Solar Powered Pinstripes: The Yankee’s Strange Day/Night Splits

Last season, The New York Yankees won the American League East with 97 wins and the best record in the American League, but they did not hang around long in October, losing in the first round to the Detroit Tigers. For Yankee fans, the early exit was disappointing, but a fan paying close attention to the numbers might have seen this coming.

The Yankees were a very consistent team throughout 2011. They never fell below .500. They never won less than 15 games in any one month but also never won more than 18. Like most teams, they were better at home than on the road, but the difference was not particularly extreme. In one area though, the Yankees exhibited an extraordinarily skewed split. The 2011 team was 44-12 in day games and 53-53 in night games, the most extreme day/night split in baseball. With such an extreme split, were the entire season played in daylight, the team would have won 126 games, while an all night schedule would have resulted in just 81 wins.

With all but one of the playoff games occurring at night, the post-season schedule played against the Yankees strength. Their playoff opponent, Detroit, who also played better in day games, was significantly better at night, with a .564 win percentage under the lights. This may not have mattered in the end, as the Yankees two playoff wins came at night, but does give the analytical baseball fan pause. Is this extreme split record an anomaly or is there something about the Yankees team that makes them better suited to day games?

There are few reasons to believe that this may not be just a coincidence. For one thing, this was the second straight year the Yankees played better in day games. The 2010 team did not have such an extreme split, but they played .621 baseball in 58 day games and had a .561winning percentage at night. The 2009 World Champion Yankees team, however, did not exhibit this same tendency, playing slightly better at night (.645 win%) then at night (.618 win%). This complicates the issue significantly; if the last three teams all better during the day, the influence of the New Yankee Stadium would be a strong suspect here. With the 2009 team thriving at night, the Stadium seems less likely to be behind these splits.

Looking at the team numbers, the first thing that jumps out is the ERA split. The team had a 3.25 ERA in 506 innings played during the day. At night, that number balloons almost ¾ of a run to 3.99 in 952.1 innings. Yankee pitchers were 10% better than the league (by opponent On base Plus Slugging+ or OPS+) in day games and 6% worse than the league in night games. Their strikeout to walk ratio was 2.97 under the sun and 2.19 under the lights. Those splits are fairly extreme. Ace pitcher C.C. Sabathia is somewhat responsible for this trend. Sabathia had extreme day/night splits in 2011 and has a career opponent OPS+ that is 22% better during the day. Freddy Garcia and A.J. Burnett have both better during day games as well, but not to the extreme level of Sabathia. Overall, beyond Sabathia, the pitching doesn’t seem inherently prone to struggling at night overall as a number of pitchers including Phil Hughes, Ivan Nova, and Mariano Rivera have been better at night, generally Sabathia does explain the trend in play though since he has lead the team in innings pitched these past two years. Still, one pitcher is not enough to explain a .286 difference in winning percentage.

Yankee hitters were also significantly better during the day, hitting 20% better than at night by OPS+. Every element of their team slash line was up during day games, but OBP was the most significant difference; the team got on base at a rate of .364 during the day and .333 at night. This was almost entirely due to a higher batting average on balls in play (BABIP) though, with the team hitting .309 on balls in play during the day and .284 at night. That leads me to believe the offensive boost the Yankees got during day games was not skill driven. BABIP is a notoriously unstable number and with the difference in sample size between the two game situations, this is probably not a skill driven outcome.

It is hard to make a definitive conclusion as to why the Yankees were so much better under the sun. Day/Night splits have been somewhat ignored outside of the baseball gambling community (which pays a great deal of attention to all kinds of splits) and they are not typically extreme enough to catch our attention. The 2011 Yankees were an excellent team, hitting and pitching well overall. The fact that they were just a .500 ballclub at night may simply be a curiosity, the product of chance, but I doubt this. The team’s pitching was very different at night and a further examination of why this happened could potentially help the team when October rolls around again. If nothing else this extreme example may be able to teach the baseball community more about the differences between playing at night and playing during the day.

Around the Web