Dan Szymborski FanGraphs Chat – 6/5/25

Dan Szymborski: Greetings, FanGrapheteers!

Harshil Jani: Hi Dan. I would like to know what the accuracy of the Fangraphs pre-game win probabilities under the ‘live scoreboard’ section. Thanks.

Dan Szymborski: I don’t actually know; there’s a lot of data on our proejction probabilities, but I don’t have access to our historical numbers of game vs. game.

Dan Szymborski: I imagine they’re best when the teams are closely matched, since they’re game state based.

Dan Szymborski: At some point, we’ll have a ZiPS in-game version of this installed (you may have seen the beta in the playoffs), and that will know more about the relative strengths of the team

John: I know you’re the ZiPS guy, so if there is someone better to ask let me know, but how does Devers have such an elite batted ball profile despite pretty middling bat tracking metrics?  Average bat speed and poor squared up %, but elite in avg exit velo, barrels and hard hit %.

Dan Szymborski: Devers using a fairly heavy bat, doesn’t he? I”m not sure how to check that out really quickly though. It’s certainly likely if his contact was more consistent that his hard-hit % would be even crazier

Matt: Hey Dan what does Zips say is a good contract offer for PCA a la Corbin Carroll or Jackson Merrill, buying out arb years and first few of free agency?

Dan Szymborski: ZiPS would suggest $225/10

Snid: Any word on Jobe?   Do you think we are looking at TJS?

Dan Szymborski: I don’t think I’ve heard anything you haven’t heard. Hearing flexor strain isn’t exactly the happiest injury to hear about, and it’s natural that TJ surgery will pop into heads

Alex: What is wrong with Atlanta’s offense? Why are they so bad?

Dan Szymborski: I”m not sure there’s a magic answer beyond the obvious; a bunch of their players are hitting like crap

Taylor Walls Can’t Hear: I’ve been thinking lately about the concept of context independent stats. I understand their utility, but do you think they’re overused in the sabermetric community? I understand FIP is useful, but players do have defenses behind them, so why dock a player who doesn’t get strikeouts but keeps runs off the board when that guy is still super valuable? Sure, that homer may have only been a homer in Yankee Stadium, but he does play in Yankee stadium!

Dan Szymborski: Because the idea is to give credit to the players who create the value.

Dan Szymborski: If you give more credit to a pitcher for what the defense does, you necessarily have to reduce credit to the defense for what the defense does.

Dan Szymborski: Meg and Matt fix my shitty run-on sentences, I’m not sure that I should get more credit for the fixing of them simply because at the end of the day, the run-on sentences are eliminated

Kevin: What’s something you know about baseball now that you didn’t know even a few years ago?

Dan Szymborski: Until I looked a few years ago, I didn’t realize that preseason projections actually predicted next year’s team record better than the team’s record in the intervening year

RJS: Just curious – why do you build a setup to run ZiPS at home instead of running it in the cloud? Seems like you could gain a lot of flexibility in speed and cost.

Dan Szymborski: For one, it creates additional coding challenges that I’m unskilled in. Second, I take care of my own security and if something happens, then I’m responsible.

Dan Szymborski: and frankly, I don’t want additional complexities when I’m tinkering

drplantwrench: i’ve been thinking a lot about Ryan Johnson – he’s got good stuff certainly, but i feel like a lot of his value is from how wacky and weird his delivery is.  How does ZiPS/you approach the value of a pitcher being weird and unexpected?

Dan Szymborski: I personally like being weird and unexpected, but the question does come up “what happens when people are used to this”

Dan Szymborski: The only reason I ever got anyone out as a pitcher as a kid is that I threw sidearm and that was unusual. But once you get well into your teenage years, batters get more comfortable with oddness and crushing your pathetic 70 mph fastball or your crappy little palmball thingy.

Dan Szymborski: ZiPS of course doesn’t really address ‘weirdness’ as a cateogry

Justin: Hi Dan. Just read Ben’s article on Luis Torrens. Has Zips bought into his hot start? Thanks

Dan Szymborski: A bit mroe than the simplified in-season model.

Dan Szymborski: the simpler models has his wRC+ going forward as 99, the more robust model 103

Insert Witty Name Here: Is Cal Ripken, Jr rolling in his grave with how bad the O’s are messing up their golden opportunity?

Dan Szymborski: He’s still alive!

Dan Szymborski: Though it’s weird how old he is

matt w: Do you have an estimate of the chances that Aaron Judge will have more career WAR than Mike Trout? (Thinking of the most WAR for a single team in the first part of the 20th century, and they may be the only contenders unless Bobby Witt sticks with the Royals.)

Dan Szymborski: I don’t offhand, but Trout’s definitely the stronger favorite

Dan Szymborski: Judge is unbelievable of course, but a 30 WAR shortfall is HARD to make up when you’re already 33

Dan Szymborski: And a run of injuries can bring things down *quickly*

Coolidge: Are the Braves just…..bad? They can’t use the injury excuse anymore. Everyone’s regressed 2 years in a row now. It’s almost like….this is just who they are.

Dan Szymborski: Not *bad*, but they’ve had so many disappointments that it’s hard to call them the class of tne NL East anymore

Hi Dan: Hi Dan

Dan Szymborski: Cassiopeia is sleeping in her cat cave, Justinian’s sitting next to me on the pillow on one of my left desks. Last I saw Constantine was sleeping on an ottoman in the living room. Don’t know what Mercutio’s doing, but I gave him his kidney food right before chat (the SDMA test in his senior blood panel showed very early kidney disease, not enough for medicine, but enough for me to immediately start him on kidney diet)

Dallas: good afternoon! Question- but underlying numbers suggest Henry Davis is (slowly) starting to make better swing decisions and contact at the plate. Considering the pitching staff is raving about his defense, is there a chance he can become a 55 level player akin to what the pirates expected when they drafted him?

Dan Szymborski: I think there *is* such a chance. Catchers are really, really weird developmentally. They disappear suddenly, they blow up suddenly. The error ranges in the median projection for catchers is actually noticeably worse than for other positions, even if we JUST look at offense

Guest: Does Cade Horton stick in the rotation all season for the Cubs? And when will we see Bubba Chandler for the Bucs?

Dan Szymborski: He’s interesting, but I’m not *confident* in Horton sticking. He’s come a long way in command, but he still has to work on the early-Eovaldi problem of having trouble getting ot that strike three

Dan Szymborski: Which is really dangerous; the longer a hitter lingers without being punched out, there more likely they’ll find their pitch

TKDC: If a team could be convinced that Bauer would change how he presents himself publicly and at least pretend to be a halfway decent person, do you think he could be signed, or is his past baggage and current talent level still make him a negative value asset?

Dan Szymborski: This is a hard question for me to ask simply because at this point, I can’t claim to have any particular insight in how a *typical* fan sees these things.

Dan Szymborski: At the end of the day, the average fan doesn’t appear to have any issue with Ozuna or Chapman as long as they’re playing well for their favorite team.

Dan Szymborski: But I’m not sure how mainstream the Bauer story actually was. After all, I live in a baseball journalism bubble.

Dan Szymborski: I’m cynical, so I actually kinda think team’s would be *more* worried about his discipline on social media

Dan Szymborski: So…I dunno?

Boobic: Boy I didn’t expect this hot start from K. Bubic! Confidence he can keep up a sub-3 ERA ROS?

Dan Szymborski: Under three is a REALLY big ask for anyone

Appa Yip Yip: Are you buying what Addison Barger is selling?

Dan Szymborski: I don’t think he’s *this* good. It’s funny, I just mentioned this issue on twitter pre chat

Dan Szymborski: https://twitter.com/DSzymborski/status/1930652332108517504

Dan Szymborski: I think he’s a good part of a team. I don’t think he’s a 130 wRC+ guy

Dan Szymborski: Though it DOES make me happy in that some OOTP players were really annoyed with me that ZiPS was projecting a nobody like Barger to be pretty good

admin: Skenes K numbers are down from last year. Is this intentional, concerning, neither, both?

Dan Szymborski: Not at all concerning. His contact/swinging strike numbers are actualy better than last year.

Jac Caglianone: You tell the Royals to bring him up, and they bring him up. What are you going to tell them to do next?

Dan Szymborski: Give me a few million dollars? I lik emoney.

Alex: What percentile outcome was the Giants’ 2021 season in Zips? Was that the biggest positive deviation from the median projection?

Dan Szymborski: Oh, it was 99.something. I mentioned it on Twitter back then. I think it was 99.7, but let me see if I can find it real fast

Dan Szymborski: Whew, found my tweet from November 2022

Dan Szymborski: 2021 Giants (99.9%)
2012 Orioles (99.7%)
2012 Athletics (99.4%)
2009 Marlins (99.4%)
2010 Blue Jays (99.2%)
—————————-
23 teams in 540 teams (4.3%) hit their 95th percentile projection.

Dan Szymborski: wtf

Dan Szymborski: why is it wrong color

Dan Szymborski: 2021 Giants (99.9%)
2012 Orioles (99.7%)
2012 Athletics (99.4%)
2009 Marlins (99.4%)
2010 Blue Jays (99.2%)
—————————-
23 teams in 540 teams (4.3%) hit their 95th percentile projection.

Dan Szymborski: OK, copied it to text document and then to here

Dan Szymborski: It was the most in terms of actual wins too

Darrenasu: You can start your franchise by picking from the following. Langford, Holliday, Merrill. Who ya got?

Dan Szymborski: Merrill. I love Holliday, but he’s playing at superstar levels right now and only a year older

Dan Szymborski: He’s a plug and play guy. He doesn’t need to get any better.

namenamename: Which of these three players make it to 300-300: Jose Ramirez, Francisco Lindor, Jose Altuve?

Dan Szymborski: I think the first two very likely

Dan Szymborski: Altuve I think it’s closer. I think ZiPS had him at like 60% for 300 HRs (which is of course his chance since he already has the SBs)

Justin: Hi Dan. The framing numbers on Ryan Doumit are unbelievably bad. One year he had a 123 wrc+ and was still worth -3.1 war. Is it even possible that he was that bad?

Dan Szymborski: It’s certainly POSSIBLE

Philip: Yarbrough has gone 5+ innings in his last 4 starts. Is he doing anything different that makes you believe he can be a real starter rather than a bulk guy type?

Dan Szymborski: I’m not at that point yet. I still think of him as a quality inning eater

namenamename: Do either Coby Mayo or Heston Kjerstad work out for the Os? Seems like Mayo is totally lost, although still quite young, but Kjerstad is a few weeks away from being a decent role player

Dan Szymborski: I think they’ll work out still

DJ Tofu: What do you think is wrong with LaMonte Wade? What are some players with similar steep and rapid drop-offs in performance?

Dan Szymborski: The BIG dropoff in hard-hit percentage is concerning

Dan Szymborski: Even with some BABIP bounceback, which is inevitable, the lack of power this year is supportable

Dan Szymborski: I think the downgrade in the projections have him right

Dan Szymborski: But Wade’s problem with the Giants is that he’s a role player/borderline starter who HAS to hit to have value

drplantwrench: do you think fWAR should include xBA type stats?

Dan Szymborski: No

Guest: is ZiPs too hard on Ben Rice defense (-12.6 ROS)? Or is this just a function of DH that I’m misinterpreting ?

Dan Szymborski: ZiPS is certainly twoo hard on the defense

Dan Szymborski: too

Dan Szymborski: (there’s some position mixed in there)

Dan Szymborski: but he had absolutely oddly poor numbers in ZiPS probabilistic method of evaluating minor league defense PBP

Dan Szymborski: he should have a much sunnier defensive projection this fall

drplantwrench: i guess i should follow up – why dont you think fWAR shouldnt include xBA type stats?

Dan Szymborski: Because we’re going too far into expected value rather than what actually happened. In a different way than FIP with pitching

Dan Szymborski: People sometimes describe RA9 WAR as what actually happened, but that’s a bit different because in that case, it’s what happened to the team. The pitcher has to share responsibility with defense.

Dan Szymborski: There’s not the same issue with an individual batter

Dan Szymborski: Now, we do have to figure out batter vs. *opponent* valuation

Dan Szymborski: But batters have a mroe repeatable tendency to perform better/worse than their expected stats than pitchers do in their FIP vs. ER stats

Dan Szymborski: Which can be seen in how ZiPS *mixes* the expected stats with the actual ones as a better predictor of the future than just the expected

Spider: Why are there some players who ROS ZiPS has for negative projected WAR but ZiPS Depth Chart has them for positive projected WAR? I understand they have different playing time projections but the rate stats are the same, so wouldn’t they both always be either negative or positive for each individual player?

Dan Szymborski: Can you give an example? I don’t actually run anything for this part as I have little to do with the depth charts (unless both Jason and Jon get ebola or something) and I don’t do the FG implementation of the depth charts into a model

Insert Witty Name Here: Thought experiment. If you watched every single game, every single pitch of one team for the entire year, how close would you be able to predict their WAR?

Dan Szymborski: I expect very poorly.

Dan Szymborski: I like to think I’m an insightful observer of things, and I have watched as much baseball as anyone on the planet. But I’m subject to the same cognitive biases as anyone, even if I was in person for every game (violating space-time constraints) and could see stuff that the cameras don’t show you that are necessary to evaluate defense

Dan Szymborski: Plus, the human brain is really bad at this kind of thing.

Spider: For example, Tyler Phillips is projected -0.1 WAR by ZiPS depth chart and 0.3 WAR by ZiPS. Different amount of innings, yes, but how can this be correct with the same ERA/FIP/etc.?

Dan Szymborski: Hmm

Dan Szymborski: Ah, it’s because the different roles.

Dan Szymborski: FG is still a bit awkward at players in different roles than projected. Still trying to figure out a technological solution for that

drplantwrench: id argue that xBA is what actually happened to a certain extent, but also completely understand it too. it would be fascinating to see how an expected-based WAR (xWAR) would fare over different sample sizes against the other two WARs.  also i just want more types of WARs to confuse people

Dan Szymborski: I think zWAR is a good middleground between fWAR and bWAR approaches to run prevention (ZiPS WAR gives more credit to the pitcher the longer they out/underperform their ERA relative to team defense expectation). But the *value* of a very slightly more accurate WAR is more than cancelled out, I believe by the negative value of creating more confusion

Smokin’ Jim Leyland: This is probably more of a question for Jay Jaffe…but assuming that the Cubs and Reds are out due to his minimal time spent with those orgs – which cap do you think Nick Castellanos wears at Cooperstown, Tigers or Phillies? And will he go in the “Nicks” wing of the hall, alongside Mick Markakis?

Dan Szymborski: Phillies

Dan Szymborski: If he somehow had a Hall of Fame career, it’s probably going to have to happen from an explosion of offense coming SOON, so it would be more PHillies-related

Dan Szymborski: And if that explosion doesn’t come before 2027, it just wouldn’t come – I doubt anyone’s giving him a significant contract after 2026

Vacay: What is your favorite type of vacation? Mountains, beach, fun city, road trip…?

Dan Szymborski: Beach

Dan Szymborski: I like road trips too

Dan Szymborski: I’m a fat dude with a degenerative back condition and a hernia; I don’t have much interest in long mountainous hikes!

J: I think drplantwrench needs a reminder that xBA doesn’t account for spray angle

Dan Szymborski: I’ll note that zBA does

jjq: xBA/xSLG doesn’t take into account where the ball what hit does it?  I think until they do then they aren’t as useful as we would like them to be. Hard hit to CF is just a lot different than hard hit to pull side.

Dan Szymborski: correct

All Star: Who is your obligatory choice for an All Star from the lowly Rockies? Jake Bird?

Dan Szymborski: Bird or maybe just Hunter Goodman

Guest: Do you buy Chris Bassitt’s recent struggles as his normal for this season? Or was his April sta

Dan Szymborski: He’s a decent mid-rotation inning-eater who had a hot April and a cold May. Something the cake is just a cake

drplantwrench: how valuable is Jo Adell? he’s an enigma wrapped in a mystery to me.  had this debate recently, could the Angels net a 40-45 FV prospect for him in a trade? or more or less?

Dan Szymborski: Less I think. I don’t think his value is perceived as vaery high

Jacob: Hey Dan, how does the “2025 Updated In-Season Projections” work? Is it a periodic update (at what frequency?) based on season to date performance? I presume it’s not just prorating the pre-season projections through the end of the season.

Dan Szymborski: The updated in-season projections are run every night using the simplified in-season ZiPS model

Dan Szymborski: so ROS is always changing!

tacothecat: Do you think the Giants will bring up Eldridge to solve their 1B issues?

Dan Szymborski: It’s a realistic possibility, though I think they’re goign to want to see him mash at AAA first

Addison Barger fan club: Back to Barger for a second. Does ZiPS take into account his quality of contact? His exit velocity is elite.

Dan Szymborski: more robust model does have him 117 rest of season rather than 112

Sirras: Is today a food day?

Dan Szymborski: IT is not

Bookworm: Do you read books? And if so, what are you currently reading or anything you read recently you liked?

Dan Szymborski: LIke many, I’m reading Original Sin

Dan Szymborski: (no comment on the politics, naturally)

Dan Szymborski: I’ve actually been a little behind on reading for me, which is unusual

Dan Szymborski: A large percentage of my reading is military history

Dan Szymborski: I”m a little slow right now reading because one of my kindles isn’t working

Dan Szymborski: so I’m mostly reading on the bathroom kindle at the moment

Dan Szymborski: That reminds me I need to order a new paperwhite

Insert Witty Name Here: How’s Cistulli doing?

Dan Szymborski: I’m not sure. This sounds weird, but I don’t usually talk much with people in baseball if they’re people I already know, simply because I don’t want to put them in an awkward position talking things with me

Bob Loblaw: Any concerns on PCA’s low walk rate? I like the aggressive approach, but the chase rate is really bad.

Dan Szymborski: No, I think it’s just a player profile thing

Dan Szymborski: On that note, it’s time for me to hit the ol’ dusty trail for another week

Dan Szymborski: Thanks for coming everyone!