The Warriors Aren't Lucky
The Golden State Warriors won late Wednesday night, in yet another flurry of Stephen Curry three-pointers. With that they ended the regular season with a record of 73 wins and 9 losses -- the best in National Basketball Association history.
If this were baseball or hockey, and a team had just set a regular-season wins record, we’d be taking bets on who would knock them off in the playoffs. The two Major League Baseball teams with the most wins, the 1906 Chicago Cubs (116-36) and the 2001 Seattle Mariners (116-46), both failed to win the World Series -- the Mariners didn’t even get there. The 1995-96 Detroit Red Wings, who hold the National Hockey League record for most wins in a regular season, with 62, didn’t make it to the Stanley Cup Finals.
If the Warriors get knocked out in the playoffs, on the other hand, it would be something of a shock. The four teams that before this week topped the list of best NBA regular-season records all went on to win NBA titles, as did eight of the top 10.
That’s because, of the major team sports, basketball is the one in which skill counts for the most. Here’s the well-known skill-luck continuum assembled by Michael Mauboussin -- then of Legg Mason, now of Credit Suisse -- in 2010:
One of the main metrics Mauboussin used to determine a sport's skill level was simply to calculate the variance -- a measure of how spread out a set of numbers is -- of teams' winning percentages for a season and compare that to what the variance would have been if wins and losses were determined by coin flips. The best NBA teams have higher winning percentages, and the worst have lower winning percentages, than their counterparts in most other team sports.
Why is basketball so much less dominated by luck? One theory, proposed by economists David Berri, Martin Schmidt and Stacey Brook in their 2007 book, "The Wages of Wins: Taking Measure of the Many Myths in Modern Sport," is that there just aren't enough tall people out there:
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.
Still, soccer also scores high in skill and imposes no such extreme physical restrictions -- the world's best player is 5'6". So the very nature of the game must play a role. The nature of basketball is that a small group of players does the same danged thing over and over again in a small playing area. Here's Mauboussin comparing college basketball to college lacrosse (in which the games last 60 minutes to college hoops' 40) in his 2012 book, "The Success Equation: Untangling Skill and Luck in Business, Sports, and Investing":
In basketball, each team gets the ball almost twice a minute, while in lacrosse each team gets the ball only once every couple of minutes or so. The size of the sample of possessions in basketball is almost double that of lacrosse. That means that luck plays a smaller role in basketball, and skill exerts a greater influence on who wins.
An interesting question is whether skill-dominant sports are more or less fun to watch than those where luck plays a big role. Tastes obviously differ, but my impression is that, while highly knowledgeable basketball fans can find something to savor in every game, most people find the NBA regular season to be a bit of snooze in its predictability and seeming endlessness. It's only in the playoffs, when more is at stake and the best are matched up against the best, that the sport comes into its own. This year the Warriors will probably win it all (FiveThirtyEight's CARM-Elo prediction model currently gives them a 42 percent chance, though I'm willing to chip in at least 9 more percentage points), but if they come up against the San Antonio Spurs -- whose record this year tied for seventh best all-time -- in the Western Conference finals, it could get pretty interesting.
I of course cannot end this without a mention of investing, which according to Mauboussin's continuum (it's represented by the piece of paper with a chart on it) is an endeavor dominated by luck. This doesn't mean that skill plays no role in investing results -- just that it's a smallish role, hard to separate out from all the noise. In many other modern occupations (such as mine) we don't even consistently measure the outcomes of our work, meaning we can't even really begin to put them on a skill-luck continuum.
That may be part of the near-universal appeal of Warriors star Curry. We know that he's making all those crazy shots not because he's lucky but because he's really, really good.
Because hockey has tie games, most wins doesn’t necessarily mean best record. The 1976-77 Montreal Canadiens won 60 games and hold the NHL record for most points (two for a win, one for a tie) in a season. They did go on to win the Stanley Cup. Still, only two of the past 10 Stanley Cup winners also had the best regular-season record, so the general argument still stands.
This is from the July 15, 2010, Mauboussin on Strategy report from Legg Mason, titled "Untangling Skill and Luck." It used to be easy to find online, but doesn't seem to be anymore.
Football is a weird exception, because there are so few games that undefeated and winless aren't unheard of. But if you average over several seasons, or apply other statistical screens, it becomes clear that it's a high-luck sport.
The usual disclosure: This book was published by the Harvard Business Review Press, and I used to work there. In fact, I think I was editorial director when it was acquired. But my involvement didn't go much beyond that.
The Warriors' pursuit of the wins record added some excitement to this year's regular season, but that sort of makes the point: It usually isn't that exciting.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
To contact the author of this story:
Justin Fox at firstname.lastname@example.org
To contact the editor responsible for this story:
Zara Kessler at email@example.com