The Talent Evaluation Secrets of the NBA
Thursday night is the annual NBA Draft--and the 30th anniversary of Michael Jordan's first day as a basketball pro. (Jordan, after his junior year at the University of North Carolina, was the third pick of the 1984 draft.)
In search of the next Jordan, NBA teams devote tons of time and cash to in-person scouting. They also deploy analytics, hoping to find and exploit "moneyball" inefficiencies in the talent supply. But finding top talent isn't easy. After all, when you watch a young player, all you really is how he fares against weak competition. What lessons, then, have NBA teams learned about discovering high-upside performers? Plenty. Here are two of them, which you can apply to your own recruiting and hiring efforts.
1. Align leadership at multiple levels. Basketball organizations generally have three major leaders: The owner (who signs the checks), the general manager (who selects players and oversees talent and compensation decisions), and the coach (who manages the players on a daily basis).
No doubt, your own organization has its version of these three management tiers. In which case, you know how important it is--culturally and functionally--to agree on talent evaluation practices. It's easier said than done, since all three tiers have different priorities. Coaches want performers who can help right now, usually at any cost. General managers want to win too, but they're often willing to make small sacrifices in performance for the sake of developing younger talent and trimming the payroll. Owners, for their part, have their eyes on larger questions, like branding and the business model.
How, then, can these three tiers find alignment on talent decisions? The same way top teams always find alignment: Bring conflicts to the surface. Do not let them simmer. At the end of the day, the coach works for the general manager, who works for the owner. A decision will be made. But it shouldn't be made before an honest discussion.
If there's one NBA team where all three tiers are aligned over talent evaluation, it's the Dallas Mavericks. Hard as it is to devote another sentence to owner and headline-maker Mark Cuban, the fact is, his organization is on the same page when it comes to analytics. Cuban and his coach Rick Carlisle hired a renowned analytics guru named Roland Beech. You might think, so what? Doesn't every NBA team have an analytics guru or three?
Yes. But in most cases, that guru doesn't sit on the bench during games. The bench and the games are generally considered the private province of players, coaches, towel boys, water boys, and team doctors. General managers, owners, and other team employees seldom tread there.
But Beech actually sits on the bench during games, notes Thomas Davenport in the MIT Sloan Management Review. This prevents him from making his analytics decisions in a vacuum. He's not an isolated number cruncher. Instead, he's exposed to the in-game realities that Carlisle has to manage. He can see, for example, that sometimes a coach needs to use a tough player with lousy stats, rather than a soft player with a shimmering analytics profile. To win a game, sometimes a team needs physicality more than it needs scoring efficiency.
The business equivalent to this, notes Davenport, would be "for CEOs, middle managers and analytical specialists to be working closely together and consulting frequently with each other on key decisions."
2. Analytics are great, but there's no substitute for repeated, in-person talent evaluations. As a player in the 1960s and 1970s, Jerry West had a Hall of Fame career. As a general manager for the Los Angeles Lakers in the 1980s, 90s, and 2000s, he amassed the talent that led to eight titles.
One of West's biggest skills was old-school scouting: Traveling to games, observing players in different in-game circumstances, taking notes, exchanging opinions with other scouts. "He revolutionized the position of executive in the sport," writes Roland Lazenby in his biography, Jerry West: The Life and Legend of a Basketball Icon. "West dug in and did the nitty-gritty work, loved doing it, and he would continue to do this his entire executive life. Whereas other GMs tended to stay in their offices, West remained in the field, intensely studying the game."
In modern-day terms, you might say West put in his 10,000 hours of scouting. You could even say he learned how to spot a player who'd logged 10,000 hours of practice. He did it in 1996, when he scouted a teenage Kobe Bryant. "There was length, there was strength, there were physical talents, but to go with them was a beautifully polished set of skills, the kind of skills that a 17-year-old could possess only after long hours of dedicated work," writes Lazenby.
So West took a major risk: He traded the Lakers' starting center in order to draft Bryant with the 13th overall pick. Even though Bryant had never played college basketball and was coming straight from high school. Bryant, of course, went on to become the best player in his draft class--and one of the best 20 players in the sport's history. Yet 12 teams in the 1996 draft ignored Bryant's potential before West pounced.
The lesson? Put in your 10,000 hours of talent evaluation. The more candidates you assess--and the more you assess each individual candidate--the more you'll hone your own eyes for talent.
Professional service firms like McKinsey and Goldman Sachs have made extended evaluation a part of their hiring process. Charles D. Ellis wrote about their methods in his most recent book, What It Takes: Seven Secrets of Success from the World's Greatest Professional Firms. Capital Group, the legendary investment-management firm, has as many as 20 different employees interview serious candidates. McKinsey believes that the optimum number of interviews is eight.
In sum, then, there are two lessons. First, if you're conducting only two or three interviews for each candidate, you should conduct more. Second, evaluate your "magic number" for interviews. At McKinsey, it was eight.
Yours might be more, or it might be less. But it's probably more than two.