Peter Lynch is one of my idols, and along with Warren Buffett, he's one of the most famous stock investors in the world.  In his two bestselling books, Beating the Street and One Up On Wall Street : How To Use What You Already Know To Make Money In The Market, one of the central questions is "How can a layman expect better results than a professional paid millions to run a mutual fund?"  And his explanation might also explain how you or I can (sometimes) make better decisions than the coaches of the NBA and NFL who are paid millions for it.

Would probably make a decent basketball coach.

You see, one of the things that running a mutual fund, a basketball team, and a football team all have in common is that there are politics involved.  The buck doesn't actually stop there.  Making the right decision alone isn't enough; one frequently has to justify the decisions one makes.  Especially if any of those decisions have disastrous consequences.  The boss, the owner, the board of directors, perhaps the media -- they will all want to hear about why you did X instead of Y.  Especially if the decision you made defies conventional wisdom.  You may think that a professional should not care what any of these people think as long as he or she knows that the decision was the correct one (after all, he/she is the expert, and everyone else is just a layperson playing armchair quarterback), but the truth is that professionals often have little choice.  Your boss might fire you to provide a scapegoat.  The media can get you fired in a lot of markets.  The players on a lot of professional sports teams can get you fired if they don't buy in to your decisions.

In stock trading, Lynch's argument boils down to this:  a mutual fund manager often has to follow the masses, even when he strongly suspects the masses are wrong.  Because if he doesn't, and it's one of those times where the masses are right, he'll have to face the boss saying, "Why the hell didn't you buy X?  Everybody was buying X, dammit!"  Furthermore, if the masses are, in fact, just as wrong as you suspect, he will not be punished for making the same mistake they did.  He can look at his boss and say "Well, that sucked, but you know, everybody was buying X so you can't blame me for doing it too."  So the pressure to follow conventional wisdom is enormous.  Sound familiar?  I talked about this concept in my first article about the draft:  there are some picks that are "safe", and the manager who doesn't take that pick is always risking the wrath of his superiors and/or the public, so the pressure to make that pick is great, even if he's not certain it is the "right" pick.  If you pass on the consensus #1 and your pick doesn't turn out, woe be unto you.

There are lots of ways that we fans second-guess coaches and managers, but there are two situations that come to mind where an educated fan might think the coaches most often make the wrong decision for precisely this reason.  In the NFL, coaches very rarely go for it on 4th down, even though they nearly always should (Dre Alvarez makes a very good roundup of the NFL coaches' week 1 decisions here), and in the NBA, coaches almost always take a player out of the game with his 2nd foul, even though they almost never should.  And in both of these cases, the decision probably has a lot more to do with covering one's bases than it does with making the right choice.

When facing a 4th down, coaches consider going for it the "risky" play and punting the "safe" play, but fail to consider that giving the opponent possession of the ball carries it's own risks; often the points you give up by punting outweigh the value of going for it -- even if you'll fail a good percentage of the time.  And they often fail to consider field position -- if you punt from the opponent's 43, is the opponent really going to be much worse off than if you go for it and fail? In basketball, coaches worry about players getting in foul trouble, and thus being unavailable or less effective in crunch time.  But this is a self-fullfilling prophecy;  if you take LeBron out of the game at the 5 minute mark with 2 fouls, you are replacing him with a much worse player for about 7 minutes of game time.  In other words, you just lost the very efficiency that you were trying to save.  This mistake is exascerbated even more if you adhere to the "superstar bonus" that referees give players, meaning LeBron is unlikely to foul out anyway if he stays in.  In any case, removing him only makes sense if one believes that minutes at the end of the game "matter" more than minutes at the beginning.  And I suspect this is a common belief, but has very little basis in evidence. Replacing LeBron for about 7 minutes with an average player loses you .279 WP48 per minute, or .005 wins per minute, or .035 wins.  3.5% of a win might not seem like a lot.  But over an 82 game season, if you do this every night, you'd lose about 2 more games than you'd have to.  And that's only if you don't do something equally silly, like take out LeBron if he gets a third foul early in the second quarter.

The explanation is very similar to our mutual fund manager's.  Conventional wisdom is that you remove a player when he gets his 2nd foul, because this is "safe", and "you need him in crunch time".  You never see media articles with the press writing "If only that fool hadn't taken LeBron out when he got his second foul," or even more telling "Wow, they got lucky that they won even though the coach foolishly took out LeBron when he picked up the 2nd foul." But if the coach leaves him in with 2 fouls, and he later fouls out, he'd better have a change of clothes, because a media shit-storm of second-guessing is coming his way.  After all, everybody knows you need your best player in the fourth quarter, right? We forget that, actually, to win games, you need your best players to play the lion's share of minutes.