Not Enough Basketballs? The Too-Much-Talent Effect

MIAMI, FL - April 20: Al Jefferson #25 of the Charlotte Bobcats shoots against the Miami Heat during Game One of the Eastern
MIAMI, FL - April 20: Al Jefferson #25 of the Charlotte Bobcats shoots against the Miami Heat during Game One of the Eastern Conference Quarterfinals of the 2014 NBA playoffs at the American Airlines Arena in Miami, Florida on April 20, 2014. NOTE TO USER: User expressly acknowledges and agrees that, by downloading and/or using this photograph, user is consenting to the terms and conditions of the Getty Images License Agreement. Mandatory copyright notice: Copyright NBAE 2014 (Photo by Nathaniel S. Butler/NBAE via Getty Images)

The NBA playoffs are underway, and the Miami Heat are the odds-on favorite to "three-peat." If they do -- or if they don't, for that matter -- the outcome will fuel an enduring debate about how best to build a sports franchise. Back in 2010, the Heat opted to wager hundreds of millions of dollars on the Big 3 -- signing superstars LeBron James and Chris Bosh on top of pricey local favorite Dwayne Wade. James boastfully predicted a Heat dynasty, while cynics chanted a more skeptical mantra: "Not enough basketballs" for those super egos.

Is there such a thing as having too much talent? This is the perennial question facing the owners of big-time sports franchises -- not to mention the managers and coaches and players and fans. Does adding more and more talent add up to ever better team performance?

Surprisingly, this question has never been studied in a rigorous, scientific way -- until now. A team of psychological scientists -- headed up by Adam Galinsky of Columbia University and Roderick Swaab of INSEAD -- decided to explore the possibility of a too-much-talent effect--the notion that at some point adding one more superstar actually becomes detrimental to the team. They reasoned that internal jostling for team dominance would eventually undermine the coordination needed for team performance. They ran a series of experiments to test this idea, including one that focused on the NBA.

Galinsky and Swaab looked at the regular season play of all NBA teams over a decade, from 2002 to 2012. They computed individual talent, and team talent, by using the so-called Estimated Wins Added, or EWA, formula, which estimates the victories that any given player adds over and above what a replacement player would contribute. They had access to comprehensive play-by-play data from all the games, which they examined to tally team coordination -- an amalgam of total assists, field goal percentage, and defensive rebounds. Team performance was simply winning percentage at the end of each season.

Then they crunched all the data together, with interesting results. Increasing talent was linked to better team performance -- but only to a point. After that critical point, the benefit of more talent decreased and eventually turned negative. What's more, it was clearly the diminished team coordination that hurt performance. That is, too much star talent undermined the selflessness that leads to team excellence. Or as sports analysts say, not enough basketballs.

But what about soccer balls? Perhaps there is something unique about American hoops that lends itself to the too-much-talent effect. To rule this out, the scientists ran a similar study of international soccer, using data from FIFA. They assessed talent in a somewhat different way, but basically it was the same study concept. And they came up with the same result: Simply stacking a team with the best talent in the game -- this strategy only gets a team so far. Then it turns sour.

Okay, but here's the really interesting part. Recall the scientists' prediction that the too-much-talent effect would only emerge when individual jockeying harmed team coordination. But what if interdependence were not so critical to a team's success--in baseball, for example? Obviously baseball players have to coordinate a lot on the field, but game outcomes have much more to do with individual performances than they do in basketball and soccer. That's why it has been called "an individual sport masquerading as a team sport." So Galinsky and Swaab ran yet another analysis, this one of talent and team performance using MLB data. And as expected, they got a different result: Accumulating talent did not hurt team performance at the ballpark. There is no too-much-talent effect at work in America's national pastime.

So all this data, reported in detail in an article to appear in the journal Psychological Science, document the predicted effect and explain why it occurs. But despite this, as the scientists show in a final study, people believe the opposite. They intuitively believe that their favorite team will get better by piling on more and more top talent.

Just ask Miami Heat fans. Even a lackluster regular season has not diminished their hope and faith in a dynasty yet to come. And who knows -- maybe they will win another. But contract options are looming, and it's the owners who soon will have to decide if it's worth that big price tag to keep the Big 3 intact for years to come.