And Then There Was One

“Say it ain’t so, Joe.” – unknown little boy allegedly outside of the courtroom

When I ran my simple series simulator 100,000 times for the 2005 World Series the White Sox swept the Astros about 9% of the time. That’s a pretty healthy percentage, but even so the simulator lacked the ability to take into consideration the great run of pitching the White Sox enjoyed, their good fortune with the boys in blue, and their knack of getting the big hit at the right time. Those all combined to lead the White Sox to an 11-1 postseason run in which they outscored their opponents 67-34; and if you stretch back to the regular season, a run of winning 15 of their final 16 games.

And now Shoeless Joe Jackson can rest.

Dave Studeman has written an excellent article highlighting the White Sox season, so in this article I’ll briefly look at Game 4 from a Win Probability perspective.

The first thing you’ll notice when you look at the graph below is that in a scoreless game the win probability oscillates back and forth, usually between around 40% for the visiting team in the bottom of the inning to 50% at the top of the inning, and then nudging over 50% as the visitors put men on base. The reason for this is obviously because when the home team bats the visitors have already had their chance in the inning, so with more opportunity comes a higher probability of winning.

In last night’s game the White Sox nudged themselves over 54.2% on Jermaine Dye’s leadoff single in the 3rd but their chances quickly dropped 10.2% as Brandon Backe struck out Paul Konerko, A.J. Pierzynski, and Aaron Rowand to retire the side. The two White Sox low points in the middle innings came when Brad Ausmus singled to lead off the bottom of the 5th when their win probability stood at 38.9%, before an Adam Everett double play increased their odds by 9.6% to get them back near 50%. In the 6th, the walk to Lance Berkman with Willy Taveras on second base lowered their win probability to 37.4%. The strikeouts of Morgan Ensberg and Jason Lane served to level the playing field at 50% once again.

The balance changed of course when Willie Harris led off the top of the 8th with a line drive single to left, pushing the Sox to 56.7%. Although the sacrifice by Scott Podsednik actually lowered Chicago’s win probability by 2.4% and a subsequent groundout by Carl Everett dropped it to 48.1%, the Dye single that followed scored Harris and was the biggest play of the game, raising their win probability from 48.1% to 71.4%.

The Astros, however, had an excellent scoring opportunity in the 8th when Willy Taveras was hit by a pitch, advanced to second on a wild pitch, followed by an intentional walk to Berkman. Alas that would be the Astros’ high point, as a fly to center by Ensberg and a ground out by pinch hitter Jose Vizcaino combined to raise the Sox’s win probability 24.1% to 83.4%. The leadoff single by Lane in the 9th registered a 13% swing and got the Astros to 33.7% and temporarily raised their fans hopes, but Chris Burke-tober couldn’t work any magic; nor could Orlando Palmeiro and the Astros went down on the strength of two outstanding defensive plays by Juan Uribe that registered +13.9% and +14.7% to bring the World Championship to the Windy City.

For the game Freddy Garcia led the World Champs in win probability added at .349, while Dye, who would get player of the game and World Series MVP (.438/.526/.688) honors came in second at .280. Konerko left runners on in the first, fourth, and eighth innings and so was at the bottom for the Sox at -.136. For the Astros there were no shortage of negative win probability added values with Ensberg in the cellar at -.214 and Adam Everett at -.160. For Ensberg and Everett it was a World Series to forget, as combined they were 3 for 33 with 2 RBIs.

One of the interesting aspects of Game 4 and Astros manager Phil Garner’s strategy was his insistence on bunting and neglect of the running game. In Game 4 the Astros sacrificed twice. The win probabilities for those events were:

Inning   Batter         Play      WP
   1     Taveras  54/SAC/BG    -.015
   9     Ausmus   34/SAC/BG    -.051

What’s interesting about this is that the sacrifice in the first actually registered less negatively than the sacrifice in the 9th. While it makes no sense to give away outs in the early innings despite what Joe Morgan or Tim McCarver might say, I was a bit surprised at first that the sacrifice in the 9th was so low. The reason is that the cost of losing one of the three remaining outs in the 9th inning, even with only a one-run deficit, far outweighs the gain from moving a runner only to second base. Had the runner been moved to third base the win probability for the Astros would have risen by 8.2%.

One of the lessons of the Red Sox 2004 championship discussed in the book Mind Game is that speed is a strategic weapon and using it at the appropriate time can make all the difference. The 2005 Astros didn’t use the speed they had strategically (Taveras had one stolen base in the postseason) nor did they structure their roster to allow for using speed when they needed it most. As a result they wasted outs in sacrifice attempts—outs they couldn’t afford to lose in the context of a struggling offense.

References & Resources
Mind Game: How the Boston Red Sox Got Smart, Won a World Series, and Created a New Blueprint for Winning – Baseball Prospectus writers edited by Steven Goldman, 2005