Pirates, Cardinals defying Pythagorean record
BY JOHN GROCHOWSKI For Sun-Times Media August 12, 2013 7:56PM
A fan holds up a flag for the Pittsburgh Pirates as the Pirates face the Colorado Rockies in the first inning of a baseball game in Denver on Friday, Aug. 9, 2013. (AP Photo/David Zalubowski)
Updated: August 13, 2013 12:28AM
The major-league playoff races are nearing the three-quarter pole. By the end of the week, all teams will have reached the 121-game mark, which means only a fourth of the season remains.
No one is doing quite what the Orioles did last season, when they finished 11 games better than their Pythagorean projection and rode that into the playoffs. That’s an aberration, and most teams are going to win at a rate a lot closer to the one implied by their runs scored and allowed. The formula is (runs squared)/(runs squared plus runs allowed squared) = winning percentage.
This season, no team is more than six games better or worse than its Pythagorean record. Coincidentally, the teams that are six games better and six games worse than their Pythagorean records are in the National League Central. The Pirates are 70-47, three games ahead of the Cardinals’ 67-50. By run data, the Cardinals are projected with a 73-44 record, nine games better than the Pirates’ 64-53.
Part of that is record in close games. The Pirates have been one of the most successful teams in baseball this season in the close ones, going 23-16 in one-run games and 9-5 in extra-inning games. The Cardinals are 13-14 in one-run games and 1-5 in extras.
Not only has the Cardinals’ record been mediocre in games decided by one run, but they’ve played the fewest one-run games in the majors. Perhaps their incredible .334 batting average and .874 OPS with runners in scoring position have been taking victories out of the one-run category.
A lockdown bullpen or one that gives up too many leads can cause a team’s record to stray from the projection. So can an unusual run of clutch hitting in close games. That tends not to hold up from season to season, and teams that greatly exceed or fall short of the projection one season usually play closer to Pythagorean the next.
There’s no indication managers have any effect. Tony La Russa, the winningest manager in baseball since the expansion to a 162-game schedule, was only 11 victories above projection for his 33-year career. Bobby Cox, second in victories during the era, was on the negative side, losing 12 more games than projected. So was No. 3 Joe Torre (minus-11). All are insignificant amounts, less than a half-victory a season. Of the 25 winningest managers whose careers fell entirely within the 162-game era, only Felipe Alou reached two victories per season above Pythagorean, and even that’s within the bounds of chance.
Nor is there any indication first-year managers have success relative to Pythagorean that differs from their more experienced competitors. In the decade 2003-2012, 31 men managed their first complete season. Sixteen had records below projection, 13 above and two right on. Their cumulative record was one game above projection.
What is significant is the effect winning a little more or less than projected can make in a division race. Just look at the Cardinals and Pirates.