Last night, Evan Longoria came to the plate twice with men on first and second once and the bases loaded once. In both instances there were 0 outs. In both instances he preceded to ground into a double-play.
After Carl Crawford's "single," the Rays win expectancy was 69.1%. When Longoria grounded into the double-play, their win-expectancy dropped to 59.3%. Had he struck out, the Rays would have had a 63.1% chance of winning the game. Longoria's GIDP was drastically worse than if he had struck-out.
Longoria's third trip to the dish followed a CC infield single. Once again the young slugger was put in a great position to extend a lead and keep a rally going. With the bases loaded and no one out in the 5th, the Rays should win the game against the Red Sox 79% of the time. Longoria once again GIDP, at least providing one run of benefit this time. The win expectancy still dropped to 73%. Had he merely struck out, the win expectancy would have lied somewhere between the two. The win expectancy estimator has for all games from 1977-2006, has the drop from 82.8% to 81.9% while the GIDP drops it to 80.9%. Thus in all games played over those 30 years, the double play was twice as costly.
As we saw, the Rays went on to win the game anyway (and Longoria helped with a homer and a great defensive play), but Longoria needs to learn to avoid the double play in these situations. They truly are a rally killer. Players like Jim Rice are Hall of Famers because they believed that "putting the ball in play" was there job when there were runners on. Today, we know that merely putting the ball in play can sometimes hurt more than it helps, and it may be better to just keep the bat on your shoulder or just try to put one in the outfield.