A few weeks ago, two games ended unusually. On Thursday, August 31, the Twins’ Max Kepler faced the White Sox’s Juan Minaya in the last of the ninth inning, with the bases loaded and two outs in a 4-4 game. Minaya hit Kepler with his first pitch, forcing in the winning run. It was a walk-off hit by pitch.

Two nights later, the Mariners’ Mitch Haniger faced Oakland’s Blake Treinen in a similar situation: Last of the ninth, bases loaded, two outs, tie game (6-6 in this case). Ahead 0-1, Treinen threw a wild pitch, allowing Jean Segura to score the winning run from third base. It was a walk-off wild pitch.

I decided to determine how common walk-offs are in general, and what kind of walk-offs occur the most frequently. I mostly used the Baseball-Reference Event Finder. It tracks walk-offs only for plays in which the batter completes his plate appearance, so I used other sources for walk-off wild pitches, walk-off passed balls, walk-off errant pickoffs, walk-off stolen base attempts, etc.

I looked at walk-offs this season, as well as over the entire 30-team era beginning in 1998. Through Saturday’s games, there have been 192 walk-offs this year and 4,242 since 1998. That works out to a walk-off every 11.6 games this year and every 11.4 games since 1998.

Below you'll see the frequency of walk-offs, starting with the most common in the 30-team era. Note that often a walk-off is the product of more than one event, like a single with a runner on first base, after which the outfielder airmails his throw to third base, allowing the runner to score. In cases like that, I considered the batter’s action to be the key, scoring plays like that as walk-off singles rather than walk-off errors. After all, the batter’s going to be the guy who gets water dumped on him.

Singles: 80 in 2017, 1,920 since 1998. Melky Cabrera is the walk-off single champion, with 11 since 1998. Mark Trumbo has three this year to lead the majors. There are four teams with six walk-off singles this year, and they’re not the ones you’d expect: Angels, Braves, Phillies, and White Sox.

Homers: 68 in 2017, 1,334 since 1998. Manny Machado leads the majors with three this year. The 30-team era king is Albert Pujols, whose 12 walk-off homers give him one more than Jim Thome and David Ortiz. The unlikely 2017 team leader is Oakland, with seven.

I also looked for a subset: Walk-off grand slams with the home team trailing by three runs. There have been nine in the 30-team era, most recently the Blue Jays' Steve Pearce against the Angels’ Bud Norris on July 30 of this season. Six occurred with one out, and three with two down. Two—the Braves' Brooks Conrad vs. the Reds' Francisco Cordero on May 20, 2010 and the Astros' Brian Boqusevic vs. the Cubs’ Carlos Marmol on August 16, 2011—were struck by pinch-hitters.

Doubles: 17 in 2016, 304 since 1998. Adrian Gonzalez has five in his career, the most in the 30-team era. Fenway Park accounts for 17 of the 306 walk-off doubles since 1998, the most in the majors. Given the park’s dimensions, that isn’t a surprise.

Sacrifice flies: 7 in 2017, 204 since 1998. Carlos Beltran and Joe Crede, one of whom is a strong Hall of Fame candidate, have hit three, the most since 1998. Somehow, a walk-off sacrifice fly doesn’t capture the excitement of a walk-off the way a base hit does. Well, get ready for a bunch of even less exciting walk-offs.

Walks: 9 in 2017, 133 since 1998. This is another low-wattage kind of walk-off, isn’t it? When the home team comes storming out of the dugout, on whom do they throw the Gatorade and talcum powder? The guy who scores? The guy who walked? The pitcher? The umpire?

Mike Cameron and Jorge Posada have three each since 1998. Two fun ones: Adam Dunn, on April 2, 2014, and Yuniesky Betancourt, on August 31, 2010, had walk-off walks with runners on first and third. In both cases, the pitcher threw a wild pitch for ball four, allowing the runner from third base to score the winning run.

Errors: 1 in 2017, 86 since 1998. (Note: To avoid double-counting, this does not include players who reached base on an error but were credited with a sacrifice hit or sacrifice fly.) Walk-offs are thrilling for the home team. They are less so for the visitors, and this must be one of the most painful ways to watch your team lose. The leaders since 1998 are Mark Loretta, Rod Barajas, and Geoff Jenkins, with three each. Here’s an example:

Wild pitches: 4 in 2017, 63 since 1998. So when Segura scampered home against the A’s, it wasn’t all that unusual.

Fielder’s choice: 0 in 2017, 50 since 1998. Here’s a bar bet you could win if you can find anybody to take it: This might be the first season in the 30-team era with zero walk-off fielder’s choices. You can imagine how these work: Runner on third base, batter hits a ground ball, fielder (obviously) goes for the play at the plate rather than the play at first base. For instance:

Triples: 0 in 2017, 35 since 1998. Yeah, I had the same reaction. A walk-off triple? What’s the story with that? Did the runner on first trip, or was it Jose Molina? Shouldn’t a double suffice to score the winning run?

Well, three of them occurred with nobody on base. In each case, the cutoff man threw wildly, allowing the batter to score with the help of an error, and the same happened once with a runner on first base. In every other case, there was a slow guy on first base (e.g., Michael Cuddyer) or a speedy batter who made it to third base quickly (e.g., Rajai Davis, the only batter since 1998 with two walk-off triples).

Outs on bases (other than sacrifices): 3 in 2017, 33 since 1998. As you might expect, there’s more to these than, say, a simple ground ball to shortstop. On April 19, Tampa Bay’s Logan Morrison batted with one out and the bases loaded in the bottom of the ninth inning, down 7-6. He hit a grounder to second base, forcing Brad Miller at second as Kevin Kiermaier scored the tying run. The shortstop’s wild throw to first base on an attempted double play allowed Peter Bourjos to score the winning run.

On July 29, the Red Sox’s Eduardo Nunez batted with one out in the bottom of the 10th, runners on second and third. He grounded out to shortstop, but Sandy Leon, who’d held at third base, broke for home and scored on an acrobatic slide:

You get the idea. These are pretty crazy plays, attributable to fielding and baserunning rather than batting.

Hit by pitch: 1 in 2017, 26 since 1998. What Kepler did wasn’t unique. In fact, there were two walk-off HBPs in August of 2016: Yadier Molina against the Reds on the eighth, and Rougned Odor against the A’s on the 16th. Longtime Effectively Wild listeners will not be surprised to know that the leader in career walk-off HBPs in the 30-team era is Jonny Gomes, with two.

Sacrifice hits: 1 in 2017, 25 since 1998. Wouldn’t you have thought there would be more walk-off bunts than walk-off hit by pitches and walk-off triples? I sure did. And here’s the craziest part: There was a runner on third base in only six games. There were times in which there was a runner on first base who scored: Three E1s, two E3s and an E2. Similarly, there were six games in which a runner on second base scored on a throwing error, and seven with runners on first and second, the lead runner scoring on an errant throw.

Those totals apply only to plays scored as sacrifices. There were 15 other bunts (including one this year) that resulted in a walk-off. Eleven were scored as a single, three as a reached on error, one as a fielder’s choice, and none as sacrifices.

Balks: 0 in 2017, 11 since 1998. This isn’t about the batters, it’s about the pitchers. They are, alphabetically: Al Alburquerque, Taylor Buchholz, Santiago Casilla, D.J. Carrasco, Aaron Crow, Justin Duchscherer, Keone Kela, John Rocker, Mike Stanton, Esmerling Vasquez, and Jeff Zimmerman.

Passed balls: 0 in 2017, 8 since 1998. Here’s the most recent:

Errant throws to bases: 0 in 2017, 5 since 1998. These are fun. As with balks, no pitches were thrown! The throws that allowed runners to score were uncorked by catchers Henry Blanco, Mike Matheny, Derek Norris, and Keith Osik; and pitcher Brandon Lyon (runner scored from second base).

Strikeouts: 1 in 2017, 4 since 1998. Let’s just list them all:

Jeff Sullivan of FanGraphs discussed the latter walk-off when it happened. His title was, “The Angels Won on a Walk-Off Strikeout*,” explaining the asterisk by writing: “If we want to be totally honest, the Angels didn’t win on a walk-off strikeout. They won on a walk-off error that took place in the immediate aftermath of a strikeout.” Which is, yeah, technically true, but what’s the fun in that?

Pitcher drops the throw back from the catcher after a pitch: 0 in 2017, 1 since 1998. Really. It happened on August 11, 2005:


Summing it up, here are the walk-off champions (counting only plays that complete a plate appearance):

Thanks to Rob McQuown for research assistance.

Attention Play Index truthers: The Event Finder includes walkoff fielder’s choices in the Reached on Error bucket. Walkoff home runs don’t include two inside-the-parkers. Outs on Bases includes eight reached on error.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
Fun article, Robs!
Thanks, John! I appreciate it. I was pleased with the end product, though I do not recommend trying to suss out some of these buckets if you're looking for something fun to do.
Nice article!

In the period you studied, did any hitter have a walk-off in each half of a double-header? We had Blake Perkins do it in Hagerstown this past season- my back-of-the-envelope calculation pegged this as a once-in-50-year event or rarer in any given league and maybe once a decade over all 5 full-season leagues (MLB, AAA, AA, A+, A) combined.

Of course, the odds were tilted slightly since we had so many doubleheaders this year to make up for rain-outs!

What about walk-offs by the same hitter in consecutive games, games in the same week, etc.?

Thanks for reading it!

I looked only at walkoffs in aggregate. But you've given me some ideas!

Just to hazard a guess, I'll bet you that there have been cases of batter getting a walk-off in both ends of a DH, but not since WWII. There were a *lot* more doubleheaders before the war.
Would the "Fielder's Choice" include the case with a double play possibility (e.g., first and third, one out) where the out is recorded at second but the batter beats the throw at first?
Good question. I should note up front that my definitions were somewhat circumscribed by what was available in the dataset I used.

Here's an example: SF at WAS, 7/5/12, tied 5-5, last of the ninth, bases loaded, one out, LaRoche at the plate. He grounds to second, forcing the runner from first, but LaRoche beats the throw to first, allowing Harper to score. Using the Play Index Event Finder, that's in the "Non-SO Outs (to Pos)" category, and the description reads "Groundout: 2B-SS/Forceout at 2B; Harper Scores."

I know that this is splitting hairs to an extreme. Yes, we'd score that play a fielder's choice. But I think that in this unique case--game-ending run threatening to score--I'm OK with discerning between a play like the one above, which I put in the Outs on Base bucket (and didn't have an automated way of doing otherwise), and one where the fielder takes a grounder and ignores the batter, firing the ball to the catcher in an attempt to get the runner trying to score, which I put in the Fielder's Choice bucket.
Thanks for the explanation. This is a very fun article to read. Now we just need a catcher's interference walk off. (I remember the crazy Jason Kendall dash to home when Francisco Rodriguez muffed Jose Molina's return throw to the mound. One of the most unique plays I've ever seen.)
I completely agree--I was disappointed to see that the only walkoff catcher's interference in the Event Finder database was on August 1, 1971, long before the 30-team era. (Reds at Dodgers, Willie Crawford awarded first in the bottom of the 11th and the bases loaded in a 4-4 game).