JotCast | Ben Clemens FanGraphs Chat

Ben Clemens FanGraphs Chat - 8/25/25

View Full Chat Next Page

Ben Clemens

2:01

Hey everybody, let's get the chat started. I'm aiming for a longer, more relaxed vibe today. I've got the US Open on in the background, no games until the afternoon, doing some random research that I'm happy to take a break from, let's talk baseball

Mike

2:01

Probably a user-error, but how do I access Ohtani’s pitcher player page? I can’t find any link to it from his main batter page

Ben Clemens

2:02

So, there should be a batting/pitching toggle at the top left, just below the search bar, on desktop

If not, um, ping me again?

Todd

2:02

This is pedantic, but: BWJr went 1-4 with a walk yesterday, giving him an OBP of .400 and a slug of .250. His game log shows a wRC+ of just 33. I can’t see how that would be accurate?

Ben Clemens

2:02

No, we live for pedantry here~

2:03

The reason is that Witt was intentionally walked. wOBA and wRC+ treat IBB's as non-PA's, so his wRC+ is calculated only on the four PA that didn't end in IBB's

Lars

2:03

It’s the first inning. Runner on first, your second batter - and best hitter on the team - is up. You’re facing a starter (not an opener). If you could guarantee X number of pitches for the at bat in exchange for a strikeout, at what point would it be worth it? Example: a 25 pitch strikeout is a positive outcome (I think) because it adds so much to the pitch count.

Ben Clemens

2:04

This is a fun question, and honestly I don't know the answer to it at all. I'd guess that it's a high number, yeah, probably at least higher than 15. The tradeoff is not amazing but honestly, the most likely outcome is that your guy makes an out anyway, so why not get some ptiches guaranteed?

Guest

2:04

The Medvedev outburst last night was insane. No idea how he didn’t get a point penalty

Ben Clemens

2:05

We were channel surfing and happened to see that, and we were like "okay we have to watch this match until the end now"

if you haven't seen this, everyone, go check it out. Medvedev got in an argument iwth the umpire (reasonable argument imo) and kind of egged the crowd on to boo the umpire for his decision. but the booing got out of hand and actually paused the match for 7 minutes because it was so loud that Medvedev's opponent wouldn't serve

Oaktown Blues

2:06

Re: The IBB not counting as a PA in wRC+ or wOBA: They still get WAR credit for the PA, right? Just counted as an additional PA at their normal wRC+, not as a walk?

Ben Clemens

2:06

That's correct. We basically go back and credit the PA to their WAR at their seasonal wRC+

Insert Witty Name Here

2:06

Who ya got for the US Open? Going chalk to Sinner/Sabalenka. What are your thoughts on how the finals in tennis are usually the same 3-4 people?

Ben Clemens

2:07

I'm on Mirra Andreeva on the women's side, but I can't tell you whether that's because it's a measured opinion or b/c I went to Indian Wells this year and she was on fire then

I'm okay with it. I actually watch more tennis, like sheer hours watched, early in the slams, because it's so fun to watch all these marginal players trying to get to the top of the heap

but at the end of the day, yeah, gimme a Sinner/Alcaraz slugfest as a final over Taylor Fritz getting absolutely steamrolled every day of the week

El Tim

2:07

Hey Ben, I stumbled back upon a 5 Things from May where you wrote about the Tyler Soderstrom left field experiment. Very prescient to predict that he'd have more OAA than Steven Kwan at this point in the season. Okay, you didn't actually do that, but still... it seems like you were right to like it and that he's holding his own out there.

Ben Clemens

2:08

Ha, I'll happily take credit for that even if I didn't predict him being a positive out there

Necco Wafer

2:08

Skubal is running an 80+% strand rate for two years now. Is it possible he's tougher to hit or steal on due to the lefty factor?

Ben Clemens

2:08

Yeah, also higher-K pitchers run higher strand rates

The Terminator

2:08

How long before you are replaced by AI?

What aspect of your job will take the longest for AI to learn?

Do you know where I can find Sarah Connor? She's a friend of mine.

Ben Clemens

2:09

I'm hoping not, and honestly, I don't think it's all that likely

the idea generation and weeding out garbage from signal part is still pretty tough for AI to do

2:10

I know because I am constantly trying to use GenAI to come up with article ideas, help me parse data in more ways

the hardest aspect isn't the actual analysis, it's determining which analysis is interesting

Crashed Scion

2:10

The Brewers were projected for 81 wins preseason by Playoff Odds. They have 81 wins today. They are already an MLB best +71 wins versus the FGPO preseason win totals for the eight completed full seasons in the database, this will (presumably) be their fifth season with a +12 or greater margin. Any thought for recalibration in the offseason? Even manually adjusting their ERA-FIP to like -0.25 or -0.33 versus the -0.10 it is at now would buy some extra wins. Last three years they've run a -0.48, since 2016 its -0.26.

Ben Clemens

2:10

Yeah, no way

We're not gonna start recalibrating specific adjustments for teams, that seems like a very bad way of doing this

2:11

If we're talking a bigger overhaul of the model, that's a definite maybe

or at least, a goal

one thing we really like about it is how lightweight it is, though

it runs a LOT, like all day every day

Mariners Fan

2:11

The Mariners are 20-7 while I’m traveling internationally this year and 49-54 while I’m state side. How do I break it to the guys that I don’t have any more trips planned and they’re going to fade down the stretch?

Ben Clemens

2:11

I mean I think the obvious conclusion is that you need to schedule more trips

2:12

you call yourself a fan? find a way to get sent to Paris next week

Jeremy

2:12

We tend to draw a line between true talent and luck/noise. Hard to figure out where to draw the line, especially with small samples, but we assume there's a line to be drawn even if we don't know exactly where to draw it. But in a world where players are constantly adjusting (especially in response to adjustments from other players), maybe it would be better to see "true talent" and "noise" as two ends of a continuum that vary on different timescales? Like, maybe "true talent" is something that only changes with age, if it changes at all? But maybe there's also "current adjustment-dependent talent" that changes on a timescale of weeks to months? And then there's "luck/noise" that varies from pitch to pitch or day to day?

Ben Clemens

2:12

Oh man, I've been thinking about stuff like this in general of late and yeah, I tend to agree

there's like long-term baseline talent

Connecting…