12:01 |
: Greetings!
|
12:02 |
: I’m sad to report that the chili I was making last week when we chatted did not turn out well.
|
12:02 |
: I got careless with the salt and the whole thing was way too salty.
|
12:02 |
: So I had to eat it with VERY improper cheese and sour cream added in to mute the saltiness.
|
12:03 |
: Can Gil be a front line guy? Or Schmidt? Both have looked great
|
12:04 |
: Both are certainly making their cases. Pitching development is weird, so when a guys’ working out a lot of the time it’s just him working out
|
12:05 |
: Ther’es nothing particularly head-scratching in the peripherals or statcast data
|
12:05 |
: What do you think about Volpe? Could be better under the hood but seems like xwOBA underrates him given his spray angles
|
12:05 |
: the zstats do use spray information and like him better
|
12:05 |
: With the Martian on his way back, why is it that ZiPS seems so down of him? ZiPS has him for a 98 wRC+ despite his high prospect pedigree and stellar performance (even if it was only 8 games) last year.
|
12:06 |
: I mean “even if it was only 8 games” is a pretty big if. It’s a bit like saying Abraham Lincoln only REALLY had a couple bad minutes during the play
|
12:06 |
: ZiPS and Steamer are very close on this. I know ZiPS loves his long term potential
|
12:06 |
: but he was only OK in AA last year
|
12:06 |
: Now, the young part is why the long-term upside is huge
|
12:07 |
: do you believe in pork chili? bison? chicken?
|
12:08 |
: As long as its still primarily meat-and-chili peppers sure
|
12:09 |
: Weirdly, there’s no chili alignment chart
|
12:09 |
: Do you think baseball analytics has gone a bit too far, and sucked some of the fun out of the game? I loved sabermetrics but now am worried the increased strikeouts and pitch designs are creating an unbalanced game that is less fun to watch. How can we model fun factor?
|
12:09 |
: I think aesthetics are in the eyes of the behodler
|
12:09 |
: the more interesting question is: let’s assume for the sake of argument that it’s true, now what?
|
12:09 |
: I posed this on Twitter the other day
|
12:10 |
: and as expected, the “solution” for most people was to not use analytics
|
12:10 |
: Which is an absolute cretinous reponse
|
12:10 |
: response
|
12:10 |
: If a portion of a game is “solved” in a sense that the winning strategy is known and excludes alternate strategies, and that strategy isn’t fun, then the solution is to change the rules
|
12:11 |
: Because for many people, the solution is to just expect teams to un-know some basic things, which isn’t going to happen
|
12:11 |
: Would you convey one non-baseball societal opinion that (a) you feel strongly and (b) perhaps we don’t know?
|
12:12 |
: OH god, are you trying to get me into trouble?
|
12:12 |
: OK, on opinions I haven’t expressed before, let me think
|
12:12 |
: The problem is I’m opinionated and a big mouth
|
12:13 |
: hmm…MAYBE my belief that no election ballots should list a political party as I consider that advertising, and names should be in randomized order on each ballot
|
12:14 |
: Am I real? K rate down 12% and currently 8th in MLB in average exit velo..
|
12:14 |
: The power probably. K should be improved, but not this much as still below average contact
|
12:15 |
: How is injury data incorporated into statistical analysis (both ZIPS and teams’)? Is it just a categorical flag for each (torn UCL, etc.) or is it something more fine-grained?
|
12:15 |
: The former
|
12:15 |
: The data doesn’t really exist to do it fine-grained
|
12:15 |
: does knowing the inherent flaws / gaps in ZiPS give you an edge on baseball betting? Like at this point, the sports books’ odds are probably derived similarly to ZiPS, so do you ever see “Oh, here’s why they put the over here, that’s a flaw in the system that overvalues X” so then you bet the under?
|
12:15 |
: No, I don’t try to eleventy-dimension chess it
|
12:15 |
: If I see a good EV+ I take it, and if I don’t, I don’t care
|
12:16 |
: errr +EV
|
12:16 |
: Did you see that the state of Indiana declared that tacos are in fact a sandwich. Actually. It was a legal decision to settle a mini mall lease dispute over what constitutes a sandwich.
|
12:16 |
: Only a sandwich in the statuatory sense
|
12:17 |
: What goes into projecting a players BABIP? Is it just speed and exit velo?
|
12:18 |
: line drive rate, spray; opposite field hit rate actually is a better predictor than speed!
|
12:18 |
: Hey Dan. How is your back treating you?
|
12:19 |
: Much better since disposing of some of the extra weight
|
12:19 |
: Are we all going to forget about Brennen Davis in a few weeks, or is he an excellent prospect again? If it’s too early to tell, how many more weeks/months of good play would it take to give a better answer?
|
12:19 |
: I’m sorry to say I’m skeptical
|
12:19 |
: Can’t really give X weeks because that depends how good he continues hitting
|
12:20 |
: If he continues a 170wRC+, it’s a lot shorter time than if he continues hitting, say 130 wRC+
|
12:20 |
: What is the difference between Full Fat ZiPS and ROS ZiPS?
|
12:20 |
: Use of Statcast and plate discipline data is a big one
|
12:20 |
: and integrated minor league translations
|
12:21 |
: See, here’s the problem:
|
12:21 |
: To do a full ZiPS run takes about a day each for the PC that it runs on, so it can’t be updated daily
|
12:21 |
: And even on my main PC, it takes about 16-18 hours, during which I can’t really do much with my main PC
|
12:22 |
: bonus question: if you run Full Fat ZiPS, how much of 3YR ZiPS becomes ~untrustworthy (ie what % of the ZiPS player pool do you think goes through significant changes)?
|
12:22 |
: I don’t think I can give a hard and fast rule for a generality
|
12:22 |
: It’s OK, Kevin. We know your chili normally works out.
|
12:22 |
: Hey, I have more hair than him!
|
12:22 |
: Whenever you run projections using past data, how do you handle 2020? Do you use the numbers as they are or maybe a projection model of what they could have been over a full season? Was just curious!
|
12:22 |
: Poorly?
|
12:22 |
: I guess that’s not a good answer
|
12:23 |
: ZiPS considers 2020 as a mix of actual 2020, and the ROS projection calculated at the moment the season ended
|
12:23 |
: and thankfully, it’s a really small part of projections now
|
12:24 |
: Does ZiPS think the Royals can keep this up at all?
|
12:24 |
: In terms of a >.600 WP? No
|
12:24 |
: but even at the .485ish winning percentage expected the rest of the way, that still makes them a playoff team like 60% of the time
|
12:25 |
: BTW Hokius, flipping in MLB the Show is getting closer to normal levels of profitability, especially off peak
|
12:26 |
:
|
12:26 |
: I’d also like an update on how Cold Steel III is going for you, though you don’t necessarily have to give it here
|
12:26 |
: Speaking of Oneil Cruz, what are your thoughts on his performance thus far and likely development?
|
12:26 |
: He hit ball really ahrd
|
12:26 |
: Hit ball hard good
|
12:26 |
: what’s for dinner?
|
12:26 |
: Nothing, it’s a fast day
|
12:26 |
: Can you run an alternate Zips where Chris Sale never got hurt?
|
12:26 |
: I can!
|
12:26 |
: And I may at some piont
|
12:27 |
: What does it say about the Braves that their DH has nearly double any other hitter’s fWAR?
|
12:27 |
: Nothing good
|
12:27 |
: I should be all in going after Mason Miller, right?
|
12:27 |
: Yeah
|
12:27 |
: Any reason for some concern about the Orioles? 9-8 in May, the offense is really struggling, and now injuries are starting to crop up.
|
12:27 |
: Not really. Every team is going to have some runs
|
12:27 |
: Why did Mike baumann cost so much? And will the mariners have the answer to make him great?
|
12:27 |
: He writes great headlines
|
12:29 |
: I have just accidentally discovered that Mike Baumann’s player page is sponsored by FanGraphs with fun facts! Are there other such players? Will you tell me who they are?
|
12:29 |
: I was unaware this was a thing
|
12:29 |
: Re: Analytics – it seems like hitting analytics has lagged pitching analytics. Maybe the hitting side is going to get better and do a better job of countering the pitching.
|
12:30 |
: It’s possible, but pitching has an advantage in that it’s not reactive
|
12:30 |
: Does ZIPs look at Roman Anthony first few years and see Andrew Benintendi as possible outcome? Or more like Abreu (Phillies)?
|
12:30 |
: I guess you want a comp? hang on
|
12:31 |
: I’ll get back to this
|
12:31 |
: Do you and or ZiiPS think Vientos is a better hitter than Baty (now and long-term); if so, do whatever numbers you trust suggest that his defense at 3B is too terrible to make up for his offense? What are the odds that he could make a Wilmer Flores-esque transition to 2B if 3B is too much?
|
12:31 |
: I’m actually a bit worried that they’re coming down with Kevin Witt Disease
|
12:32 |
: Not good enough to hit at 1B, not good enough to field 3B
|
12:32 |
: Leading question, but what impact do you think his NPB posting signing deadline had on the whole “Shota is underpaid” thing? Caused by the signing deadline full stop, caused by having a signing deadline while other lefty SPs didn’t, just a scouting miss from the market?
|
12:32 |
: Probably not a huge one
|
12:32 |
: But so far, it’s not looking good
|
12:33 |
: as in the scouting miss
|
12:33 |
: ZiPS frigging LOVED him
|
12:33 |
: and I’m a little annoyed that a few teams bought some really detailed projections for him and then showed no interest
|
12:34 |
: As I expressed then, my feeling was that GMs that led the Cubs get him for that little really screwed up
|
12:34 |
: and I called it — I think, I hope I’m not remembering wrong — the best signing of the offseason
|
12:34 |
: And naturally, 2024 hasn’t done anything to make me feel worse about it
|
12:35 |
: I’d really like to have an Original Tommy’s double chili cheeseburger TBH
|
12:36 |
: I have never been there as I’ve spent very little time in CA
|
12:36 |
: shoudl’ve thought of it when I was in Vegas for winter meetings
|
12:36 |
: Is a lasagna a casserole?
|
12:36 |
: Yes
|
12:37 |
: Red Sox broadcasters are starting to talk up Wilyer Abreu for ROY. Thoughts? I do think he’s better than the prospect analysts thought, particularly defensively, but still mostly a platoon guy, surely?
|
12:37 |
: I mean, it’s not the craziest thing given that he’s like 2nd or 3rd in rookie WAR at least check
|
12:38 |
: and projectiosn already basically had him as a league-average player before the season
|
12:38 |
: like ZiPS was at 1.7 WAR in 474 projected PA
|
12:39 |
: and his time not against lefties is basically already “filled in” with replacement level performance
|
12:39 |
: The Orioles have one of the lowest walk rates in baseball. What are your thoughts on that as to whether it is a strategy or a problem with the young hitters. Adley in particular is walking way less.
|
12:39 |
: They’ve done so much damage when they hit things that they’re incentivized to be aggressive
|
12:40 |
: The purpose of plate discipline isn’t just to get walks, it’s to get your pitch to drive
|
12:40 |
: guys that don’t do that go down the Ben Grieve/Jeremy Hermida route
|
12:40 |
: If I were to pick out 50 ZiPS projections (or aspects of a projection) before the year for 5 years and give you my over/under take on them, what percentage would I have to be right on over that stretch to make you think that I know something that ZiPS doesn’t?
|
12:41 |
: You guys like giving me questions that don’t have straightforward answers! lol. it would depend on the magnitude and what characteristics correlate the errors
|
12:41 |
: and your argument
|
12:41 |
: where would an unopened can of kidney beans fit on this chart?
|
12:41 |
: I don’t want to say, because I’ll naturally put beans in evil and tomatoes in chaotic
|
12:42 |
: LAWFUL GOOD
|
12:42 |
:
|
12:42 |
: What does Zips see from the first two months of the season as a positive future development for the Cards? (slander aside…)
|
12:42 |
: I mean, ZiPS doesn’t think they’re *awful*
|
12:42 |
: they’re just in real danger of getting mired in the Middle Team Muck
|
12:42 |
: Another Royals question, what does ZiPS make of Salvador Perez? I’m sure he’s due for some regression, but how likely is he to finish the season as a well-above average hitter on the whole?
|
12:42 |
: his zstats are chef’s kiss
|
12:43 |
: skeptical on the BA part, but fully endorses the power
|
12:43 |
: Is the Orioles offense coming back down to earth a bit? Obviously there’s Gunnar and Adley, but I’m not sure what to think of Westburg/Cowser
|
12:43 |
: Cowser was likely to come back more
|
12:43 |
: Is someone who voluntarily eats Spam qualified to be a chili gatekeeper?
|
12:43 |
: depends
|
12:44 |
: if you’re eating it straight from the can, no
|
12:44 |
: but fried spam is really good
|
12:44 |
: For players that have consistently outperformed their projections, playing better than their age would normally suggest, does ZiPS adjust to reflect their slower aging curve or does it always project a decline?
|
12:44 |
: I haven’t foudn the guys that do that
|
12:44 |
: Issues of *bias* are far easier to wring out than questions of *accuracy*
|
12:45 |
: The problem with sabermetrics giving the ‘winning strategy’ is that winning hasn’t budged. The average team still loses half its games. The point of any spectator sport is entertainment. Overall, baseball seems to be falling down on that score.
|
12:45 |
: But each team is still going to try and pick the best winning strategy
|
12:45 |
: so you have to change the meta
|
12:46 |
: This entire chat is slanderous…..
|
12:46 |
: Hi Dan! Which team’s success do you think is more sustainable? The Royals or the Guardians?
|
12:46 |
: Guardians
|
12:47 |
: My controversial and correct opinion – The Beatles are not a good band.
|
12:47 |
: In regards to Travis’ issue with the analytics going too far, my mini solution was to just sit and watch a game on mute. Or go to the ball park and watch a game. We don’t NEED you Dan to enjoy the game.
|
12:47 |
: Nobody needs me for anything ever, really
|
12:47 |
: Even the very nice woman my age who puts up with me doesn’t REALLY need me
|
12:48 |
: I know wRC+ isn’t your wOBA/league average wOBA. So what is it?
|
12:48 |
: is kenley jansen a one pitch pitcher now?
|
12:48 |
: Yes
|
12:48 |
: But if you have only one good trick, make it a good one
|
12:48 |
: Has Cody Bellinger done enough this year to convince you that last year’s performance is sustainable?
|
12:48 |
: I have no idea what to make of him ever anymroe
|
12:49 |
: His whole career is batshit random
|
12:50 |
: OH CRAP I FORGOT ROMAN ANTHONY COMPS
|
12:51 |
: Right now, ZiPS has his top copm as Aaron Hicks
|
12:52 |
: Tuffy Rhodes
|
12:52 |
: Dwayne Murphy
|
12:52 |
: Reggie Thomas
|
12:52 |
: Larry Whisenton
|
12:53 |
: How does zips handle someone like Lugo who goes from reliever to starter? Is it as confused as Iam?
|
12:53 |
: ZiPS has long-term data on all moves between
|
12:53 |
: Oh, I should note one thing on an earlier question
|
12:53 |
: While ZiPS doesn’t factor in the history of being right or wrong on a player’s projection *generally* (since that has no predictive value)
|
12:54 |
: on things like FIP relative to ERA over team and expected stats, it will naturally believe in a player’s ability to over or underperform those as time goes on]
|
12:54 |
: So ZiPS would have a different view of Tom Glavine near the end of his career
|
12:57 |
: In early 2017 I remember projections favoring Conforto and Benintendi over Judge because he made so little contact. Was Zips in agreement or did any of them believe in the giant?
|
12:57 |
: after 2016 season
|
12:57 |
:
|
12:57 |
: No hard and fast rule for 3Yr ZiPS usability; does that mean you haven’t done a comparison to season 1 3YR ZiPS to season 2 3YR ZiPS? (Not trying to troll, genuinely curious, and I know you usually check most of your other statistics on a long term basis)
|
12:58 |
: It’s because you’re asking for a very specific answer to a rather non-specific position
|
12:58 |
: Just realized ZIPS never particularly loved Brennen David anyway. Top comp, Carmelo Martinez? Oof.
|
12:59 |
: Carmelo Martinez had a career most players would be very happy with
|
12:59 |
: Also, is a sushi roll a sandwich?
|
12:59 |
: No
|
12:59 |
: Re:Vientos not hitting well enough for 1B: 14 teams currently have a collective wRC+ at or below 100 from the 1B position. Not an endorsement of Vientos, more of an indictment of the position.
|
12:59 |
: Yeah, but you want to replace those guys
|
12:59 |
: We can put this all to rest with a functional linguistic test. A lasagna is not a casserole because no one in history has pointed at a lasagna and said “pass me the casserole”. Likewise no one in history has pointed at a hot dog and said “I’d like that sandwich please”
|
1:00 |
: Nobody has ever said “pass me the alcoholic beverage” to ask to have a beer handed to them, but that doesn’t mean that beer isn’t an alcoholic beverage
|
1:00 |
: Am I wrong for thinking Christopher Morel can become an elite hitter?
|
1:01 |
: Not necessarily, some of the components of an elite hitter are present
|
1:01 |
: though it’s nto the average result
|
1:01 |
: Do you believe in Randy Arozarena or is he a ghost? Maybe there’s a scale, from tooth fairy (low) to evolution (high) for faith that he hits projections?
|
1:01 |
: He’s alive, plus I don’t believe in ghosts
|
1:01 |
: What percentage of your work motivation is your enjoyment of it, vs you need to work to survive, vs you think the product is useful?
|
1:01 |
: I really enjoy my work
|
1:01 |
: and I’d probably do SOME of it for free if I had another career
|
1:02 |
: but I like doing it, I like people giving me money to do it, and I hope that people find it useful
|
1:02 |
: because if my work isn’t useful, it’s weird to be paying me.
|
1:02 |
: because ain’t nobody giving me money for my personality
|
1:02 |
: Why do you hate the legumes, man? Beans have a considerably higher EV than meat
|
1:02 |
: I like beans. But it’s not chili
|
1:03 |
: I enjoy tacos al pastor. I don’t want a tacos al pastor cheesecake
|
1:03 |
: Do you have to slow down while typing to accurately capitalize ‘ZiPS’?
|
1:03 |
: No, but I type everything too quickly in chat so I can answer more questions
|
1:03 |
: So there’s a new lady in the Szym life? That explains your more cheery tone.
|
1:03 |
: No. Am I more cheery?
|
1:03 |
: I’m in Cincinnati for two weeks. I’ve had goetta and Skyline Chili
|
1:03 |
: I’m sorry your trip hasn’t gone well
|
1:04 |
: Re: Analytics sucking the fun out of the game, as a data scientist, it has always seemed to me to come down to asking people to optimize the wrong thing. Like, obviously each team is trying to optimize defensive posistioning to maximize the fraction of time a ball put into play is turned into an out, and have each plate appearance put the mostemphasis on increasing run value. But in a closed-sum game, the total number of wins will stay the same, but the enjoyability of what is fundamentally entertainment changes in a way that is not being accounted for at all. So people fretting that they don’t like the game as much is, to me, totally valid, and something the sport should try to change. And they are, which is good!
|
1:04 |
: I think MLB, if they think it’s a problem, needs to be even more aggressive
|
1:05 |
: big bases? how about *80* feet to first base
|
1:05 |
: and I’d love to see them test out all kinds of stuff in indie leagues, even stuff that’s absolutely idiotic, just to see what happens. because a failure might give them an idea for another thing that might work
|
1:06 |
: what IF the playing field should have a wider arc?
|
1:06 |
: what IF a batter doesn’t have to reach first base with two outs on a grounder for the runner on third to score?
|
1:06 |
: what IF it was four strikes and you’re out?
|
1:07 |
: I don’t know if any of these are good. But someone should at least TRY
|
1:07 |
: (In fact, I dont’ think any of these are good, except maybe the first thing)
|
1:07 |
: the change in base distance
|
1:07 |
: what if it was a cheesecake with some flavor of drizzle in the shape of tacos al pastor on top
|
1:07 |
: It’s not tacos al pastor then
|
1:08 |
: If I eat a chocolate truffle in the shape of a poop, I’m not eating poop
|
1:08 |
: Cat friends cat friends how do they do
|
1:08 |
: They’re being really damn lazy today
|
1:08 |
: Constantine started off chat asking me for dinner (which doesn’t come until 4) but he got tired and dozed off from the effort
|
1:08 |
: Re: analytics and fun. I do think MLB has an enjoyability problem. League average OPS is under .700 right now. The offensive run environment is super depressed. I think we should go back to juiced baseballs and spider tack. 2018-2019 was a really fun time for the league imo
|
1:08 |
: My wife found a recipe for pickle cheesecake in an old recipe book. It was … interesting.
|
1:09 |
: I forgot about Mat Latos! also I’m a couple months older than him but he’s been out of the league 7 years
|
1:09 |
: didn’t you ever see the Cat Latos picture?
|
1:09 |
:
|
1:10 |
: as for pickle cheesecake, well, I have some AI horrors for you…
|
1:10 |
:
|
1:11 |
:
|
1:12 |
: I’d like to make a motion for all the spider tack, all the steroids, juiced balls, and all of the cocaine
|
1:12 |
: cocaine seemed great for baserunning
|
1:12 |
: What does a CJ Abrams extension look like? Presumably at or below the 192/9 you projected for Elly last month?
|
1:12 |
: below, but it’s moving up
|
1:12 |
: I heard about his cat named Cat Latos, don’t remember a picture
|
1:13 |
: Isnt that kind of insulting to the players in indy leagues ?
|
1:13 |
: It doesn’t have to necessarily be there, but it should be SOMEWHERE
|
1:13 |
: How do you choose which questions to answer? Is it because you like some of us more than others?
|
1:13 |
: It’s all chaos
|
1:13 |
: though I try to answer questions I can answer quickly
|
1:13 |
: and that don’t have answers very specific to the other person
|
1:14 |
:
|
1:15 |
: and on that gross note, I will head out for another week
|
1:15 |
: thanks for 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.