Hang it in the Louvre: Oh, and this one too: The playoffs were absolutely wild this weekend. Out of the six games played Saturday and Sunday, two were all-time classics. First, the Yankees and Royals traded blows before Alex Verdugo produced a game-winning single after a controversial stolen base call. Then the Phillies and Mets traded home runs and blown leads right up until the last play of the game, Nick Castellanos’s walk-off hit.
If you wanted to, you could read our game stories for these games, or any number of other fine pieces about them across the internet. You could watch highlights or condensed recaps. But this is FanGraphs, so I thought I’d cover another angle: where these games fit in the history of wild playoff games.
We have win probability charts going back to 2002, which means we have data on total win probability changes going back to that year as well. If you take the absolute value of these and sum them up, you can see exactly how much each team’s fortunes changed throughout the contest. The more total win probability changes, the wilder things are.
For example, the least exciting game by this measure occurred on October 9, 2019. The Cardinals beat the Braves 13-1 in the NLDS, and they opened things up by scoring 10 in the first inning. No drama, and thus very few changes in win probability. A 2023 contest between the Diamondbacks and Dodgers (11-2 Arizona, 9-0 after two innings) is the runner up.
On the flip side of things, the wildest playoff game of the 21st Century happened in the COVID-shortened 2020 season. The Braves and Reds played a game as frustrating as it was classic, a 13-inning affair that ended 1-0 on a Freddie Freeman walk-off single. That game obviously didn’t feature a lot of lead changes – just one, to be precise – but playing an extra half-game of tied, last-inning baseball really amps up the drama.
If you’re putting everything in terms of how much the odds of winning the game changed, that Braves-Reds game was the equivalent of 13 Cardinals-Braves games from the year prior.
Why am I telling you all of this? To set up how bonkers this weekend’s two contests were. For a sense of scale, the Cardinals-Braves laugher featured a grand total of 0.6 WPA change. It started at a 50% likelihood for each team to win and went more or less straight to 100% Cardinals (0.5.
WPA change) with minimal deviation on the way. The extra-inning Braves-Reds thriller featured a whopping 7.92 cumulative change in win probability. It started in the middle and then see-sawed wildly, on and on: The average playoff game has a sum change in win probability of around 2.6, to give you a rough idea of the numerical scale.
To instead give you a rough idea of what that feels like, think of Saturday’s Mets-Phillies game, a tight but orderly 1-0 Philly lead through seven innings. The Mets broke through for five in the eighth and held on uneventfully. Some drama, an exciting lead change, and nothing more: That’s the average playoff game.
Game 1 of the Royals-Orioles Wild Card Series, which Kansas City won 1-0, and of the Mets-Brewers Wild Card Series, which New York won 8-4 after erasing a 4-3 deficit with a 5-run fifth, have been roughly average in terms of total excitement as well.
This weekend’s two wild games featured far more twists and turns than average. To be specific, Yankees-Royals Game 1 had a total WPA change of 4.415, and Phillies-Mets Game 2 had a 4.414 WPA change. Where do they stack up out of all the playoff games that have been contested since we have play-by-play win probability data? They’re 56th and 57th out of 797, respectively — among the top 8% all games in the set.
That might not sound very impressive, but both games managed to pack all those reversals of fortune into a regulation nine innings. Extra innings are more or less a cheat code for chaos; more plays mean more chances for each team’s likelihood of winning to change, especially when every inning could be the last one of the game. It isn’t always the last inning, of course, but WPA doesn’t look back; it measures each play without knowing what will happen the rest of the way. Every single home half of an extra inning is monumentally important. That’s how it feels in real life, too.
To be clear: I’m not saying this is a bad thing, but it explains why the top 12 games on the list all went to extras. If we limit ourselves to what happened in the first nine innings, the two games from this weekend stand out. I left all the extra-inning games in our sample and just lopped off everything that happened after the ninth.
With that equal footing, these two games are 16th and 17th out of the 797 games in our database. We haven’t seen a wilder regulation game since Game Four of the 2020 World Series. You might remember that one, even if you don’t remember the series overall. The two teams traded the lead back and forth, with the Dodgers scoring in six different innings and the Rays scoring in five. There were three lead changes and a tie in the last three innings, capped by Brett Phillips sending everyone home with a two-out, two-strike walk-off single.
That game gets undersold because it happened in the weirdest year of baseball in my lifetime. Even more than that, the 2020 Rays are one of those forgettable pennant-winning teams; we remember the Dodgers won the World Series that year, and that they lost a close game during it, but those Rays are nothing more to our memories than the maddeningly non-specific unnamed opponent. This weekend’s games probably won’t fall into that same trap. This Mets run is storybook stuff already. The year the Yankees got Aaron Judge and Juan Soto to team up is going to be one we remember, too.
And each game had its own hook, which adds to the ease of remembering. The Phillies-Mets game is fifth in the whole sample in terms of WPA change from the sixth to ninth innings. In other words, during the part of the game that we expect it to be at its most exciting, it came through in spades. There were two ties and three lead changes. Castellanos and Mark Vientos each delivered more win probability than the total amount contained in that Braves-Cardinals blowout I’m using as the low end of the scale.
The highest-leverage moment for each team produced something brilliant: Bryson Stott’s eighth-inning triple and Vientos’s game-tying, two-run homer in the ninth. This one followed a classic arc: a slow build-up of stakes in the early innings followed by a series of impactful clashes that culminated in a walk-off.
The Yankees and Royals, in contrast, spread out their drama. Only one run scored in the last three innings. The Yankees allowed just a single baserunner during innings 7-9, a two-out Garrett Hampson walk. There wasn’t even a bottom half of the ninth to allow for a walk-off. But the middle innings of this game were jam-packed with hairpin turns. No game that we have data for featured more change in cumulative win probability in the fourth through sixth innings.
It’s not hard to see why. MJ Melendez’s go-ahead two-run homer in the fourth inning got the party started. The Yankees tied the game in the fifth on a bases-loaded, no-out walk and felt like they might run away with things. The Royals got two huge outs to wriggle most of the way out of that jam, only to give up another walk to put New York back on top. Not to be outdone, however, Kansas City stormed back in the top of the sixth. After Anthony Volpe threw the ball away to put runners on second and third, Hampson singled to drive them both home to put his team back in the lead. The Yankees didn’t need much time to equalize; Austin Wells laced…
