*The Charlotte Observer*

	Posted on Sun, Jul. 04, 2004


*COMMENTARY*


  Managers might not realize bad odds for walking Bonds


*STAN OLSON*

In most cases, teams are better off pitching to San Francisco slugger
Barry Bonds than walking him.

That's not an opinion. That's hard statistical evidence, developed by
Duke's Jerry Reiter.

A little background: Bonds is easily the most feared hitter in the game.
In 2001, he set the season home-run record with 73. He also set the
record for walks in a season with 177. The following year, he broke that
mark with 198 walks, and added 148 last year.

Through Friday, he led the majors with 117 walks, 62 of them
intentional. That put him on pace for 237 this year, and he entered
Saturday four walks short of Rickey Henderson's career record of 2,190.

Because of his phenomenal success, more and more managers are walking
Bonds when traditional baseball strategy says not to, most notably with
no outs and no one on base.

They shouldn't be.

Reiter, an assistant statistics and decision sciences professor, was
watching Bonds being walked yet again a few years ago when he decided to
use the numbers to determine if that strategy was effective.

"I just wanted to flat-out let the data do the talking," he said.

He charted every one of Bonds' at-bats over a three-year period
(2001-03), taking each game situation into account, and determining
whether the Giants had been more likely to score after Bonds was walked
or after he was pitched to. Reiter presented that information in a
recent article in The Baseball Research Journal.

Reiter's research showed the Giants scored in 46 percent of innings that
Bonds began with a walk. They scored in 36 percent of innings in which
he was pitched to.

With one out and a man on first, San Francisco scored 28 percent of the
time following a Bonds pass, but just 13 percent when he was pitched to.

Reiter analyzed all 24 variations of outs and men on base that a batter
can face, and found that "with a runner on first base, pitching to Bonds
is better than walking him, no matter how many outs there are."

Most situations, Reiter said, were too small to be statistically
significant. For instance, Bonds was walked seven of nine times when he
batted with no outs and runners on second and third.

There are certain times when walking Bonds is prudent. Those primarily
occur with no runners on and at least one out.

"With two outs and nobody on, the Giants scored about 13 percent of the
time when he walked," Reiter said. "That went up to 16 percent when he
was pitched to. With one out and nobody on, it was 18 percent when Bonds
walks and 29 percent when he's pitched to."

For the most part, Reiter's analysis shows walking Bonds can get a team
into more trouble than pitching to him, particularly at the beginning of
an inning. Yet we see Bonds flip aside that bat and stroll to first,
game after game.

"One possible explanation for that is the `SportsCenter' phenomenon,"
Reiter said. "On there, you never see him making outs. There's this
perception that gets perpetuated by all these highlights, a sort of, `Oh
my God, every time he comes up he's going to hit a home run.' Maybe the
reaction to that power and ability is not to deal with it."

Keep in mind as well that Reiter's conclusions have not been widely
publicized. And many managers still prefer their hunches to statistics.
So they keep sending Bonds down to first base.

And the Giants, more often than you might expect, keep making them pay. 

Stan Olson