90% of this game is half mental.

Dumbo isn’t the only guy to need help with a little performance anxiety. Slate has an interesting piece on whether or not sports psychology really works. The answer seems to be somewhere between “It depends” and “No one really knows.”

The fact that baseball shrinks can’t back up their work with numbers is at odds with the trend toward rational decision-making among baseball managers. Egghead GMs like Billy Beane and Theo Epstein have revolutionized the sport by using objective measures to build their teams. You might expect this new breed of executives to demand the same rigor from their psychologists.

But, say the shrinks, why not just ask the players if it works?

If John Smoltz says Llewellyn turned his career around, then Llewellyn turned his career around. Why shouldn’t we believe A-Rod when he says therapy helps him through his slumps?

This reminds me, of course, of the scene in Bull Durham when Crash lectures Annie about respecting the streak. “If you believe you’re playing well because you’re getting laid or because you’re not getting laid or because you’re wearing women’s underwear, then you are.” But Slate writer Daniel Engber is having none of it.

In fact, we have every reason to doubt the testimony of professional athletes. Baseball players in particular are notorious for ascribing their success to inane rituals, astrological signs, and other hokum. Wade Boggs used to eat chicken before every game. If Boggs says that eating drumsticks helped him get hits, should we believe him, too?

Well, within baseball’s internal if-you-build-it, still-we-believe, calling-his-shot logic, I think the answer is yes. The idea behind Dumbo’s feather wasn’t that he didn’t need it to fly. He did need it—he needed to believe in himself in order to fly, and he needed the feather in order to to believe in himself. It seems pretty simple to me: a little therapy never hurt anyone, and if you think it’s helping your game, then it is. Maybe Engber is just thinking about this too hard:

Full-on experiments—with players assigned to different treatment groups—would yield the best data, but even that level of rigor isn’t necessary. Mental trainers could learn a lot just by keeping careful logs of all their cases, with statistical outcomes for each player.

Oy. I’ll give the last word to Crash: “Don’t think. It can only hurt the ballclub.”