Dipping into the e-mailbag for the first time in a while ...

Hi Rob, Just browsing through the recent draft choices and I noticed something interesting (or maybe not interesting). Of the first 41 picks (before the second round began), 18 were high school players. That's 44 percent. Oakland had seven of those first 41 picks, and they used each of those seven picks on college players. I'm just wondering if you think that this was done on purpose? As in, the A's go for guys that are a bit more tested, older, more experienced, whatever. When you think of the kind of investments teams are putting in some of these 17-18 year-old kids, it seems to me that the wise thing would be to go after the older players.

Just wondering what your thoughts on this are.

Regards,
Shawn Goodell

On purpose? Everything the A's do is on purpose, Shawn. And it goes even beyond those first seven picks. As reader Ben Bremer points out, the A's didn't take a high school player until the 19th round, with their 25th pick -- both of which have to be records.

The Athletics obviously decided to concentrate on drafting college players, which makes sense because the risk factor is lower with college players. This is so obvious, assuming of course that you bother studying the issue, that it continues to amaze me that teams continue to spend millions of dollars on bonuses for high school players.

But what if you ignored high school players completely, didn't even bother scouting them?

Sounds crazy.

It's not, though. Almost exactly 10 years ago, at the SABR convention in St. Louis, Bill James presented a study he'd done of a related issue. Bill didn't break it down, college vs. high school. Instead, he simply looked at this question ... Would you do better in the draft if you arbitrarily ignored half the players in the draft, but devoted more of your energy to scouting the players you did not ignore?

Bill's answer?

"At all levels of the study a little bit of knowledge in depth proved to be of more value than a great deal of knowledge in breadth."

I'm seriously short-changing Bill's efforts here, due to space considerations and also my inability to simply reprint his work (with luck, someday he'll publish it himself). But I suspect that many of you won't believe the results I'm reporting, or that you'll recoil in horror when I tell you that Bill's study was not based on historical data, but instead was based on a computer simulation that he designed.

So let me frame this issue in a way that I think most of you could relate to ...

Let's say you've just been stranded on an island in the Pacific, just like Tom Hanks in Cast Away (which is showing, yet again, on HBO this morning). But where Tom Hanks had only a volleyball and a small photo of Helen Hunt for companionship, you've hit the jackpot. There are people on your island. And the king of the people is so impressed with your blue eyes, not to mention your ability to name the last six U.S. Presidents in chronological order, that he's offered you your pick of the island's 1,000 most beautiful unmarried young maidens. What's more, the king of this island has made all the arrangements for you to interview your prospective brides.

There's a catch, though (isn't there always?) ... You're allowed only 10 hours for the interviews, after which you have to make a final decision. And on this island, divorce is illegal and infidelity is punishable by death.

So here's where you have to make a decision. Do you want to interview each of the 1,000 candidates -- which works out to 36 seconds per candidate -- or do you want to limit yourself to, say, only 100 candidates? If you do that, you're going to meet only 100 of the island's most beautiful women ... but you'll get to meet each of them for six minutes.

Well, I don't know about you guys, but I'll take my chances with 100. Shouldn't that be enough, if I learn enough about each of them?

And it's the same way with the baseball draft. Depth of knowledge is better than breadth of knowledge. Except where Bill surmised that you could do better if you arbitrarily ignored half the players, I'm suggesting that you ignore the high school players. That's right; don't even scout them. Employ just as many scouts, but send them to the colleges.

Theoretically, a team that completely ignored high school players would, I believe, do well. But practically, I don't think that's necessarily the best strategy. If Alex Rodriguez or Junior Griffey is eligible for the draft, you should probably do your homework (though of course in their cases everybody knew they were the best players in the country, so you could have done all the necessary homework with an issue of Baseball America).

I know, it sounds crazy. If you got all 30 major-league scouting directors in a room and suggested to them that they essentially ignore high school players next spring, all 30 of them would sign a petition to have you locked away in the monkey house.

Bill addressed this issue in his study, writing,

On an intuitive level, the conclusion urged by this research is almost impossible to accept. In any draft situation, one of the main things you think about is the fear of overlooking somebody; you always think "I hope I didn't miss anybody." What I am suggesting here is that you might be better off to deliberately overlook half of the players in the draft, and that's a very unsettling idea.

But when you think about it, I think you can see the logic. On the one hand, you think "What if there's a superstar over in that group, and I don't even look at him." But on the other hand, what if there's a superstar in this group, and you miss him because you don't take the time to look carefully enough at him? Since superstars often are picked in the second round or later, that's a realistic possibility.

Is this radical. Yes, it is. But today's radical is tomorrow's moderate. And if the A's aren't going to draft high school players, then why should they bother scouting them?