It’s officially fall on the calendar, which means they are about to lie start lying to you. The last weekend of the regular season is upon us and the playoffs are right behind that. It’s the fun time of the year when individual games can mean the end of someone’s season or a giant trophy. It means we’re about to watch some particularly emotional games in the next few weeks, too. A season’s worth of hopes and dreams may come down nine innings. It also means we’re about to indulge in one of baseball’s most time-honored traditions: absolutely baseless claims about the stretch run and the postseason, how they work, and how one can predict them.
Pro tip: You really can’t.
The playoffs aren’t exactly a crap-shoot, because craps relies on rolls of the dice that are totally random. Someone’s a favorite in each playoff series, so it’s not a 50/50 shot, but it’s a lot closer to 50/50 than most people want to believe. Then again, that’s the fun of the playoffs. “Anything can happen” is more or less true. But people don’t like uncertainty, and so to cope with what is existential randomness, we begin seeing patterns where none exist. Well, the broadcasters do, and since you don’t have a microphone, you just have to accept what they say.
Maybe that’s fun for you. But, maybe you want to be that fan. The one who says, “None of this is actually real and the numbers can prove it.”
Well …
Warning! Gory Mathematical Details Ahead!
Myth 1a: Teams that lock up their playoff berth early are able to rest regulars and therefore have an advantage in the postseason.
Myth 1b: Teams that have to battle down the stretch for a playoff spot are able to keep the edge they need to perform at a high level in the postseason.
You’ll hear both of these down the stretch and as the playoffs begin. You’ll somehow think that both of them are true when someone says them. When you put them next to each other, do you notice something about them? The problem is that they can’t both be right. What’s more, they’re both wrong.
I’ve previously looked at how many “meaningful games” (defined as being within three games of an available/non-clinched playoff spot in either direction) teams played in their final 15. It turns out that when they get to the playoffs, players on teams that coasted in and players on teams that fought all the way down the stretch perform pretty much like we would have expected them to. The playoffs are a collection of good teams, which means that good hitters are more likely to be facing off against good pitchers, and not the random Triple-A filler that the rebuilding team in their division was throwing out there that helped pad his stats a bit during the regular season.
Batters hit a little worse than we might expect based on their regular-season record, and pitchers pitch a little worse. It’s what happens when an irresistible force meets the immovable object. It made no difference what their team did down the stretch. Even more, when you look at how teams do in terms of winning series, the answer is that teams that played no meaningful games down the stretch tend to win about half of their series. Teams that fought down the stretch? About half of their series.
How you get to the playoffs has no bearing on how you will do once you get there.
***
Myth 2: A big come-from-behind, walk-off win in a playoff game gives teams momentum.
Nope.
The day after a game in which a team comes from behind in the eighth or the ninth inning, their hitters and pitchers perform exactly in line with their regular-season stats, although between 1995-2017, teams that won a game in this way won about 59 percent of their next games. That might seem like a win for “momentum,” but it’s mostly explainable by the fact that teams that win a playoff game in any fashion are about 54 percent likely to win the next one against the same team, and teams that win walk-offs are more likely to play at home the next game, because walk-offs happen for home teams only.
If there’s a “momentum boost” it’s much smaller than people make it out to be. The reason is that while we always look at the team jumping up and down after hitting the walk-off home run, we never stop to think what the other team does when they get into the clubhouse. Sure, it probably stings for a little while to lose a game like that, but you move on with life.
Consider what happens after something actually bad (like, in the real life sense, not losing a game) occurs. It doesn’t mean that you’re going to be down for the rest of eternity. Humans have ways to come back from adversity. And on the other side, the team that hit that home run can’t sustain an emotional high forever. Eventually, it’s 0-0 in the top of the first inning the next day. Sometimes—in fact, about half the time—the team with “the momentum” wins, and so you have an example where you can say, “See, momentum works!” But that’s not how things actually are.
***
Myth 3: I’ve figured out the way to win in the playoffs.
And now a brief list of things that have been “proven to win” in the playoffs:
- Having an ace pitcher.
- Having two or three shutdown relievers (alternately, having Andrew Miller).
- High contact percentages.
- Three True Outcomes guys.
- Magic.
- Having a group of experienced veterans who have been in the playoffs before.
- Having a group of young players who are hungry.
There are ways in which a playoff roster plays differently than a regular-season roster. High-leverage relievers work more often and for longer stretches (see Miller, Andrew). Your fifth starter is now a sponge/long reliever or a tandem starter. There’s room on the bench for a guy who’s mostly a pinch-runner because your fifth outfielder isn’t needed to give someone a day off during the series and that’s all he usually does. So, in some ways there are rosters that are more built for the playoffs than others, but other than the bullpen thing, they are largely at the margins.
The types of “strategies” that get talked about this time of year are either or one of two things. One is a a variation on “have good players,”which isn’t a strategy. (Quick test: does the opposite of it sound silly? Then it’s not a strategy, it’s a goal.) The other is a contradictory statement that is simultaneously believed along with its opposite (experienced players are better vs. young players are better). Most of those contradictory statements are in the “they’re both wrong category.
***
Myth 4: There are some guys who are “good at October.”
If you tried to sneak this one by me as “there are some guys who have been good in October” then I might allow it. There are guys who have had some good October games in the past. The cruelty of October is that if someone has a bad series, it’s more likely that his team will lose and he will get no more chances to redeem himself. If he has an amazing series, it’s more likely that his team will win the series and he will get more games in which his performance is likely to regress back to his true talent level. Eventually, everyone disappoints in the postseason. There’s also the obvious sample-size issue around evaluating previous October appearances. Even for someone who’s been around for a while, we might be talking about 20 or 30 games total. And they might have been five years ago.
There probably are players who are better or worse at handling the pressure and the lights. As a psychologist, I believe this wholeheartedly. The problem is figuring out who those guys are from that limited sample. The difference between a .260 hitter and a .300 hitter is one additional base hit per week. If you randomly hit a blooper that isn’t caught during a playoff series, you’re going to look amazing. There’s just too much noise for us to get a good fix within the numbers on who might be good at this whole October baseball thing and who might be bad.
The problem with quoting numbers is that people tend to assume all historical numbers are equally indicative of future performance. During the regular season, eventually if someone gets to be .300 hitter over the course of a few months, we’re not surprised if he hits like a .300 hitter going forward because we have a good sample on which to base his performance. Stay out of that trap. Random small-sample splits are not the same thing as good research methodology.
***
Myth 5: There’s any team that’s a favorite to win the World Series.
Sure, some team has probably got the best chance, but of the eight teams that make the Division Series, they all pretty much have a 10-15 percent chance of emerging as the champion. Maybe one’s got a 20 percent chance. It means that they’ve got an 80 percent chance of going home sad.
It’s worth remembering that if you look at a playoff team in isolation, they’ll look pretty good. They are good, but let’s remember that some of the 90 wins they put up were against teams playing for next year. Everyone else in the playoffs is good.
Even if you had two teams that were such a mismatch that one of them would win 60 percent of the games played against each other (that would be a 97-65 record over 162 games), you’d still see the underdog pull a series upset 31 percent of the time in a best-of-five scenario. A few years ago, after an Effectively Wild e-mail, I estimated how good a team would have to be to register as a 50 percent favorite to win the World Series at the outset of the playoffs. I came up with a team that had the quality to win 113 games during the regular season.
On the flip side, it really does mean that anything can happen in the playoffs. So, my recommendation is that you just enjoy the ride.
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
I tend to think that most of these players are weeded out early. If that were true, it would provide small support for the "experienced players are better" theory, as experienced players would be less likely to succumb to the pressure now since they didn't before.
On the flip side, assume that all professional athletes are capable of handling all routine pressure they face, even a World Series. Even then, the person going through a rough divorce, family illness, or undiagnosed condition of some sort might drop below that threshold at an untimely moment. I infer from this that the effect we'd see would look a lot like random noise ...