In a Pinch, “Running” Is a Relative Term

Welcome to another episode of Baseball Writer Notices Something Totally Mundane, Starts Thinking Way Too Hard About It, and Turns it Into an Article, brought to you by FanGraphs, WGBH Boston, and Viewers Like You. I’m your host, Bob Vila.

A couple weeks ago, I was at a Phillies-Astros game where Don Mattingly chose to pinch-run for Luis Arraez with Garrett Stubbs. It’s not every day you see a catcher pinch-run for a second baseman, though it made sense for reasons I’ll get into in a second. A week later, I was at home watching a Guardians-White Sox game, and Cleveland manager Stephen Vogt inserted David Fry as a pinch-runner for Patrick Bailey in a high-leverage situation.

Fry came up as a catcher, and while he’s made a handful of appearances behind the plate this year, these days he’s more of a four-corners bat-first guy. Even so, he still runs like a catcher. Still, Vogt — who’s a good manager with exceptional analytical integration with his front office — chose to put Fry’s leaden getaway sticks on the basepaths with the division title on the line.

Substitutions like this happen all the time; I’m just picking out two examples that got the rusty gears in my brain rolling. Pinch-runners are supposed to be fast, though. And catchers, generally, are not.

There are exceptions, of course. Stubbs is one of them, which is to be expected. There are very few active big leaguers I’d describe as literally unique — Aaron Judge, Shohei Ohtani, Tyler Rogers — and Stubbs is one of them. He’s one of the smallest (5-foot-9, 170 pounds) position players in the league, and he’s not average at anything. He’s one of the best bunters in baseball and one of the best catchers at controlling the running game. But he’s also one of the worst hitters of the 2020s.

The Phillies, for whatever reason, have been categorically unable to fill out a bench this season, so Stubbs has remained on the big league roster as Aaron Nola’s personal catcher and a sort of warm body of last resort at every other position. After he pinch-ran for Arraez, Stubbs stayed in the game to play first base, for example.

Stubbs, who’s 33 now, has lost a step since his prime, but he put up sprint speed numbers in the 70th percentile range for most of his career — he can scoot, especially for a catcher. Arraez, by contrast, is a 33rd-percentile runner, though his baserunning value numbers have been about average this year.

That said, Stubbs pinch-ran at first base with the Phillies down two runs with two outs in the eighth inning. He had one job in that situation: to score from first on a double, should one present itself. (It didn’t, but that’s not Stubbs’ fault.) He’s absolutely better-equipped for that task than Arraez, placing the 2026 Phillies in what is surely a rare historical class: teams whose backup catcher is definitively faster than their starting second baseman.

The Fry situation a week later was about the same. He’s only a 24th-percentile runner, but Bailey is an eighth-percentile runner. And when he entered the game, he went to second base with two outs in the ninth inning and the Guardians trailing by one. With Daniel Schneemann at third base, Fry represented the winning run. And in that situation, every marginal microsecond counts.

At least, it could have. As it turned out, there was no play at the plate:

Again, Fry was Vogt’s best option. Schneemann was himself a pinch-runner, and after accounting for day-to-day injuries, the only other set of legs on the Cleveland bench belonged to Austin Hedges. And Hedges is even slower than Bailey.

I just want to know how we got here, man. How are two contenders pinch-running with catchers late in pivotal games?

I mean, I know how. Left to their own devices, modern GMs would build the entire roster out of one-inning relief pitchers. Major League Baseball had to change the rules to limit teams to 13 pitchers, or else we might’ve seen Chase Shugart or Slade Cecconi on the bases in those games. Perish the thought.

Even with September rosters allowing for 14 position players instead of the usual 13, major league teams are running short benches. Which stinks. The 2015 Royals weren’t too far in the past, and they not only had a starting lineup full of fast guys, they had multiple designated pinch-runners. Jarrod Dyson could at least plausibly hit and play the outfield, but in case he wasn’t fast enough, they kept Terrance Gore in reserve for special occasions.

Time was, a fast guy could make a career out of being a theoretical outfielder without ever touching a bat. Sometimes those guys would use the Herb Washington role as a springboard into everyday playing time — shout out Rajai Davis and Michael Bourn — but even Scarborough Green had his place back in the day.

Now we’ve got catchers pinch-running. Routinely. Yikes.

My expectation was that we’d be seeing fewer pinch-runners now, with shorter benches in general and the demise of Ned Yost’s Flying Circus in particular. So I compared this year’s pinch-running numbers to two seasons for context: First, 1998, the season in which Major League Baseball expanded to 30 teams. Second, 2015, which is the first season of the Statcast era, in addition to being the pinnacle of the Nine Innings of Hell Royals:

I was surprised to learn that pinch-runner usage has changed almost not at all. Or something like that. Because there has been fluctuation in pinch-runner usage over the 30-team era; it just wasn’t picked up in that three-season snapshot. Here’s a season-by-season line graph of pinch-running appearances since 1998, with 2020 projected out to a 162-game season:

If you forget 2020 (and boy, does everyone want to), we had a consistent downward trend in pinch-runner usage through the late 2010s, heading into the early 2020s. Then, in 2022, we kicked right back up to the normal range from the first part of the 2000s: between 850 and 1,000 pinch-runners per season.

That uptick coincides with the advent of the universal DH, but it does not, as I had expected previously, coincide with the move to 26-man active rosters in 2021. The year rosters expanded, we had the lowest pinch-runner usage of the 30-team era.

In the interest of following the scientific method, here’s my hypothesis: In the 2010s, pinch-running declined because, as relief pitchers took over more and more of the roster, there were fewer fast guys on the bench to pinch-run with.

In the 2020s, the reserve army of fast guys has remained thin, but pinch-running has returned to previous levels as managers have started to pursue more marginal upgrades. Imagine, like, Bobby Cox in 2006. He’s not going to bother pinch-running for a slow catcher with a slightly less slow catcher. A manager like Vogt, armed with more detailed data on his players, is willing to spend a substitution to chase an extra couple tenths of a foot per second with the game on the line.

Testing this hypothesis is tricky. Collecting outcomes from pinch-running appearances is kind of beside the point. Take the Stubbs-Arraez situation from earlier. Stubbs entered the game on first base with two outs in the bottom of the eighth. And that’s where he stayed, when Alec Bohm struck out to end the inning. That’s probably the most likely outcome for an at-bat that pits Bohm against Bryan Abreu, but that doesn’t make it any less the correct decision to pinch-run Stubbs for Arraez.

Then there are measurement concerns. Fast players don’t always show up on speed leaderboards because Baseball Savant’s criteria for measuring sprint speed are quite narrow. In 2016, for instance, neither Gore nor Dalton Pompey came up with a dedicated sprint speed, because they didn’t play enough.

Nevertheless, I wanted see if specialist pinch-runners were getting slower.

Before I show you the chart, let’s define some terms. For the purposes of this exercise, pinch-running specialists are players with five or more appearances as a pinch-runner in a given season. Mean, median, maximum, and minimum are all terms you’re familiar with; these values are for average sprint speed among the population of pinch-running specialists, not in the league at large. I also noted the sprint speed one standard deviation above and below the pinch-running specialist mean.

Finally, the column in the right: This is the number of players each season with five or more pinch-running appearances and an average sprint speed at or below 27 feet per second. That’s what Baseball Savant reports as the league-average sprint speed:

As you can see, high-usage pinch-runners are a couple ticks slower on average than they have been throughout the 2020s. We’re also seeing a greater spread in speed among such players; one standard deviation either side of the mean is a spread of 2.2 fps this year, the widest gap in a 162-game season since Statcast came online in 2015. In some seasons, that number has been as low as 1.6 fps.

We’ve also got eight average-or-worse runners with five or more pinch-running appearances this year. That’s as many as the previous three seasons put together. Here are those eight slowpokes:

You can see the logic with some of these guys. Hayes, Allen, and Palacios have all put up above-average sprint speed numbers in the recent past, and have just slowed down of late. Perhaps more to the point: Those three, Espinal, and Walton are all short infielders. Apart from Hayes, who’s 5-foot-11 and plays third base like a shortstop, they’re all middle infielders who are listed at 5-foot-9 or under.

I just assume any short middle infielder can run. I’m sure there are a few major league managers and coaches who’d make the same inference. Like, I found out Walton wasn’t a plus runner when I made this chart.

That leaves three pinch-running slowpokes: Schneider, Piper, and Valenzuela. Schneider’s basically an average runner, and the Blue Jays’ starting lineup is both platoon-heavy and heavy of foot. At any rate, Schneider has pinch-run once since the first week of May; it seems like Toronto has some better options.

The other two are catchers. All six of Piper’s pinch-running appearances have come at the expense of Liam Hicks, who’s a marginally slower runner. I don’t know if Piper, who’s made all of two big league starts in his career, is actually a defensive upgrade on Hicks, but I get where Kevin Cash is coming from. If you’re going to make a defensive substitution anyway, you might as well get a jump on it by pinch-running the inning before. And Hicks sure is on base a lot.

But Piper pinch-running for Hicks is equivalent to running for Bailey with Fry. Valenzuela is not just slow, he’s got 11th-percentile sprint speed, which is marginally slower than Hicks and more than a foot per second slower than Arraez.

Valenzuela is the slowest semi-regular pinch-runner in the Statcast era, by a margin of…well, let’s line Valenzuela up in the 100 meters against any of the other players I’ve mentioned so far and see what the gap is. That kind of margin.

So why is Toronto manager John Schneider using Valenzuela as a pinch-runner? Well, in five of his six pinch-running appearances, he’s replaced the same player. I’d like you to take a few seconds to guess who it is, if you haven’t figured it out already:

Valenzuela couldn’t outrun the parliamentary process, but he looks like Donovan Bailey next to Alejandro Kirk.

That’s about the state of pinch running in modern baseball. It’s a bit of a counterintuitive breakthrough, but managers and front offices all seem to be coming to the same conclusion: Don’t let the glacial be the enemy of the slow.