Exploring extra innings in the World Series

To celebrate Friday’s (Saturday’s?) 18-inning affair, let’s break down each of the 63 World Series games that ended in extras.

The Dodgers and Red Sox played a long game Friday night that set all sorts of records.

Los Angeles’ 3-2 victory in 18 innings set the record for longest World Series game by both time and innings, and it wasn’t close. The seven-hour, 20-minute affair marked the longest game in any championship series among all four major professional sports. Also, it set the record for longest MLB postseason game, breaking the record over the 2014 NLDS game between the Nationals and Giants by 57 minutes. That game was also 18 innings.

Interestingly enough, that very Nationals-Giants game inspired this article which I wrote in August 2017. Entitled “Exploring extra innings,” it was one of the most top trending articles on Beyond The Box Score as a result of the Dodgers-Red Sox marathon.

Today, I am going to continue to explore extra innings. Rather than look at extra innings trends during the regular season, however, I will completely focus on extra innings during the World Series specifically.

Thanks to help from fellow BtBS writer Daniel Epstein, I was easily able to break down each of the 63 extra inning World Series games.

As you can see, most World Series extra innings games end fairly quickly.

Of the 63 games, 32 of them — 50.8 percent — ended by the 10th inning. If the game was not over by the 10th, it was most likely over by the 12th — 93.7 percent of World Series games went no more than 12 innings.

Using the research that I did for my study last year, which encompasses about 5 ¹⁄₂ seasons worth of data, here’s how the World Series extra inning breakdown compares to the regular season extra inning breakdown.

Clearly, the World Series data matches up pretty well, considering it’s just a 63-game sample out of a 1,200 “population” data set, if you want to call it that. It’s quite interesting that nearly 94 percent of World Series games end by the 12th, compared to just 82 percent of regular season games. It’s hard to pinpoint exactly why this may be the case — perhaps it’s just due to the sample.

If you create a 90 percent confidence interval for the number of World Series games that should end by the 10th inning (as trying to do this for 12 innings would not make it normally distributed), the true proportion could be between 38.4 percent and 63.1 percent. Thus, it’s quite possible that this discrepancy is due to just a statistical error and sample size. There is no evidence that World Series baseball is no different than modern baseball in terms of games ending by the 10th inning. (Thanks to my AP Stat teacher, Mr. Grossman, for helping me with this!)

Anyway, considering that this is the 114th World Series in baseball history and that there have only been 63 extra inning games, this means that we see, on average, a total of one extra inning World Series game per every two seasons.

The 1991 World Series between the Twins and Braves is the only World Series in MLB history to feature three extra inning games. Kirby Puckett homered in the bottom of the 11th inning of Game Six to force Game Seven, which was then then won 1-0 in the bottom of the 10th on a walk-off single from Gene Larkin. That seems like it was quite the series.

While we know that Friday’s game was the longest in World Series history, the shortest World Series extra inning game was Game Four of the 1939 World Series between the Yankees and Reds. It lasted just two hours and four minutes and went 10 innings. At that rate, those two teams could have completed 35 innings (!) in the time that it took the Red Sox and Dodgers to complete 18.

The most runs scored in an extra inning game actually came last year in Game Five between the Astros and Dodgers. The 13-12 final featured 25 total runs. My favorite part about that game? The starting pitchers — Dallas Keuchel and Clayton Kershaw — were both former Cy Young award winners.

What do we do with this information? I’m not really sure. But it doesn’t really matter. All it does is reiterate a lesson that we seemingly learn every single day: you just can’t predict baseball.

---

Devan Fink is a Featured Writer for Beyond The Box Score. You can follow him on Twitter @DevanFink.