about once a year.
Posted by: Jacob Grier at May 13, 2008 2:37:20 PM
Boos/cheers surely have a direct effect too.
Another testable hypothesis: do sold out / loud crowds increase the home team's win percentage?
Posted by: sd at May 13, 2008 2:57:55 PM
The difference is due to evolution - home players have a higher level of testosterone. This has been measured in soccer matches in England. Players spit into cups just before the match.
From an evolutionary game theory perspective, this is a stable strategy of being a hawk at home and a dove away - leads to fewer confrontations and genetic collusion to progress the species... (or something like that if I remember right)
Posted by: Nate at May 13, 2008 3:03:34 PM
Home or road, I don't think NBA players have to go to great lengths to... calm down...
Posted by: Roger at May 13, 2008 3:07:11 PM
another reason why homecourt advantage is important:
http://sports.aol.com/fanhouse/2008/05/13/did-the-hawks-go-out-boozing-the-night-before-game-7-in-boston/
"Word is, Boston entertainment prince Patrick Lyons might have played a role in the Celtics' blowout Game 7 win over the Hawks a week ago that vaulted them to the second round. The Hawks stayed at the Liberty Hotel May 3 in Cambridge before playing the Celtics. A league source said Lyons, who operates the Alibi Lounge in the hotel, instructed his bar staff to give anyone affiliated with the Hawks a double shot in any alcoholic beverage they ordered."
Posted by: jumper at May 13, 2008 3:33:43 PM
Casey Stengel was a supporter of the sex theory. Asked if sex the night before a game affected players' performance, he said,
"It's not the sex. It's the staying up all night chasing it."
Posted by: Bernard Yomtov at May 13, 2008 3:34:01 PM
I (and others in the field of sabermetrics) have argued that HFA is higher in basketball becaust he game is "longer". There are about 100 possessions for each side, and points are scored frequently. This is long enough for one team's talent advantage to come out much more frequently than in basball, where their is much more variance of outcomes and the game is "shorter".
Put another way, if basketball games were only half as long, the HFA would be much smaller. If thy were twice as long, HFA would be larger.
HFA is a function of the variance of outcomes per "atom" of game play, and the number of "atoms" in a game.
I would argue that there's nothing special about home court advantage in basketball vs. other sports, except in terms of the "length" of the game.
My own argument for this is on page 3 here.
Posted by: Phil at May 13, 2008 3:56:17 PM
I should clarify: my argument applies to why NBA home advantage is larger than in other sports, not why it exists in the first place.
Also, the argument in my link above is for competitive balance, but the argument is the same. The longer the game, the more likely the better team wins. And HFA makes the home team better, which and so much more likely that the home (now better) team wins.
Posted by: Phil at May 13, 2008 4:04:53 PM
Not sure how isolating out Clippers/Lakers games goes when they play in the same arena, though (both play their home games at the Staples Center).
Posted by: Michael Fisk at May 13, 2008 4:22:47 PM
As someone who has made a good bit of money betting on sports, the biggest home court advantage in the NBA
comes when the road team has to travel to a different time zone on short rest, or is on an extended road
trip through an unfamiliar time zone.
Posted by: Jay at May 13, 2008 4:34:41 PM
Phil: Baseball is a completely different beast. In baseball the road team is at a huge disadvantage as
it has to bat first.
Posted by: Jay at May 13, 2008 4:38:39 PM
_I recall reading that Larry Bird became a truly great player only once he...um...calmed down a bit._
Narrative fallacy, anyone? Causation problems?
I expect better out of you, Tyler.
Posted by: porteno at May 13, 2008 4:53:07 PM
I heard there was a baseball player who had five young children at home and when he was on the road his batting average went up about 100 points.
Posted by: jack at May 13, 2008 6:20:42 PM
Anyone who thinks that NBA players have to go to any sort of great lengths to get sex on the road isn't paying attention. . .
Posted by: Matthew C. at May 13, 2008 7:07:38 PM
It's now deep into playoff season in the NBA, which means that opponents are much closer to each other in true ability and talent. Other factors outside of team talent, such as home court advantage, would mean the difference in games more often, and therefore appear to be especially strong. The effect of the home court is the same, but the teams are so closely matched that other factors begin to dominate--home court
Also, Phil is absolutely right about HCA being based on the number of iterations. The NBA game is 8 minutes longer than college basketball, and it has a quicker shot clock. This leads to many more possessions per team. Each possession can be thought of as just a mini-game, in which HCA has a small effect. The greater the number of possessions, the greater the cumulative effect of the advantage. This is based on research specific to basketball, not baseball.
Posted by: Brian at May 13, 2008 10:45:04 PM
This Celtics team relies on perimeter shooting, and it's top shooters (Pierce and Rondo) shot over 40% in the regular season. However, they are barely shooting 30% in the playoffs, and this is reflected in the home/away splits.
Whatever is killing the relaxed mindset, confidence, flow, and momentum of perimeter shooting on the road will also be the demise of this team in the playoffs.
Posted by: caveat bettor at May 13, 2008 10:54:34 PM
1. Testosterone levels and effort on defense. This one seems strong.
2. Foul shots and referees: See, e.g. the Utah Jazz, who are highly dependent on free throws and not fouling, and have huge home/road splits. Referees are affected, they are only human. Also seems strong.
3. Crowd noise and effort: see #3, but slightly different. This seems like it should affect bench players more.
4. Length of game and variance: I have a problem with that theory, which is football. Home-field advantage is incredibly strong in football, but games only have about as many possessions total as a baseball game has outs per side. Clearly you're right that it matters, but still we observe a much stronger home field advantage over time in basketball and football than we do in baseball. Going into a game between two even teams I believe WPA (win probability added) starts with the home team at a .55 chance to win.
5.
Posted by: Joel W at May 14, 2008 12:11:18 AM
Joel - I'm not sure I agree that home-field advantage is 'incredibly strong' in football. I'll have to look up the stats, but if I recall correctly, the winning % of home teams in conference championship games (typically sellout crowds between evenly matched teams) is about 50%.
This was a good attempt at quantifying the effects of HFA, and they wound up with a standard deviation twice the average effect.
Posted by: Independent George at May 14, 2008 7:44:59 AM
That link is pretty clear that the effect is strong. HFA is generally considered worth 3 points in football.
Posted by: Cliff at May 14, 2008 9:45:56 AM
"I recall reading that Larry Bird became a truly great player only once he...um...calmed down a bit." No. Larry Bird was born a great player. Go Celtics!
Posted by: Ben at May 14, 2008 9:53:54 AM
One of the economics profs who taught me at university spends a lot of time studying sport, including statistically determining the existence or absence of referee bias. The evidence is pretty overwhelming that referees are biased in favour of the home team, because they're influenced by the crowd. One study showed referees identical gameplay situations and asked them whether a foul was committed. Some saw it with the sound off, some with the sound on. Referees tended to award fouls more often when they could hear the crowd, even when it wasn't there in the room. Another study found that referees in the German Bundesliga were signficantly less biased than those in the Premier League; this was because German football stadiums tend to have running tracks between the stands and the pitch, which means that the crowd is further away from the ref.
Of course, basketball courts are smaller and have fewer spectators than football stadiums, but the effect must surely exist even if it's smaller.
Posted by: Sam B at May 14, 2008 11:01:13 AM
San Antonio in game four of the series (a home game for San Antonio) shot 12 more free throws than New Orleans. In game give, New Orleans (the home team) shot 15 more free throws than San Antonio. Does anyone really believe that in three days that the swing would really be 27 free throws unless the referees were not calling the game the same.
P.S. one of the reason for more upsets in the NCAA is that a team that usually gets really good calls at home like Kansas or Duke will get a set of referees who does not show preference.
Posted by: superdestroyer at May 14, 2008 1:14:54 PM
Here's a link to an article summarizing the testosterone theory of home advantage:
http://www.newscientist.com/article/dn2050-testosterone-surge-linked-to-sports-home-advantage.html
Posted by: Nate at May 14, 2008 2:24:55 PM
Tyler,
I'm sure that your general point is correct -- players get to sleep earlier when they are at home with their wives/girlfriends than when they are out on the town in a strange city with unlimited temptation -- but the Larry Bird example is all wrong. Bird was an awesome college player, and a great pro from his first season: rookie of the year, 4th in the MVP, NBA championship.
Jayson Williams would be an example of a player who didn't attain his potential until after Charles Barkley left his team so he wasn't out with Charles all night. Barkley doesn't need sleep, but Williams did.
Posted by: Steve Sailer at May 14, 2008 11:15:26 PM
The lakers beat the Utah Jazz in game five of their series while shooting 14 more free throws than the Jazz. Of course, the game was in Los Angeles.
Want to bet that the Jazz will shoot at least ten more free throws in game six when the game is in Salt Lake City.
The NBA is about two steps away from being as scripted as Professional Wrestling. Different rules for all stars versus role players. Different rules at home games than away games. Different rules in the first quarter than the fourth quarters. And the last, manipulation that certain cities get the best draft picks.
Posted by: superdestroyer at May 15, 2008 8:31:37 AM
Losing teams tend to foul more, so it's hard to separate out the effect of refereeing. Plus, during the regular season, the Lakers went to the line 3 more times on average than the Jazz and sent their opponents to the line almost 6 fewer times. Considering they had very similar WL records and scoring differentials (not sure about their pace stats), we should witness a difference of about 9 FT/game. So far it's been LA +6 on average.
superdestroyer,
That's true for every sport. Greg Maddux gets more leeway on the strike zone, top corners get away with mugging receivers, Brett Favre gets fewer intentional grounding calls. The calls in the 4th quarter of the NFL are more lax than in the 1st and the playoffs are called less stringently in the playoffs.
As for the draft, why did Yao become a Rocket instead of a Knick, why did Chris Paul get past the NBA favorite cities of Atlanta and Salt Lake City? Why was Lebron drafted by Cleveland? What about Greg Oden and Kevin Durant? I can't believe you still believe the Patrick Ewing conspiracy theory.
Posted by: Mo at May 15, 2008 11:17:12 AM
Give you a 中古車|二手車,which one you will choice,i like all.
Posted by: 租車 at Oct 2, 2008 10:20:08 PM
