Matthew Yglesias has posted a comment on Competitive Balance in the NBA. His comment echoes one of the stories we tell in The Wages of Wins – specifically that the NBA suffers from a “Short Supply of Tall People.” In his comment, Yglesias links to The Wages of Wins Journal. Unfortunately, I do not think we have specifically gone over this particular story at our blog. I did post a comment on baseball that mentioned the story, but there is no post directly examining the competitive balance in the NBA.
Rather than write something, I thought I would just offer an excerpt from the book that details the evidence for our argument (for further evidence one is referred to our academic articles on this subject).
from pp. 61-62 of The Wages of Wins
At the other end of the extreme lies the NBA and the ABA. Each of these leagues is less competitive than any other league considered. At first this seems odd, since basketball is possibly the second most popular sport in the world. The problem with basketball, though, is not the population of people who are interested in playing. To play basketball you can’t just be interested. To play basketball at the highest level you generally have to be tall.
The average player in the NBA is 6’7”. The average American male stands about 5’10”. If we look at player rosters from ten consecutive NBA seasons beginning with the 1994–95 campaign, we find fewer than ten players logging significant playing time and standing 5’10” or smaller.34 At the other extreme we would note that only about 2% of adult males in America are taller than 6’3”, and a miniscule number are 6’10” or taller. Yet nearly 30% of the players employed by the NBA over the ten years we examined were at least 6’10”. So although basketball may be popular, to play at the NBA level you first must be tall. Unfortunately, as people in the NBA often note, “you can’t teach height.” As a result, the players the NBA requires are in short supply.
In the “Short Supply of Tall People” we compared competitive balance both before and after 1990. In the AL, NL, NFL, and NHL, competitive balance in the post-1990 era had improved. In the NBA, though, the 1990s was the least competitive decade in its history. Why? The NBA keeps adding teams but the short supply of tall people—yes, we love saying this—persists. There is only one Shaquille O’Neal. No matter what policies the NBA adopts, they cannot manufacture quality big men for every team. This rule doesn’t just apply to players like Shaq, but to smaller players like Michael Jordan. There is not an abundance of 6’6” people in the world either, and when you ask for a person of this height with Jordan’s skills, you are not going to unearth many candidates.
Given the supply of talent the NBA employs, there is very little the league can do to achieve the levels of competitive balance we see in soccer or American football. When your league depends upon a small population, the number of athletes close to the biomechanical limit will also be small. Consequently, games will often be between the good and the not so good.
– DJ
Wages of Wins Journal 
Tom Hurka
November 20, 2006
I’ve long wanted to ask you this — well, since I read The Wages of Wins over the summer.
In the book you attribute differences in competitive balance between sports to facts about the supply of talented players. But a striking fact about the table on p. 61 is that the ordering of sports by competitive balance tracks the ordering by number of scores in an average game.
Soccer is the most balanced sport and also the lowest-scoring — just a couple of goals per game. Next come football and hockey, and my seat-of-the-pants sense is that there are a few more goals in an average hockey game than touchdowns in an average football game. Then comes baseball — more runs per game than goals per hockey game — and finally basketball, with way more scores per game than any other sport.
This suggests another hypothesis: the fewer scores in a game, the more opportunity there is for the outcome to be determined by luck, and since luck is neutral between the two sides, the more competitive balance there will be. On this view basketball is the least balanced sport because, with 50 or so baskets per team per game, it’s less likely that luck will cancel out the skill differences between teams.
(This would also explain why there’s more consistency in basketball performance than in, say, football QB performance. Basketball statistics from a given year reflect hundreds of shots over 80 or so games, and are therefore highly predictive; football QB stats reflect many fewer attempted passes, partly because there are many fewer games per year, and are therefore less reliable.)
Anyway, the question of whether average-scores-per-game affects competitive balance in a sport seems at least worth considering.
dberri
November 20, 2006
Hi Tom,
This is an interesting observation. But the pattern we observe in competitive balance is not just across leagues, but also across time. If all that was going on is the number of scores in the game, why did baseball become more competitive in the latter half of the 20th century? Our focus on population changes can explain that pattern. I am not sure the scoring story has the same explanatory power. Still, this is an interesting observation.
Jake
November 21, 2006
I would like to add that the differences in competitive balance, both over time and across leagues, are probably resultant from more than one factor.
It is entirely possible that both theories have an impact. The notion that one lucky goal can win a soccer game, while one lucky basket rarely wins a basketball game makes a lot of sense. The more luck is involved in the game, the closer the outcomes will be to chance, or a 50-50 split.
Harold Almonte
November 21, 2006
I´m agree with the tall people argument, but the number of scores argument is very good. Those sports have more scoring difficulties (goal stoppers, more steps to score), players need to joint the good side of their averages with the few decisive opportunities, then every player can change the balance in a “jointed” moment. Basketball have too much moments, and no decisive until the final seconds of a tied game. I would like to know, How many come backs from a more than 15 points trailed game happens in basketball? How many bad teams do that?
Tom Hurka
November 21, 2006
Like Jake I was assuming that both theories might explain some of the facts — just suggesting that the number-of-scores theory shouldn’t be ignored completely.
And it wouldn’t be hard to test it even for one sport. Baseball scoring was low in the 1960s, wasn’t it, compared to earlier and later? And hockey scoring was at its peak in the 1980s. So you could see whether balance was greater or lower in those periods than at other times. The hockey results might not favour the theory, though, since the 1980s were the decade of first the Islanders and then the Oilers dynasties, and I think winning teams in those years had very high winning percentages. But there were also confounding factors, like the admission of new teams from the WHA just before the decade started (though of course the Oilers were one of them) that gave the good teams some weak ones to feast on.
Still, the fact that the ranking of sports by competitive balance tracks the ranking by average scores per game — and I’ll bet has always tracked it, despite changes in the availability of players — is surely suggestive: how likely would that result be if the only determining factor was the supply of athletes?
Jason
November 21, 2006
I’m not sure how to separate it out, but that in any given era, ballparks can and do have noticeable impacts on average runs scored, it seems like there may be a way of teasing out the effects of scoring on “competitive balance” with some independance from the talent pool.
Baseball also has a nice break-point immediately surrounding 1968 when the pitcher’s mound was lowered, immediately giving a boost to offense while the talent pool stayed more or less the same. Was there more competitive balance in 64-68 than in 69-73?
Tom Hurka
November 22, 2006
Sorry — one bit in my last comment had things backwards. If the NHL in the 1980s had high scoring (as it did) and less competitive balance (as I suspect) that confirms, rather counts against, the scores-per-game explanation. And re baseball, I was thinking precisely about the change in the height of the mound: the years before that were very low scoring.
Beamer
November 23, 2006
Another reason why sports may change their competitive balance is a change in the rules. In baseball one of the primary reasons for a shift in competitive balance has been the introduction of the FA market, the expansion draft and more money into the game.
Let me illustrate. If you play a game of basketball for just five minutes what will the end season standings look like? Probably a lot more even that it is at the moment. This is an instance whereby one changes the context and you change the probability of the better team winning. The other point is talent distribution take 5 NBA players and 5 economists. If they play basketball for a five minute game, who wins? The NBA players every time. If you mix the teams up, say 3 NBA + 2 economists a team (OK, you need an extra MBA player) then the chance of winning will be closer to 50/50.
Context and spread of talent are the two most critcal factors.
Beamer
November 23, 2006
Sorry. That last post was unfinished — posted by accident. Here it is again, completed.
******
A reason why sports may change their competitive balance is a change in context. In baseball one of the primary reasons for a shift in competitive balance has been the introduction of the FA market, league expansion and greater disribution of income in to the game. This shift in context can account for the change in competitive balance.
Now, among different sports by varying the rules one can radically impact the probability of the outcome. Let me illustrate. If you play a game of basketball for just five minutes what will the end season standings look like? Probably a lot more even that it is at the moment. This is an instance whereby one changes the context and you change the probability of the better team winning. The other point is talent distribution: take 5 NBA players and 5 economists. If they play basketball for a five minute game, who wins? The NBA players every time. If you mix the teams up, say 3 NBA + 2 economists a team (OK, you need an extra NBA player) then the chance of winning will be closer to 50/50.
Agina Chidiebube
December 21, 2006
i need to get more information about the things i saw here
jessica
January 9, 2007
Many tall people are there.
Ashleigh
June 27, 2007
there are enough tall people to stock 30 teams
the NBA could institute policies that makes sure they’re good players
argentina in 04 and greece in 06 didn’t beat the USA at basketball because of three point shooting 7 footers, they had solid 6’7 – 6’11 guys who were there because they could play basketball, not because of potential (KWAME)
Annon
July 2, 2007
To the people commenting on baseball read Moneyball before you talk about the sport. The author makes it clear that a lot the problems in baseball stem from false perceptions of what a good baseball player is and what factors contribute most to wins. It still continues today and I think that if the “powers at be” (the management) start looking for players who will help contribute most to wins and not ones who just look good (you will know what I mean if you read the book) then the sport will even out be a lot more 81/81 (theoretically) with teams deviating standardly from there (there will be very few 100 win teams). What would really happen would be: the baseball teams with the most money would win the most games because it has no salary cap. The best player would be paid a premium and the Yankees would have a pretty good chance of winning it all every year (although the playoffs are a crapshoot).
Interesting facts
May 23, 2010
The former Yugoslavia had the best guard, but with the disintegration of the country, disappeared the system of work with the players. Now create a tall players, but they do not have that quality.