Dan Szymborski FanGraphs Chat – 2/16/23
| 12:02 |
: SzymChat is *back* now that I’m firmly in the domain of the living.
|
| 12:02 |
: As a Braves fan with very little to complain about, I’ve thought of something that is driving me nuts. Why is Ozzie Albies continuing to switch hit? He’s bad left handed and much better right handed. Are players stubborn about this? Do front offices not suggest this? I feel like I’m missing something obvious because it seems like such a no brainer.
|
| 12:02 |
: Players tend to be stubborn about dropping pinch-hitting.
|
| 12:03 |
: And with Albies, he isn’t *so* bad left-handed that the issue is likely to be forced.
|
| 12:04 |
: I mean, he’s still at around a .750 OPS left-handed
|
| 12:04 |
: Dan !!! Can you please please please tell us when Prospect Week is ?!??! We don’t know how much longer we can last without it !! |
| 12:04 |
: Next week, I believe.
|
| 12:04 |
: How does ZiPS perform H2H sims in your model. Does it run through actual 9 inning games for each matchup, or does it generate single game WAR scores for each players and compare to determine a winner?
|
| 12:05 |
: For simming entire seasons, I don’t find getting too in the weed H2H is all that useful. I just do probability for odds ratio (plus home/road adjustment) based on whatever roster strengths ZiPS has for the teams that year
|
| 12:05 |
: What were your major takeaways from Evan Drellich’s book?
|
| 12:05 |
: I haven’t read it yet!
|
| 12:05 |
: Clarke, German or not on roster to replace Montas?
|
| 12:05 |
: Pretty sure it’s going to be German
|
| 12:05 |
: What uncertainty margins do ZIPS projections typically have?
|
| 12:05 |
: Depends what you’re talking about
|
| 12:06 |
: it’s a more complex q than what is your shoe size!
|
| 12:06 |
: Dan, what is the correct ratio of bacon to lettuce to tomato on a BLT?
|
| 12:06 |
: Fill up a sandwich with as much bacon as it will hold. Then take out only enough bacon to put in a slice of tomato and/or a lettuce leaf.
|
| 12:07 |
: How worried should we be about Aaron Ashby ?
|
| 12:07 |
: Some, but I wouldn’t panic unless something more substantial is announced.
|
| 12:07 |
: If baseball were played with 27 consecutive outs instead of 3, which types of players would become more/less valuable?
|
| 12:07 |
: I would imagine ball in play people would be because you’ll have a much higher percentage of at-bats with runners on base
|
| 12:07 |
: Is the Bally bankruptcy the first sign of the long predicted (by yours truly at least) burst of the MLB broadcast $ bubble?
|
| 12:07 |
: I don’t think it’s going to be a massive burst, but there will be some adjustment.
|
| 12:07 |
: You’ve noted before that ZiPS does not attempt to project playing time. I don’t think I’ve ever heard you say why that is the case though. Why is that?
|
| 12:08 |
: A hammer is for nails, not for cleaning glass.
|
| 12:08 |
: ZiPS *does* actually do a pretty good job if you’re *only* looking at established players
|
| 12:09 |
: Since age, history, position, specific injury info, is a good predictor
|
| 12:09 |
: But ZiPS doesn’t do “team decisions.” I don’t see the use of a projection system that decides when the Cubs will start using Matt Mervis.
|
| 12:10 |
: Now, I have to do that for seasons, but my philosophy is as little human intervention as possible.
|
| 12:10 |
: And frankly, I’m more interesting in using a projection system to estimate how good, say, Miles Mastrobuoni *would* be rather than using a projection system to estimate how much the Cubs like Mastrobuoni.
|
| 12:11 |
: .000/.000/.000 in 0 PA conveys no useful information in the context of a projection system
|
| 12:11 |
: Will ZiPS DC be adjusted to account for recent injuries and trades? Thinking Montas and AJ Puk etc.
|
| 12:11 |
: ZiPS DC is automatically adjusted by the depth charts
|
| 12:11 |
: Ranger Kyle Cody seems to have been left out of the ZIPS projections both on the team page and his player page.
|
| 12:11 |
: Yeah, he’s on the to do list for the next submission
|
| 12:11 |
: The ZiPS projections for Masataka Yoshida and Zoilo Almonte are awfully similar…
|
| 12:11 |
: Yeah, that’s an error. Zoilo Almonte shouldn’t actually have a projection
|
| 12:12 |
: I’ve already submitted that to be fixed.
|
| 12:13 |
: Basically, I first projected Yoshida before an MLBAM ID was ready and with changes to how it was mapped, I didn’t see it assigned him the last MLBAM ID for a name alphabetically for a player projected over the last five years (Zoilo Almonte). So there were two Yoshida projections, one with Almonte’s MLBAM ID and one with Yoshida’s.
|
| 12:13 |
: What Orioles will surprise people the most (in a good way) this year?
|
| 12:13 |
: I’m certainly hoping Kremer because ZiPS has a fetish for him
|
| 12:14 |
: I still think DL Hall could break out very quickly if things go the right way
|
| 12:14 |
: Do we have to call him IL Hall now?
|
| 12:15 |
: How much can a player getting a cup of coffee in the big leagues affect their projection for the next season? For example, how much better is G. Henderson’s projection given his performance in 132 ABs?
|
| 12:15 |
: It’s not really a big difference given how good his translations were
|
| 12:15 |
: Why do you think ZIPs is down on Tony Gonsolin comparing to his results LY. Was last year really a career year for him?
|
| 12:15 |
: I mean, he already was a run better than his FIP last year and didn’t have a history of being this good
|
| 12:15 |
: ZiPS is the *kindest* to him!
|
| 12:16 |
: 3.54 ERA in ZiPS vs. 4.25 THEBAT and 4.38 Steamer
|
| 12:16 |
: What do you think is the best way (or a way many haven’t thought of) for a dynasty/fantasy baseball enthusiast to use your projection system to clean useful info?
|
| 12:16 |
: Well, ZiPS DC.
|
| 12:17 |
: Harold Baines had a 4.3 BWAR (4.8 FWAR) in 1984. Did any HOF’er have a worse best season than this?
|
| 12:17 |
: If you hang on, I can get the worst best fWAR for HOFers.
|
| 12:17 |
: I’m just going to do hitters because reliever and stuff
|
| 12:17 |
: During Drew Rucinski’s 4 years in the KBO he had BB/9s of 2.28, 2.80, 2.77, and 1.52, but ZiPS projects him for a 3.73 BB/9. Is this due to his age, the fact that he put up those numbers in a lesser league, or his previous MLB experience being weighted more heavily than I expect?
|
| 12:18 |
: ZiPS thinks of the KBO as an AA 1/2 league, so there’s a translation hit
|
| 12:18 |
: How does ZIPS account for Eddie Rosario’s mid season eye injury and surgery? Is there an adjustment for specific injury types when trying to project future seasons?
|
| 12:18 |
: There is, but it’s not granular enough to capture something like this
|
| 12:19 |
: I can’t believe Zips predicted the 2023 Phillies to miss the playoffs? /s
|
| 12:19 |
: Punches!
|
| 12:19 |
: What player or team projection does Dan the human disagree with ZiPS on the most?
|
| 12:19 |
: I haven’t really sat down and thought about it. My brain’s basically been fried for three weeks.
|
| 12:20 |
: The last couple team of ZiPS was a race against the clock because both my mom and sister were violently ill and I could feel my lungs start to get cloudy
|
| 12:20 |
: It’s why you haven’t seen as much of me as normal last few weeks.
|
| 12:20 |
: Mom eventually was in the hospital for three days with double pneumonia
|
| 12:21 |
: Re: Drellich’s book. Take a shot each time he describes Kevin Goldstein as “punk rock”
|
| 12:21 |
: I *do* have Kevin in WWE2K
|
| 12:22 |
:
|
| 12:23 |
: I hope you’re feeling better! Do you watch spring training games in a typical year, and do the rules changes make you any more or less likely to watch any this year?
|
| 12:23 |
: I do, though I most follow individual players *within* games than really the game itself.
|
| 12:23 |
: More likely to watch spring training because I’m curious about the effects
|
| 12:24 |
: I don’t have a problem with the rule changes except for zombie runners becoming permanent.
|
| 12:24 |
: I think it is a bit of a problem — though overrated by many — that MLB offense is basically “solved” so I don’t mind them altering the meta to possibly incentivize other styles of offense
|
| 12:24 |
: Many of us are probably wondering about the Giants projections in ZiPS. Were you surprised by how the NL West looked wrt to SF?
|
| 12:25 |
: Yeah, Giants about 3 wins higher than expected
|
| 12:26 |
: What’s the most impressive baseball play that you witnessed live on television? Thank you Dan, very cool!
|
| 12:26 |
: Devon White’s WS catch maybe?
|
| 12:26 |
: When will you be re-running the Yusei Kikuchi ZiPS projections? He has a beard now and I don’t think ZiPS knew that when you ran them before.
|
| 12:26 |
: heh
|
| 12:26 |
: Do you think that Austin Riley will maintain the gains he had last season or slide back to more of his career norm? I’m not sure there is another level he can get to.
|
| 12:26 |
: I think he’ll regress slightly and then improve from there
|
| 12:26 |
: Why is ZiPS right to be higher on the Cardinals than most other projections systems? Why is it wrong to?
|
| 12:26 |
: I don’t know yet if ZiPS is right to be higher!
|
| 12:26 |
: Though my methodology for seasonal play tends to make Cards/Rays look better than others since I try to estimate the value of depth
|
| 12:26 |
: Bally bankruptcy is a case of corporate malpractice, not anything to do with MLB or sports broadcasting viability.
|
| 12:27 |
: Are playoff odds called “odds” because it is odd the O’s have a 10% chance of making the playoffs?
|
| 12:27 |
: It’s because we’re all sometimes a little sloppy with terminology
|
| 12:28 |
: Hi Dan, you’re indeed the man. My question concerns the playoff odds, in which I know your involvement ends at the handing over of your projections. However, I thought you might still have insight into them. I’m curious about the difference between projected WAR and win totals, especially with a more balanced schedule. For instance, Cleveland is projected to be worth more wins above replacement than SF, Milwaukee, and Seattle, but fewer team wins. I’m curious how to grapple with this. Thanks!
|
| 12:28 |
: I honestly don’t know.
|
| 12:28 |
: I don’t play with the innards in there at all.
|
| 12:28 |
: The Twins seemingly solved their biggest issue from last year, depth. Am I just a homer thinking they could win 90?
|
| 12:28 |
: They COULD, it’s just not their mean projection.
|
| 12:29 |
: What was the ZiPS projection this year where you went “oh, that can’t be right” but then it was?
|
| 12:29 |
: Addison Barger. He popped up at the very start when I was doing test runs
|
| 12:29 |
: Is it just me, or does it seem like little rule changes that no one has asked for (mound visits? really?) seem to now be annual feature of MLB. If so, shouldn’t the fans be able to submit/ vote on proposals—that would be a “better game” IMHO.
|
| 12:30 |
: I’m not sure fans are particularly well equipped to evaluate the consequences of rule changes
|
| 12:30 |
: For instance, some of the stuff that’s cut down doesn’t save a lot in pure time
|
| 12:30 |
: but it saves a lot in time in which nothing is happening.
|
| 12:30 |
: It’s not the length of games that’s a problem, it’s the dead time within the games.
|
| 12:31 |
: And quite frankly, in the same way I find blitz chess more compelling to watch than standard timings, I think time pressure is a good function to have in most sports.
|
| 12:31 |
: Like those AI sites, are you teaching ZiPS how to draw pictures of, oh, I don’t know, Rob Manfred in “Land of the Lost”?
|
| 12:31 |
: Ha, integrating GPT-3 into ZiPS is beyond my skillset
|
| 12:32 |
: But I did think my Manfred ones were fine AI work!
|
| 12:32 |
:
|
| 12:33 |
:
|
| 12:33 |
: Avocado on a BLT to make it a BLAT?
|
| 12:33 |
: Meh
|
| 12:33 |
: I’m not a huge avocado guy
|
| 12:33 |
: the mouthfeel never is quite right for me
|
| 12:33 |
: I like guac in certain applications
|
| 12:33 |
: Does ZIPS take into account the Statcast data or the quality of pitching a batter faces? For example, it has Kim repeating last year despite bad EVs and barrels. And he absolutely demolished position players pitching (2 HRs and 1 2B).
|
| 12:33 |
: ZiPS does take this kind of data into account
|
| 12:33 |
: Any NL SP that you found to have better ZiPs projections than you expected? Pfaadt, Flaherty, Drey Jameson, DeSclafani, Elder all come to mind for me.
|
| 12:34 |
: That’s a pretty good list actually
|
| 12:34 |
: If Judge has several more seasons like last year and then a career ending injury for what number of seasons would the Yankees be happy they signed him? Do they have injury insurance on him to make the real answer less than one even if they can’t admit that?
|
| 12:34 |
: Depends how many they get!
|
| 12:34 |
: I’m honestly not sure the current status on injuries. Some years back, teams were having trouble getting insurance at reasonable rates as insurers were becoming more wary. I’m not sure what the current status is around baseball; I simply haven’t given it much thought in a long time.
|
| 12:35 |
: This morning a friend and I were debating the percent likelihood the Rockies would finish at .500 or better this year. What would ZIPS say?
|
| 12:35 |
: 6%
|
| 12:35 |
: Between PECOTA / Steamer / The Bat / Zips, Zips is the only projection with a higher WAR (5.8) for Adley Rutschman than for Julio Rodriguez (5.4).
Reasons for Zips relative exuberance for Rutschman? |
| 12:35 |
: It’s hard to say because I don’t run the other systems.
|
| 12:36 |
:
|
| 12:36 |
: The giants were projected to win 88 games and the padres 91. Aaron Judge should be 5 wins better than their current corner outfielders. Would the giants be favorites in the west with judge or would they have not signed a couple good players the real giants signed because they already spent the money on judge?
|
| 12:36 |
: It does seem like this would be the case
|
| 12:36 |
: On the topic of switch hitters, can ZiPS do projections for guys if they didn’t switch hit? I’m assuming this would just mean a projection where they only face one side for a full season, but would it be able to use ratios to see what to possible expect if someone gave it up like Mullins did?
|
| 12:36 |
: It can, but it’s not “stock” so I have to do some extra work
|
| 12:36 |
: Eric Thames retired this morning man was that guy fun to watch for a while |
| 12:37 |
: Any gardening in store this year Dan? I’ve managed to grow a ginkgo sapling from a seed that I collected from a neighbor; I’m quite proud of it.
|
| 12:37 |
: I’m not outdoorsy
|
| 12:37 |
: I basically have grass and trees.
|
| 12:38 |
: why is everyone down on a barely 23 year old Tucupita Marcano who has hit FAR FAR better than a 24YO Jarred Kelenic and has much more versatility and still the ability to hit 12-15 hrs and steals and nowhere near his peak?
|
| 12:39 |
: That’s extremely speculative too
|
| 12:39 |
: Neither have hit in the majors but Kelenic has the better minor league record
|
| 12:40 |
: Do you think Carlos Correa will hit 30+ home runs this year or ever? The kind of money he is getting seems to project that his best years are yet to come.
|
| 12:40 |
: I don’t think he needs to hit 30
|
| 12:40 |
: H’ell probably hit 30 *some* year
|
| 12:41 |
: Going through the remaining free agents, I was surprised to see Rougned Odor projected for a 101 wRC+ by ZiPS. Any idea what it sees in him specifically?
|
| 12:41 |
: I believe ZiPS simply thinks he has more power upside than Steamer does
|
| 12:41 |
: If a team’s players played at an average of their 60th percentile projection, how far higher than its 60th percentile projection would the team’s result be? I’m trying to get my mind around the level Rockies players would have to play at for the team to achieve the .500 record the owner said is possible.
|
| 12:41 |
: I’m not actually sure off-hand
|
| 12:42 |
: You’ve got Vaughn Grissom as a 2-win player, which I found surprising (though as a Braves fan it was a pleasant surprise!) — and with Alan Trammell and Tony Fernandez as comparable players, my head is kind of spinning. What is the system picking up on in Grissom that is pointing towards a Hall of Very Good/Hall of Fame-type career?
|
| 12:43 |
: I mean, dude’s got an >.850 minor league OPS through age 21 and he was really good in his cup of coffee
|
| 12:43 |
: While running teams’ projections do you just play noises of zippers in the background?
|
| 12:43 |
: dear god no
|
| 12:43 |
: How long do you think it will be until we have a full sense of how consolidation of the minors affected age-level adjustments in models? Or are we there already?
|
| 12:43 |
: Between this and the COVID season it’ll be hard to separate. I think we’ll have something really interesting to look back on at the end of the decade.
|
| 12:44 |
: Are you sold at all on Chas McCormick as a legit starting CF now? I love the guy…but keep eyeing what the A’s are going to do with Laureano thinking he might be a nicer fit for the Astros in CF. Especially since it sounds like Brantley may not even be a lock to be back by opening day
|
| 12:44 |
: As a decent starter, I’m sold. But he’s not a long-term piece.
|
| 12:44 |
: Aren’t people underselling the White Sox? They had a huge number of injuries last year (Grandal, Anderson, Eloy, Robert, Moncada, Lynn), and if just half of those players are healthy, the team will do considerably better. Beni signing also allows Vaughn to move to 1B and improves the defense.
|
| 12:44 |
: It’s possible, but how many White Sox players do you feel better about than last year at this time?
|
| 12:44 |
: Because injuries *will* happen
|
| 12:44 |
: And the Sox depth is not good
|
| 12:45 |
: A Tink Hence ZiPS would pair real nice with MILB.tv coming to MLB.tv for 2023.
|
| 12:45 |
: He only gets a projection for the purpose of ZiPS Top Prospects
|
| 12:45 |
: vidal brujan – nothing more than reserve pick for roto leagues, right?
|
| 12:46 |
: That’s probably accurate
|
| 12:46 |
: A take I’ve seen: Rule changes are mere tinkering, the core issue is that pitchers are too awesome, everything that isn’t lowering the mound is asinine if the goal is more on-base action. Agree?
|
| 12:46 |
: It’s possible, but there’s nothing wrong with being gradual with changes
|
| 12:46 |
: Would 360 million be a good price for 3 straight 10 WAR seasons?
|
| 12:46 |
: No
|
| 12:46 |
: Does ZiPS ever run projections for draft prospects?
|
| 12:46 |
: I only run college-only players if I really have to
|
| 12:46 |
: Good morning Dan. You ever mess around with sous-vide? It’s a GAME CHANGER.
|
| 12:47 |
: I actually haven’t
|
| 12:47 |
: Please tell the owners of the site to stop calling it a ‘ghost runner’. The runner in extra innings is very real, has corporeal form, and thus is not ghostly in any way.
|
| 12:47 |
: Well, idiotic rule runner would be most accurate
|
| 12:48 |
:
|
| 12:48 |
: I was playing what and gave myself the benefit of hindsight and convinced my hypothetical employer to give me a 233 million budget. I figure signing one of the big 4 SS instead of Brandon Dury, a second tier starter and signing the other people the same way the real life Angels did is the best case scenario and that still leaves them as a lower end wild card being their most likely result.
|
| 12:49 |
: I think Correa gets them fairly close to the Astros
|
| 12:49 |
: I’ve long thought that there seems to be a positive bump in long-term outlook for a guy who gets called up and doesn’t completely self-immolate, on-field and off. Do you evaluate “non-disastrous call up” as a factor independent of actual production during MLB playing time?
|
| 12:49 |
: No, I do not.
|
| 12:49 |
: How much do you expect the new rules to inflate SB totals? Last year Jon Berti led the league with 41 SBs. Do you think someone gets to 50 this year? 60?
|
| 12:50 |
: I think it will be more muted than in the minor leagues
|
| 12:50 |
: But we’ll have an uptick
|
| 12:50 |
: How heretical is it for me to say I like the corporeal runner rule?
|
| 12:50 |
: Could ZiPS suggest a reasonable lower standings effect for a Manfred Man extra-inning win? Like, that’s only really 1/3rd as good as a 9-inning win?
|
| 12:50 |
: I’m not sure I’m following
|
| 12:51 |
: Do you think ashby gets a starting gig when healthy?
|
| 12:51 |
: Yes
|
| 12:51 |
: Any particular thoughts on Miguel Vargas, ceiling seems a bit low for him. I don’t how much to believe his ISO numbers from the PCL
|
| 12:51 |
: Yeah, we have to take minor league hitter numbers with a larger grain of salt than through the 2000-2020 period
|
| 12:53 |
: Can the Yankees cut Montas and save money?
|
| 12:54 |
: Well, they could save the MLB minimum if another team signed him
|
| 12:54 |
: otherwise no
|
| 12:55 |
: In the past, contracts for MLB arbitration-eligible players that agreed on contracts to avoid arbitration weren’t guaranteed until opening day
|
| 12:55 |
: and teams would only have to pay 30 days or 45 days termination pay depending in timing of release
|
| 12:55 |
: but that changed in new CBA
|
| 12:56 |
: Yankees/Montas agreed on a deal to avoid arbitration hearing
|
| 12:56 |
: so that became guaranteed right there and then.
|
| 12:56 |
: Follow up – I’m thinking of a corollary to some hockey leagues that award fewer standings points for overtime wins versus regulation wins. Could a projection system be used to do something comparable in baseball — an extra innings win being a fractional “win” in the standings?
|
| 12:56 |
: Meh, I don’t like the system but I don’t see this as an improvement
|
| 12:56 |
: Odds the Orioles leaving Grayson Rodriguez off the OD roster?
|
| 12:57 |
: I think it’s fairly likely
|
| 12:57 |
: though he’ll be up fairly quickly
|
| 12:57 |
: do you ever refer to running a projection on a player using ZIPS as “zipping” them?
|
| 12:57 |
: Heh, I have not
|
| 1:00 |
: OK, worst best season for HoF hitters by fWAR
|
| 1:00 |
: Schalk (3.0, 1915)
|
| 1:00 |
: Ferrell (3.1, 1932)
|
| 1:00 |
: Waner (4.0, 1929)
|
| 1:01 |
: A man a dan a healthful recovered chat, ahcderevocerlufhtleahanadanama
|
| 1:01 |
: Seeing the updated numbers for guys like Willie Wells (dude was basically Giancarlo Stanton at SS in terms of career offensive production) was really cool. Well done.
|
| 1:01 |
: I didn’t have anything to do with it, but I’m happy that you’fe found it useful
|
| 1:01 |
: Do your projections for 2023 factor in any impact of rules changes this year?
|
| 1:01 |
: No, it’s pure guesswork. I’d rather people know that the projections don’t include rule changes rather than people having to guess how much of a projection is the Szymborski Arbitrary Rule Change Effect
|
| 1:02 |
: I feel like you’re the most likely FG chatter to have a friend group who bombards these chats with silliness to make you laugh.
|
| 1:02 |
: It’s likely. I admittedly do things in life because they entertain me
|
| 1:02 |
: So worst HOF seasons … how accurately does WAR measure the defensive contribution of a catcher in the dead-ball era (Schalk?)
|
| 1:02 |
: Probably not very
|
| 1:02 |
: But it goes both ways; it’s a problem of accuracy not bias
|
| 1:02 |
: cutting an injured player to save money. What an awful thing to do, so I wouldn’t put it past billionaires to try.
|
| 1:02 |
: I mean NFL
|
| 1:02 |
: Non-Twins question….those 70s dinners were “interesting”. Have you asked the AI to make hearthstone cards like you used to do in the day?
|
| 1:03 |
: I still make Hearthstone cards!
|
| 1:03 |
: Unfortunately, text on AI stuff tends to be demonic confusing words, so it doesn’t sim word heavy things very well
|
| 1:03 |
: I’d love if they could mash together the art and text AI
|
| 1:04 |
: I have tried though
|
| 1:04 |
:
|
| 1:05 |
: That was for Ragnaros the Ice Cream Lord
|
| 1:06 |
: Okay let’s say you have a pork tender loin.
|
| 1:06 |
: Microwave – -1.0WAR Baked in oven 0.5WAR pan sear 1.0WAR Smoked 2.0WAR BBQ 2.5WAR Sous-vide 6.0WAR
|
| 1:07 |
: See, I’d probably smoke it
|
| 1:07 |
: In any case, it’s time for me to depart for another week.
|
| 1:07 |
: Thanks ofr coming everyone!
|
Dan Szymborski is a senior writer for FanGraphs and the developer of the ZiPS projection system. He was a writer for ESPN.com from 2010-2018, a regular guest on a number of radio shows and podcasts, and a voting BBWAA member. He also maintains a terrible Twitter account at @DSzymborski.

