July 6, 1997 CHAPEL HILL -- Come late July, outfielder Tony Gwynn of the San Diego Padres could best the late, great Ty Cobb as the top hitter in baseball history.
So says Dr. Michael Schell of the University of North Carolina at Chapel Hill after investigating a baseball mystery: Why do yesterday's players always look better, statistically, than today's? To Schell, that doesn't make sense.
A biostatistician at the UNC Lineberger Comprehensive Cancer Center, Schell attempted to level the playing field, accounting for differences in baseball during Cobb's 1905-1928 career with the Detroit Tigers and Gwynn's career today.
Officially, baseball still calculates batting averages the same way, by dividing numbers of hits by numbers of at-bats. Schell says that because the game has changed drastically since Cobb's day, averages must be adjusted statistically to compare players' ability fairly.
He modified batting averages with four factors never before combined: the league average, which adjusts for how easy or hard it was to get a hit in a given year; ballpark effects, or how easy or hard it is -- or was -- to get hits in a given stadium; and variability in talent among players of different eras. Finally, he limits comparison to the first 8,000 at-bats to assess hitters only in their prime.
The researcher applied the four adjustments to the game's greatest hitters, including Rod Carew, Joe Jackson, Rogers Hornsby, Honus Wagner, Ted Williams, Stan Musial and Wade Boggs. He found that Cobb and Gwynn came out on top and determined that with his 8,000th at-bat, which probably will come late this month, Gwynn could outrank Cobb as the greatest.
Baseball figures Cobb's batting average at .366, Gwynn's at .339. But with Schell's four adjustments, Cobb's changes to .342, and Gwynn's became .341 at the beginning of the 1997 season, after 7,595 at-bats. So far this year, Gwynn, who will play in the July 8 All-Star Game, is batting .400 -- better than Schell estimates he must to outrank Cobb by his 8,000th at-bat.
"Everyone thinks Cobb is the best ever in baseball, and no player today can touch him," said Schell. "But if Gwynn bats .366 through the end of July, by my estimate he'll be the best."
A lifelong Cincinnati Reds fan raised partly in Ohio, Schell developed the data on his own time while researching "The Top 100 Hitters," a book he has proposed to several publishers. He also intends to publish his findings in a statistics journal.
In his UNC-CH job, Schell uses statistical analysis to determine which cancer treatments work best with the fewest side effects. Baseball has a lot to do with the line of work he chose: "Statisticians are big fans because baseball has the most detailed numbers of any sport."
The first numbers Schell figures into his comparison of Cobb and Gwynn concern league batting averages. Those rise or fall in given years because of such changes as lowering the pitcher's mound or expanding the strike zone. The league average adjustment also accounts for the emergence of night baseball, when hitting is harder, and increased use of relief pitchers, both of which have decreased batting averages.
Ballpark effects comprise the second of Schell's adjustments. He estimates Cobb's average was helped 6.4 points by favorable hitting conditions in Tiger Stadium while Gwynn has been hurt 2.3 points by the disadvantages of Jack Murphy Stadium. Though today Tiger Stadium favors pitchers, in Cobb's day it was a hitter's park.
Thirdly, Schell adjusts batting averages for differences in overall player talent between the two eras. "Cobb played at a time when his competition was not as strong, and therefore he looked relatively better," Schell said. "Emergence of baseball as a solid career and addition of farm teams and spring training all helped make players more consistently better today than in Cobb's day."
This modification is critical, he said. "If one adjusts only for the league batting average and the ballpark effect, the top hitters on average were the early hitters. So were the top double and home run hitters and players leading many other batting and pitching categories. It doesn't make sense that they should be best in all these aspects."
A feature associated with the ballpark effect is the size of foul territory. The bigger the territory, the greater the opportunity to catch a batter out.
In his final exercise to compare averages fairly, Schell cuts off comparison at 8,000 at-bats -- using 4,000 as a minimum to qualify for comparison -- to assess hitters only in their prime. He aims to avoid favoring players who retired early over those who continued playing after age reduced ability.
He maintains that his adjustments are needed to assess true performance: "All those things are going to change how easy or hard it is to get a hit. You have to look at those if you want to compare players of all eras fairly." Schell has seen Gwynn play and soon will again. When the man Schell thinks will be king is scheduled for his 8,000th at-bat, in whatever park, whatever city, "I want to be there."
- 30 -
Other social bookmarking and sharing tools:
The above story is reprinted from materials provided by University Of North Carolina At Chapel Hill.
Note: Materials may be edited for content and length. For further information, please contact the source cited above.
Note: If no author is given, the source is cited instead.