How Often Should Baldwin Catch In The Playoffs?

Plenty of people are discussing this question. I decided, shortly before Claude kills me and everyone else, to have a discussion with him about it. Before I let y’all in on it, I want to simplify the question a bit.

First, let’s simplify the choices. It’s either Baldwin at catcher, with Smith, Dubón or Yastrzemski at DH (and Ha-Seong Kim at short if it’s Dubón) or Baldwin at DH and Murphy at catcher. I think those are the only reasonable possibilities. In particular, I don’t think Murphy at DH makes any sense at all. (Obviously, there could be other situations, like DHing Acuña for some reason, but in all of those situations Baldwin will catch.)

I observed a couple of days ago that Baldwin has hit considerably better as a DH than as a catcher. This is at least mildly unusual. Most players hit better as players than as DH. In The Book, Tango, Lichtman and Dolphin report about a 10-15 point wOBA increase when a guy plays in the field. That data is old, though, and I’m not going to try to update it here, nor am I going to try to separate catcher-DHs from other positions.

What I did do, though, is look at every player since the universal DH who had both 100 PA as catcher and 100 PA as DH in a given season. (I’m not looking at 2026 because I didn’t anticipate needing this split when I started downloading the current season data.) Here’s the data set:

PlayerSeasonPA as DHPA as COPS DHOPS CDIFF
Yainer Diaz20231491910.5658350.978883-0.413048
William Contreras20221332400.6196590.994867-0.375208
Víctor Caratini20251182000.5357910.884444-0.348654
Mitch Garver20232871080.7659901.099610-0.333620
Yainer Diaz20251064550.4883740.754369-0.265995
Gary Sánchez20241501110.5557760.800164-0.244388
Sean Murphy20221294810.5838180.805464-0.221646
Cal Raleigh20251705890.7879551.004244-0.216289
Willson Contreras20241412160.7217630.937177-0.215414
Willson Contreras20231273630.6766450.877602-0.200956
Hunter Goodman20251604130.7000000.892366-0.192366
Edgar Quero20251162710.5694400.758318-0.188878
Ryan Jeffers20241053390.5809520.764788-0.183836
Salvador Pérez20251603670.6442640.774018-0.129755
Willson Contreras20221683190.7486820.852359-0.103678
Yasmani Grandal20221022560.4895200.564618-0.075098
William Contreras20231324820.7539440.823916-0.069972
Alejandro Kirk20222173190.7561480.817304-0.061156
Yainer Diaz20241514250.7194420.768875-0.049433
Ben Rice20252131150.8696650.884668-0.015003
William Contreras20241495410.8226770.8198370.002841
Salvador Pérez20231243650.7456450.7413240.004320
Salvador Pérez20221673060.7647510.7526550.012096
Tyler Stephenson20231673180.7149860.6396620.075324
Mitch Garver20251221710.7114750.6227490.088726
Adley Rutschman20232024940.8790950.7712650.107830
Gary Sánchez20221393210.7427760.6324960.110280
Agustín Ramírez20252723110.7802840.6272600.153023
Adley Rutschman20241854550.8188070.6439940.174813
Ryan Jeffers20251223380.8864720.7089000.177572
Will Smith20221094791.1332740.7210840.412191

In line with Tango et al., the average catcher hits worse as a DH. I find this mildly surprising simply because catching is so hard. But about a quarter of catchers hit better as a DH, with Will Smith’s 2022 as huge outlier.

Here is Baldwin this year from B-Ref:

SplitPAABOPStOPS+sOPS+
as C342295.78794130
as DH264245.846107129
as PH11.000-100-100
as PH for DH11.000-100-100

Provided by Baseball-Reference.com: View Original Table
Generated 9/25/2026.

For those who don’t speak BRef, tOPS+ is Baldwin compared to himself by position, sOPS+ is Baldwin at that position compared to everyone else in the league at that position. So Baldwin’s hitting advantage as a DH puts him in the upper quarter of this distribution, about where Tyler Stephenson’s 2023 is.

At this point, in keeping with my decision to go on cruise control, here’s Claude:

Here’s how I’d set up the decision.

The comparison isn’t Baldwin’s DH split. Having Baldwin catch every day doesn’t just move his bat; it changes who fills the DH slot. The real choice is Baldwin catching plus some other hitter at DH, versus Murphy catching plus Baldwin at DH. Offensively, that becomes Murphy’s bat against whoever would DH, and on current form that probably favors benching Murphy. Baldwin’s own role split is a second-order question on top of that.

On whether Baldwin’s differential is real: I’d regress it hard. The +0.059 comes entirely from BABIP, his OBP is identical in both roles, and the prior from your sample points the other way (?0.083 on average). A rough shrinkage makes this concrete. His own difference has a sampling error of something like 0.10–0.12 OPS on about 300 PA per side. Most of the 0.19 spread in your sample is also noise, so the true between-player variation is small. Shrink his +0.059 toward the sample mean with weights like those and you land somewhere around zero to slightly negative. In other words, there’s no real evidence that catching costs him at the plate, which removes the batting argument against catching him every day.

Workload is easier in the playoffs. The usual reason nobody catches every day is accumulated fatigue over 162 games. The postseason has scheduled off days between series and often within them, so catching every game in October is a much smaller ask than catching every day in August. That weakens the strongest case for rotating.

That leaves Murphy’s defense. Framing, the running game, and game-calling with a pitching staff he knows well are the only real reasons to keep him in the lineup. How much weight to give them depends on how his receiving has held up since the injury, and whether the staff’s comfort with him is worth more than the bat you’d gain at DH. That’s a judgment call for Weiss, and it’s the part of the argument your post should probably focus on, since the batting-split evidence doesn’t settle much either way.

My response:

I agree with all of that. As to who would DH instead of Baldwin, it’s probably some combination of Dom Smith, Mike Yastrzemski and possibly Mauricio Dubón if Ha-Seong Kim plays SS.

Claude came back:

I pulled their 2026 lines from the Braves team page on Baseball-Reference:

PABA/OBP/SLGOPS+
Sean Murphy114.165/.246/.29148
Dominic Smith294.254/.311/.37589
Mike Yastrzemski397.231/.315/.39094
Mauricio Dubón612.276/.321/.39597
Ha-Seong Kim139.113/.203/.121?7

The offensive side of the decision is lopsided. Any of the three DH candidates is a roughly league-average-minus hitter, while Murphy has been about half of league average since his return. That gap of 40-plus OPS+ points in one lineup slot works out to something like a tenth to a fifth of a run per game. That’s a big enough number that Murphy’s defense has to be quite good to cover it. Even the best framing catchers are usually valued at a fraction of that per game.

Two caveats. First, 114 PA is a small sample, and a regressed estimate of Murphy’s bat would sit well above a 48 OPS+. But he’s coming back from injury, so it’s not clear his pre-injury level is the right thing to regress toward. Second, the Kim scenario deserves a line in the post. If Kim plays shortstop so that Dubón can DH, you’re swapping one weak bat for a much weaker one. His ?7 OPS+ makes Murphy look healthy. Whatever his glove adds at short, that configuration gives back most of what you’d gain by benching Murphy.

Me: Exactly. Although Kim has looked better lately. but that’s what you’d be betting on.

Claude:

Right, and it’s the same bet you’d be declining with Murphy. Both are small recent samples pointing different ways: Kim’s recent upturn against a dreadful season, and Murphy’s post-return slump against a solid track record. Treated consistently, you’d regress both of them toward their longer-run levels. That makes Kim a bit less hopeless and Murphy a bit less dire than their 2026 lines suggest. It probably doesn’t change the ranking, but it’s a fair point for the post if anyone argues that Kim’s hot week is evidence while Murphy’s cold month is noise.

Let me offer a slightly more human take on this. I think Claude is right as far as he (?) goes. But if you’re going to need Murphy at any point in the playoffs, you have to play him or he’ll be useless at the plate. If you’re not going to have Murphy catch most of these games, there’s no point in having him on the roster. If he’s only there to be a fallback for a Baldwin injury, then you ought to have kept paying Chadwick Tromp the league minimum. I don’t think that’s dispositive, but I think it’s a good example of how the numbers can’t do everything.

The Game

The Braves went with an opener, Ray Kerr, even though they also used starter Grant Holmes. The argument for this, apparently, is that if there’s a third game against the Phillies, opening with a lefty against Schwarber and Harper in the first is a good matchup. Fine. I guess there could be something psychological about the first inning — whatever. If that was the theory, Kerr gave up hits to Schwarber and Turner to start the first and then got out of the 1st, followed by a perfect second and third.

The Marlins pitched Eury Pérez, who didn’t give up a hit until the 6th when Sean Murphy led off with a single, but he was erased on a Baldwin double play. The next two hits, the first off Holmes, came in the bottom of the inning; they were stranded on another double play.

Somebody was going to have to do something eventually — that’s the way baseball works. Jimmy Mack’s brother Joe hit one 428 feet to give the Marlins the lead in the 7th. Fuentes gave up a run on a triple and single in the 8th. After two baserunning blunders eliminated two Marlins, a Griffin Conine homer made it 3-0. It’s hard to give up a triple, homer and three singles in an inning and only give up two runs. Turns out it’s the 6th time it’s happened.

Pérez went the rest of the game without giving up a hit, falling just short of a Maddux at 105 pitches.

I guess if Murphy is going to get all your hits, you have to play him.