April 26, 2016
One Fine Day
In the fourth inning, Donaldson walked again. Bautista followed with a 6-4-3 double play.
In the sixth, Michael Saunders singled, and after Donaldson lined out, Bautista grounded into a 4-3 double play.
Three at bats, three GIDPs. How far was Bautista from a single game record? Pretty close, it turns out. A quick check from Baseball Reference’s Play Index indicates that the record is four, held by Joe Torre, back when he was a Met.
That struck me as odd. Not that Torre held the record—you’d figure it’d be set by a slow right-handed batter--but that it had been achieved only once. There’ve been over 210,000 major league games played since 1876, with 18 or more individual lines for each one of them. More precisely, there have been nearly 176,000 games played since the Retrosheet era, for which we have fairly complete (and Play Index-able) records, began in 1913. And in that time, Joe Torre, on July 21, 1975, is the only player to have had a day like this:
· First inning, following a single by Felix Millan, runner on first with one out: 1-4-3 double play
· Third inning, following a single by Felix Millan, runners on first and second with one out: 6-4-3 double play
· Sixth inning, following a single by Felix Millan, runner on first with no out: 4-6-3 double play
· Eighth inning, following a single by Felix Millan, runners on first and second with no out: 6-4-3 double play
After the game, Torre said, according to Tom Verducci’s The Yankee Years, "I'd like to thank Felix Millan for making all of this possible."
How often does that happen? How many single-game records are held by just one player?
The Play Index can answer that! Here’s a list of the single game record for each batting counting stat, listing the record, the number of times it’s been achieved, and the last player to have achieved it. I confined this analysis to American and National League games that didn’t go into extra innings:
Torre’s record is one of only three that are held by just one player. The others are Rennie Stennett’s seven hits (four singles, two doubles and a triple) in a 22-0 Pirate win over the Cubs in September 1975 and Jimmie Foxx’s six walks (he had an 0-2-0-0 line on the day) in Boston’s 12-8 win over the Browns in 1938.
Notably, we’re primed to have a new one-instance single-game record any day now. Maybe. The fact that there are 205 instances of seven at-bats shows how frequently players get close, as does the fact that there are eight of eight plate appearances—in other words, the exact circumstances required for this to happen, but with a walk or sacrifice having spoiled the effort. Three no. 8 hitters have had seven ABs in a game, and six no. 7 hitters, so the leadoff man has been close enough to smell it a number of times. But when it happens, those same factors might also unsingularize it immediately, if the no. 2 hitter, or no. 3 hitter, or no. 4 hitter, or so on gets to bat an eighth time. That same reason is why our 205 record-holders for ABs in a game are congregated in just 157 total games.
That more players haven’t had seven hits, like Stennett did, is interesting. On the one hand, the odds of a .300 hitter going 7-for-7 (the only way to do it, given no 8-AB games) are around 1-in-5,000, and we haven’t had anywhere near 5,000 7-AB games. But we also know one fact about those 7-AB games that mangles the odds up: There were tons of hits! So without knowing anything else, we can probably assume the guy who got seven at-bats in a game was probably, on that day, much better than a .300 hitter. The team as a whole, after all, had at least a .500-or-so OBP. If we treat the 7-AB guys as .400 hitters on those days, then the odds go up to 1-in-625. So maybe having one, and only one, is about what to expect.
There are several single-game records for pitchers that are held by just one player, which isn’t surprising. For one thing, the numbers are a lot higher, and thus there’s a lot more space for similarly bad or good performances to be one, two, three off from each other. For another, most of these come from an era when pitchers were used in a different way, and so there’s simply a lot fewer trials. For instance, from 1913 to 1929, there are 50,000 or so pitching lines from which to draw our repeats. But from 1913 to 2016—the period during which our hitting records are fairly evenly spaced out—there are a few million batting lines. So here’s the list of single-game records for starters:
Was Babe Ruth the Jon Lester of his day, attracting bunts in order to exploit his throwing? Well, he did make seven errors in 1918, including two on May 4, when he set a record for most sacrifices allowed. Might be a factor, but then so too might be the fact that sacrifice flies and sacrifice bunts were lumped together as a collective “sacrifice hits” by statkeepers at the time.
The single game records for relievers:
There are a couple of interesting solely-held records among the relievers. A relief pitcher, Lefty O’Doul, holds the record for most unearned runs allowed in a game, starter or reliever. In the first game of doubleheader against Cleveland in 1923, he pitched three innings, giving up eleven hits and eight walks in a 27-3 Indians victory, including a record 13 runs in sixth. Sent to the minors after the season, O’Doul gave up pitching after a year, returned to the majors as an outfielder in 1928 at age 31, and went on to win two batting titles and compile a .349/.413/.532 career split (.353/.417/.539 after giving up pitching).
The inverse of O’Doul’s day in 1923 was Randy Johnson’s 78 years later. On July 18, 2001, the Padres’ game against the Diamondbacks was suspended in the top of the third when a light tower at Qualcomm Stadium exploded. At the time, Arizona led 1-0 and starter Curt Schilling had thrown two perfect innings. When the game was resumed the following day, Johnson took over, and in seven innings, he allowed one hit, one walk, and struck out 16.
What records could fall? I think there are three pretty good candidates.
Batter strikeouts: Current record five, held by 59 players. With strikeouts rising, it seems a safe bet that at some point, all those platinum sombreros will be eclipsed by a titanium one. There have been eight six-strikeout games in extra innings. It’s not hard to imagine one in nine innings. After all, strikeouts are almost exactly as common as hits these days, so if a player can get seven knocks in a game—and regularly get six—there’s no reason one can’t collect six (or seven) Ks.
Batters hit by pitches: Current record three, held by 21 players. Of the 21 times a batter’s been hit by a pitch three times in a game, 14 have occurred since 2000. This is not a coincidence.
Starting pitcher home runs allowed: Current record six, held by eight pitchers. It’s hard to imagine relief pitcher records falling, given that the average relief stint so far this season is one inning (1.03, through games of April 21, to be precise). But one of the consequences of one-inning relievers is that starters don’t get yanked early from their worst starts. In 1990, 199 starting pitchers were pulled before completing two innings. It’s happened more than 100 times only twice in the past six years. Without multi-inning relievers, starters are left in to take their lumps. That’s within reason, of course. In the second game of a late September doubleheader in 1915, the A’s were en route to a 43-109 record, last in the American League and 58.5 games behind the Red Sox. Rookie pitcher Tom Sheehan pitched a complete game in which he allowed 20 runs on 23 hits, four walks, and a hit batter. That won’t happen again. But to surpass the record of six home runs allowed in a game, held by James Shields and R.A. Dickey, among others, a pitcher could go, say, five innings and allow eight runs. That’s occurred 25-32 times in each of the past nine seasons.
And Bautista? He had a chance to tie Torre’s four GIDP record, coming to the plate with one out and runners on first and second in the eighth inning. But he was thwarted by an event almost as rare as the four-GIDP game: Koji Uehara walked him.