The Surprising Performance of Marginal Pitchers in Big Spots

The last two World Series games have featured pitching best described as aspirational. The Braves used three pitchers who barely pitched in the majors to face 37 Astros, and the results were mixed. Sometimes you’re Kyle Wright, throwing 4.2 clutch innings to keep your team in the game. Sometimes you’re Tucker Davidson, coughing up a four-run lead in two-plus brutal frames. That’s the nature of pitching, but it’s not exclusive to those fringe arms, as this postseason has shown.

Consider the worst pitchers that teams brought into these playoffs. Fourteen relievers who posted a regular-season FIP of 4.50 or higher have appeared this postseason, pitching a combined 44.2 innings. They’ve been awful! They’ve combined for an ERA of 6.00 and a FIP of 5.30. That’s the kind of pitching that sends you home early.

It’s also better than you’d expect! After all, these pitchers aren’t great. That’s how I selected them — pitchers with bad numbers during the regular season. In fact, this group combined for a 5.52 FIP in the regular season, weighted by the number of postseason innings they’ve each accounted for. They’ve actually done better in the playoffs — perhaps the reason why teams selected them in the first place.

Widen the criteria a bit, and the same thing remains mostly true. Pitchers who posted a FIP above 4.00 in the regular season have thrown 170.1 innings and allowed a 4.97 ERA and 5.05 FIP. In the regular season, they allowed a 4.54 FIP. They’ve been worse — but that’s a pretty broad category, including new Atlanta folk hero Will Smith and sometimes-ace Framber Valdez.

You can criticize sample size all day, but the results speak for themselves. Bad pitchers in the regular season are, more or less, bad pitchers in the postseason. If you prefer that bigger sample, they’re half a point of FIP (and a point of ERA) worse in the playoffs than the regular season.

Given that, you might be surprised to know that I think “bad” pitchers have pitched effectively this postseason. Didn’t I just say they did worse in aggregate? I did, but consider the other side of the distribution. Eighteen pitchers who posted a regular-season FIP of 2.75 or lower have combined to throw 85 innings of playoff baseball. They’ve been great, pitching to an aggregate 2.96 ERA and 3.55 FIP. And yet, they’ve been worse than they were in the regular season, when they had a 2.75 ERA and 2.40 FIP.

In fact, no matter how you slice it, pretty much every grouping of pitchers has done worse in the playoffs. That makes sense — the batters who make the postseason are far better than the ones pitchers face on average over the course of the regular season. Pitchers with a FIP below 3.00 in the regular season had a combined average FIP of 2.64 and an ERA of 2.81. In the playoffs, they’ve put up a FIP of 3.33 and an ERA of 3.56.

You can go down the line as long as you’d like, but the results don’t really change. Great pitchers? They’re still great in the playoffs, just less so. Bad pitchers? On average, they’re bad in the playoffs, even worse than they were in the regular season. But neither side appears to have some unique claim on the special sauce that works in playoff baseball. Everyone gets worse, more or less.

In fact, you could even argue that the reverse is true. Use ERA, and things get very interesting. Twenty-three pitchers with regular-season ERAs above 4.00 have pitched in the playoffs. In the regular season, they were pretty bad, combining for a 4.37 FIP and 4.90 ERA. In the playoffs, they’ve actually been better: in aggregate, they’ve produced a 4.82 ERA and 4.35 FIP. The crucible of the postseason takes bad pitchers and spits out those exact same pitchers, with similar statistics intact, apparently.

Meanwhile, the very best pitchers by ERA have struggled on a relative basis. For example: thirty-six pitchers with regular-season ERA’s below 2.75 have pitched in the playoffs, throwing 181.1 innings, roughly 30% of all innings in the playoffs. In the regular season, these guys were nails: they compiled a 2.22 ERA and 3.03 FIP. Maybe they were getting lucky, but they were also great.

In the playoffs, they’ve run into the same effect that seems to ding most everybody. Those dominant pitchers have a combined 3.43 ERA, 1.21 points higher than their regular-season mark. They have a combined 3.51 FIP, 0.48 points higher than their regular-season mark. They aren’t special playoff performers immune to the pressure, or pitchers who have good stuff and are thus uniquely equipped against playoff lineups; they’re simply better pitchers on average, facing the same increased difficulty as everyone else.

Try as I might, I couldn’t slice the data to prove what feels so true in watching baseball: that you can’t afford to use bad pitching in the playoffs without seeing an unending stream of balls launched into the sun. Will some of these bad pitchers get shelled? Of course! That happens to them plenty in the regular season, too. If you can avoid it, you probably shouldn’t throw a bad pitcher into a close game. But best as I can tell, there’s no effect beyond that universal truth.

This has some implications for playoff managing, but the managers left standing seem to have internalized those lessons already. If you have to rely on the bad parts of your bullpen to get outs, do it! They’re on the team to get outs, after all. If you can deploy them in a spot where the game isn’t particularly close, that’s great; but if you need them, they can pitch close innings, too, with the increased failure rate you’d predict from looking at their regular-season statistics.

That’s what made me like the Braves’ double-opener gambit so much. Sometimes, you get Wright. Sometimes, you get Davidson or Dylan Lee. Those three pitched to an aggregate 4.21 ERA and 5.91 RA/9; earned runs can be noisy in such small samples. Sometimes, that’s enough to win. Sometimes, it’s poor enough to get you rocked. In no case, though, are the playoffs some secret subset of baseball where regular rules of ERA don’t apply and aces become immortal while bad pitchers crumble to dust.

I had an agenda with this article. I wanted to show basically what I did, which I’m sure means I did the study in a biased way. The pitchers who make postseason rosters despite high FIPs are probably better than FIP thinks they are, and the same is certainly true for ERA. Likewise, pitchers who did extremely well in the regular season are surely due for some regression. I make no claims that this data is accurate and infallible. Still, I think the idea bears repeating. Good pitchers remain good in the playoffs. Bad pitchers remain bad. There’s just no particular reason that the effect kicks into overdrive when the stakes are highest. And hey, thanks for reading. Here’s a table for your trouble: