Saturday, May 17, 1997

Rangers manager looks to on-base percentage to judge talent

By Al Pickett / Abilene Reporter-News

How do you compare apples and oranges?

Baseball fans have been arguing that for years. How do you use statistics to compare players?

Even Bill James, considered baseball's top statistician and author of Baseball Abstract, said recently that there are too many theories now of how to compare baseball players statistically.

In light of that, Doug Melvin, general manager of the Texas Rangers, was asked what statistics he uses when judging talent.

Speaking earlier this week to the annual convention of the Southwest region of the Associated Press Sports Editors, Melvin said on-base percentage is the most important statistic.

That's the percentage of time a batter gets on base via a basehit, walk or hit by pitch. A hitter may only carry a .250 batting average, but if he has a good eye at the plate and draws frequent walks, his on-bat percentage may be well over .300.

But then Melvin gave an interesting statistic.

"If the slugging percentage and on-base percentage add up to .800, you have a good player," Melvin said.

Slugging percentage is the total number of bases in comparison to the number of at-bats. For example, if a batter had one hit in 10 at-bats, his batting average is only .100. But if that one hit was a home run, his slugging percentage is .400.

Melvin's theory holds pretty true if one looks at the Rangers' 1996 final statistics.

Eight regulars who spent the entire season on last year's Texas roster combined for an on-base percentage and slugging percentage of better than .800.

American League MVP Juan Gonzalez led the way with a slugging percentage of .643 and an on-base percentage of .368.

Also topping the combined .800 mark were Damon Buford, Will Clark, Rusty Greer, Warren Newson, Dean Palmer, Ivan Rodriguez and Mickey Tettleton.

All of those players had a slugging percentage of at least .450 and an on-base percentage of at least .348.

Melvin also said when he rates pitchers, the least important statistic is wins and losses.

"Wins and losses are overrated," he said. "Pitchers should have double the number of strikeouts to walks. And they should give up less hits than innings pitched. Earned run average is important, too. If they have those numbers, wins and losses will take care of itself."

Using that criteria, how did Ken Hill rate last year as the team's top starting pitcher with a 16-10 record?

Pretty well. Hill pitched 2502/3 innings and gave up 250 hits. He had 170 strikeouts and walked 95. Hill's ERA was 3.63.

Roger Pavlik struggled in the last half of the season last year, despite his 15-8 record. By comparison, Pavlik allowed 216 hits in 201 innings and had 81 walks compared to 127 strikeouts. His ERA was 5.19.

By using those same comparisons, Mike Henneman wasn't all that bad last season as the Rangers' closer. Despite an 0-7 record and a 5.79 ERA, Henneman had 31 saves. He gave up 41 hits in 42 innings and had 34 strikeouts and walked 17.

But those numbers pale in comparison to the numbers the Rangers' new closer, John Wetteland, posted last season with the New York Yankees. Wetteland was 2-3 with a league-leading 43 saves. He allowed 54 hits in 63 innings, and his strikeouts-to-walks ratio was more than 3-1. He had 69 strikeouts and walked just 21.

When you pore over the major league statistics in Sunday's Reporter-News, these are a few interesting things to consider as you try to compare baseball's top players.