Sleep well: Walking Barry /is/ nuts

Jun. 3, 2004 12:00 AM


      UP ALL NIGHT

There's a lot of stuff that keeps us awake at night.

Like wondering what happened to those two missing propane trucks in San
Antonio. You don't just misplace those, do you?

Gas price gouging bugs us, too. Have you ever noticed that prices will
spike to where everybody gets really miffed and then, for some reason,
they drop suddenly?

Then they spike back up until we get angry again, and they drop down,
but each time the low is just a little higher.

Next thing you know, we're really excited to see $2.09 a gallon, which
really ticked us off just a few months ago. They're desensitizing us,
you know.

Anyway, we also lie awake at night wondering if it really is a good idea
to walk *Barry Bonds *all the time.

Sure, he's a monster and scares the living bejeebers out of opposing
managers.

But has anybody really figured out whether it pays off to put him on base?

Turns out, somebody has, and he's a Duke University professor, so you
know he's really smart and has good basketball tickets.

* Jerry Reiter*, an assistant professor in the Institute of Statistics
and Decision Sciences at Duke, wasn't getting enough sleep, either.

"I'm a baseball fan, and I was watching a Giants game and talking
baseball with a graduate student, and we started wondering if it really
is the right strategy to walk Bonds so much," Reiter told us. "So we
decided to collect data."

Bonds will be happy to know that Reiter's analysis - of all sorts of
situations, with no outs, one out or two outs and the bases empty, one
on, etc. - of the past three seasons indicates managers should pitch to him.

For instance, with the bases empty and nobody out, Bonds was walked 80
times in 2001-03, and the Giants scored at least one run 46 percent of
those times, averaging 0.9 runs per inning.

In the same situation when pitchers threw to him, the Giants scored at
least one run 36 percent of the time and averaged about 0.6 runs per inning.

Reiter found that only when there is nobody on base and at least one out
is it statistically better to walk Bonds, which makes sense because it
would probably take a couple of hits to get him across the plate.

Reiter said he didn't bother to analyze runners in scoring position and
first base open. After all, any manager would likely walk any great
hitter in that situation.

"But with a runner at first, it turned out that pitching to him was the
best strategy in all scenarios," he said.

Baseball's a game of statistics, so we're surprised managers haven't
figured this out. So is Reiter.

"I mean, my sense is that baseball operates on two levels, a statistical
level and a gut level," he said.

"More and more teams and managers are utilizing statistics. But there's
still that gut reaction when they see Bonds on /SportsCenter/ hitting
those monster home runs. They react with their gut rather than looking
at data. The data say: 'Pitch to him.' "

Ah, but it's the gut that keeps them awake at night.