What is a stolen strike worth now?

The big picture

For about fifteen years, the quietest way to win baseball games was to catch the ball well. A catcher who could make a pitch an inch off the plate look like a strike was stealing runs every night, and nobody in the stands could see it. In 2025 Patrick Bailey had the best framing season of any catcher since at least 2018: +25 framing runs, which is about two and a half wins from receiving alone.

Then 2026 happened. MLB approved the automated ball-strike challenge system in September 2025, and this season every team got two challenges a game. The batter, pitcher or catcher taps their head, Hawk-Eye checks the pitch, and a correct challenge is kept. A strike stolen from the umpire can now be stolen back.

I already wrote about how teams spent those challenges. This post asks the other half of the question: what did ABS do to the value of framing itself? The 2026 regular season is over, so we have a full year of real data to look at.

The leaderboard got squeezed

Statcast still measures framing the old way. Per Baseball Savant’s ABS documentation, framing value comes from the umpire’s original call, exactly like before ABS, and anything that changes after a challenge is booked separately as challenge value. So the framing numbers below have not been touched by challenges at all. And they still collapsed.

Every season from 2018 to 2025 looks about the same. The best framer lands somewhere between +13 and +25 runs, the worst around −9 to −17, and the standard deviation sits between 5.4 and 7.4. In 2026 the whole thing squeezed together. Only one qualified catcher was more than 10 runs from average in either direction (Logan O’Hoppe at −10.7). In 2025 there were five.

Framing also stopped carrying over as well from one year to the next. For catchers who qualified in back-to-back seasons, a run of framing in one year used to be worth about 0.50 to 0.73 runs the next year. From 2025 to 2026 it was worth 0.36.

Key takeaway: even before a single challenge is counted, framing value shrank by about 30% in the first year of ABS.

Why would the original calls change, when umpires are still the ones calling them? I can’t prove it with this data, but my guess is that umpires know every close call can now be checked in front of 40,000 people, so the borderline pitches that framers used to win are getting called more by the book. Either way, the edges that good framers lived on got thinner.

What happened to the best framers

Here are the top framers of 2025 and what they did in 2026. The last column is Savant’s challenge runs above average for each catcher, the new skill ABS created.

Bailey went from +25.1 to +4.6. On a per-pitch basis that is a drop of almost 80%. Some of that is just regression, since nobody stays at +25 forever, but nearly every top framer fell, and the bottom of the list rose to meet them. Edgar Quero went from −12.9 to −3.9, Keibert Ruiz from −8.7 to +2.8, and J.T. Realmuto from −8.4 to basically zero.

The challenge column tells the other story. Among the 86 catchers I could pull from Savant’s challenge leaderboard, the best was Jonah Heim at +4.3 runs and the worst was Realmuto at −3.9, even though he won 81% of his challenges, the best rate among catchers with 75 or more. He only went after about 30% of the runs available to him. That is the “protecting your success rate” habit from The Challenge Problem, showing up in a single player.

So the new catching skill has a range of about ±4 runs. The old one used to range about ±20.

Then the challenges take their cut

The leaderboard above is only the original calls. What actually counts on the field is the call that stands after any challenge. Savant’s league-wide ABS numbers for the 2026 regular season show how much of the original calls got undone:

The first row is exactly what a good framer produces: strikes on pitches that were really balls. Batters took back 28% of their run value. The second row is what a bad framer gives away: balls on pitches that were really strikes. Fielders, led by catchers, won back 40% of those. ABS shaves both tails at once. The good framer keeps less of what he steals, and the bad framer gets more of his mistakes refunded.

One caveat makes this a ceiling rather than an exact number. Framed strikes tend to be the subtle ones, a fraction of an inch off the edge, and those are exactly the pitches batters see worst. In my ABS study, batters read the zone with about 2.5 inches of noise. So the clawback on framed pitches is probably lower than the league-wide 28%.

Putting a number on it

Combining the two effects, here is what a good framer, one standard deviation above average, was worth before and after ABS.

The first drop, about 30%, is measured. The second is an upper bound for the reasons above. So the honest answer is that somewhere between 30% and 55% of framing’s value disappeared in year one. Using the usual rule of about 10 runs per win, an elite framer used to be worth roughly two wins a season from receiving alone. In 2026, the very best one was worth less than one.

Where it goes from here

The clawback is going to grow. Teams challenged more as the season went on, from 3.76 challenges a game in March to 4.61 in September, according to the AP’s season wrap-up. And as The Challenge Problem showed, players are still leaving value on the table, especially in high-stakes spots where they should be challenging far more often. Every improvement there comes straight out of framing’s pocket.

But framing is not dead. Two challenges a game is a hard cap, and the large majority of the umpire’s misses still stand. A catcher who steals an extra strike in the 2nd inning on a 0-0 count is still getting away with it most of the time. What’s gone is the idea that receiving alone can make a catcher a star. In 2026 the catchers who added the most did it two ways: framing well enough, and knowing when to tap their helmet.

Is framing the first skill in baseball to be priced out by a camera?