Errors Are Hits Now

Is the "E" becoming passé?

ERIC ASKS!

I have a hunch that in recent years official scorers have become cowards and are inclined to only give errors on plays that add an additional base, rather than cost the batter a hit. Does the data support my neuroses?

SM: In 1961, the first year after expansion, a batter (on either team) reached base on an error a little more than once per game. More precisely: There were 1.15 baserunners reaching base on errors per game. In 2024, the first year after The Beekeeper, batters reached by error less than 40 percent as often: 0.44 batters reaching base per game. The reached-on-error is disappearing.

I’ll argue two things here, one quickly and one in more detail. The first is that it’s the scorers, not better defense. The second is that a lot of this was probably caused by the pandemic?

First things first: It’s Not Better Defense.
I’m not saying defense hasn’t gotten better. Athletes get better. Compared to the 1960s, kickers today make more kicks, high jumpers today make higher jumps, heavyweight boxers today make heavier boxes. Fielders today probably field better. But they’re not less error prone.

We can tell by splitting up the two types of errors: Reached-on-error (ROE) errors and runner-advanced errors. The official scorer doesn’t have much leeway ruling on runner-advanced errors. Something must account for the runner’s advance, and if the only preceding event was a throw rolling down the right-field line, well, that’s an error. Hardly any way around it.¹

But reached-on-error errors are all technically subjective. Every single one has an alternate explanation: That it was actually a hit.

This was ruled a hit.

We can see the rates of those two types of errors per game, since 2000:

(This is per team per game — so, half the per-game rate referred to at the start of this article.)

The less subjective style of error has been pretty steady², suggesting that defenses, in fact, make mistakes about as often as they ever did. But the highly subjective errors—ones that could be recorded as hits, if somebody chooses—have been declining, especially in the past five years. Errors are being reclassified as hits. The league’s batting average last year was about 5 points higher than it would have been under the 2000 scorers’ standards.

Second Thing Second: It Was A Very Large Pandemic.
You’ll see in that chart that the latest drop in reached-on-errors is steep and started in 2020. Reached-on-error plays dropped about 10 percent in 2020, and then a further 10 percent in 2021, and they’ve been pretty stable down there since.

I don’t think it’s a stretch to say that this is a pandemic effect, in two ways. The first is: During the summer of 2020, while you were probably working at home, official scorers were also working from home. Instead of sitting in the press box, they were watching live video feed, with tons of extra angles, the ability to slow things down—the same infrastructure used for replay reviews. Scorers also began using Slack, which allowed them to discuss some decisions as a group. They’ve kept access to those tools since.

Terry Francona—the Guardians’ manager at the time—worried that the extra replay tools would lead to more plays being ruled errors: “I know any time you slow it down and watch it again, it always looks like an error,” he said. But it’s probably the opposite case: A ball that’s hit right at a fielder might seem easy until you see it in slow motion, until you see the topspin. For instance, after this play,

the Padres broadcast had this exchange:

Mark Grant: With the way the Padres were positioned—just couldn’t squeeze it. Right in front of him, right in the bread basket, a good ol’ fashioned boot right there. That’s a play that’s gotta be gloved.

Don Orsillo: They award him a hit on that play.

Grant: Really?

Orsillo: I’m a little surprised.

Grant: Yeah, me too.

But the slower the replay, the more complicated the play looked. The ball hit off the mound and picked up a different spin. In one shot, you could actually see the chaotic spin as it hits off Ha-Seong Kim’s thumb:

What looked on first viewing like a clunk looked, upon closer inspection, like an odd play. As the Nationals’ broadcaster Kevin Frandsen put it, “When it goes off the mound it’s ejecto time. One-oh-five off the bat. Nasty spin right there.”

Frandsen refers to another modern tool that a scorer now has: The exit velocity. It can be hard to tell the difference between 100 mph and 105 mph on a groundball, or even 100 and 115, depending on how steep the ball is chopped into the ground. But Statcast a) tells us and b) puts an expected batting average on every play based on exit velocity and launch angle. Based on how hard it was hit and the angle at which it was hit, the ball right above had an expected batting average of .497, without even accounting for the mound’s ejecto effect.

So this was ruled a hit, and I doubt it would have been in the past,

This was ruled a hit.

and I’m fairly certain that Statcast telling us that it was hit 113 mph was a factor in that decision. A ball hit that hard, at that particular angle, is according to Statcast a hit more than 60 percent of the time (though not when it’s hit directly at a fielder).

Put another way — after this groundball

This was ruled an error, and then changed to a hit.

was initially ruled an error, the Cardinals broadcaster Jim Edmonds said, “Tough error. I don’t think it’s a tough error on the position player. It’s a tough error on the hitter.” In other words: The shortstop should have made the play (deserved an error). But the batter can’t do much better than hit a ball 110 mph, so he deserved a hit. The more information a scorer has, the more reasons he can opt to be generous. (That play was changed to a hit.)

But my pandemic theory isn’t even primarily about camera angles and Statcast data. Rather, I think the pandemic (and, perhaps, the larger era around it) changed a lot of our expectations for institutional authority. In all sorts of areas of life, you can see institutions softening, becoming less punitive, less judgmental, and perhaps even more skittish about being mean. Music critics are expected to be “poptimists” rather than snobs now. My library quit charging late fees. “Securities fraud enforcement seems a bit passé,” Matt Levine wrote in his Money Stuff newsletter this week. Hall of Fame voters are talking themselves into voting for everybody. These are scattered examples of varying merit and with different mechanisms at work—erasing library fines is a pursuit of social equity, while securities fraud unenforcement is regulatory capture—but they all point to a modern idea about authority, which is that scolds are out of fashion.

And, so: The official scorer has the tools to be generous, and the social pressure to be generous, and when given the discretion is thus exercising more generosity.

**

For a play to be an error today, rather than a hit, it must look exactly like an error: Ball hit to the fielder, clean hop, clanks off the glove; or cleanly fielded ball gets thrown directly away from the first baseman. If there’s really anything slightly off about the play, it’s a hit.

I mean it, anything. Here’s a play (all of the following are from the 2024 season):

If the ball has chunked right off Ke’Bryan Hayes’ glove, it would have been an error. But because he fielding it and then simply slipped, it shifted into “weird play,” and thus no error. This was broadcast booth discussion:

Tom McCarthy: Fourth error this year for Hayes.

John Kruk: Well, that’s a tough error.

McCarthy: I’m assuming it’s an error. Maybe it’s not.

Mike Schmidt: John. That’s a tough error?

Kruk: Well, he slipped! He can’t help a slip.

Kruk: They gave it a hit.

Schmidt: WHAT?

Here’s a hit:

Nothing happened on that play other than “Freddie Freeman bobbled a ball rolling 40 mph directly at him with plenty of time to make an easy play.” But it was a bunt; bunts are sorta weird?; call it a hit.

Here’s a hit:

Clanks are mistakes, bad throws are mistakes, but bobbles transferring to throw—can’t help a bobble! No error. Here’s a hit:

It ate him up. You can’t help it if a ball eats you up!

Here’s a hit:

No errors if you’re moving to your left.

Here’s a hit:

No errors if you just fall over.

This is a hit:

No errors if it looks goofy enough.

**

Here’s something I can say with almost 100 percent certainty: More runners will reach base via error in Philadelphia this year than in Minnesota. I can say that because it’s true every year. There have been more ROE in Philly than in Minnesota in each of the past 10 years. Philadelphia is almost always near the top of the league in ROE; Minnesota is nearly always at the bottom. That’s because errors are, obviously, a highly subjective social construct.

I’m not mad at any of those errors up there being called hits. I’m just noting that this is one of those quiet changes that happens without anybody really intending it, for reasons that we can only speculate on, and it’s interesting to see where these things lead. We never had to have errors at all. We could be witnessing the end of them.

Thanks to Lucas Apostoleris and Baseball Prospectus for research assistance!

1. When I covered local news and the police would release yearly crime statistics, the grizzled editor would remind us cubs that only one crime stat is reliable: Murder. Everything else can be manipulated by departments that want to make crime look worse (to get more staffing) or better (to make some initiative look successful). But unlike, say, vehicular burglaries, “dead body” can’t go unreported, ignored, or recategorized.

2. You’ll note the small tilt upward in non-ROE errors around the year 2019. A large part of that is simply that catcher’s interferences—which count as errors against the fielder, but not ROE for the batter—have gone way up.