Effectively Wild: A FanGraphs Baseball Podcast - Effectively Wild Episode 2532: Can the Public Still Teach Teams?

Episode Date: September 17, 2026

After a brief PSA about an open producer position (email [e-mail] with the subject line “Podcast Editor 2026” to apply) Ben Lindbergh (kind of) continues his quest to temporarily resurrect a few dearly departed baseball podcasts by bringing on MLB.com’s Mike Petriello, former host of Ballpark Dimensions, to banter about a record home run trot, which “Statcast era” records they’re most confident are also all-time records, Junior Caminero’s defense and the value of small-sample defensive splits, the latest Statcast tools, Statcast WAR, predictive vs. descriptive stats, a divisive standing ovation for Pete Alonso at Citi Field, and this week’s White Sox-Guardians series. Then (58:25) Ben chats with former Astros AGM Andrew Ball and former Phillies quantitative analyst Tom Kim about Tom’s new command metric, OpenCommand, whether public baseball research still has something to teach teams, the public-private knowledge gap, and overrated secrets. Audio intro: Philip Bergman, “Effectively Wild Theme” Audio interstitial: The Spaghettis, “Effectively Wild Theme” Audio outro: Justin Peters, “Effectively Wild Theme” Link to Ballpark Dimensions podcast archive Link to Caminero trot Link to Caminero LIDOM trot Link to Langs post Link to Tater Trot Tracker Link to exit-speed contributions Link to Mike’s post on Caminero’s defense Link to Mike on the Williams homer Link to Mike’s last EW appearance Link to 2024 bat tracking Link to 2025 swing tracking Link to Nats farm system article Link to 2026 scoops article Link to scoops leaderboard Link to “I’m tired, boss” meme Link to predicted xwOBACON Link to VAAAA Link to Baseball Savant Link to “last mile” wiki Link to standing Alons-O article Link to team 1B WAR Link to Piazza ovation Link to Seaver ovation Link to Pujols ovation Link to Pham story Link to White Sox-Guardians gamer Link to Peters PA article 1 Link to Peters PA article 2 Link to Davy on Jack Reacher Link to Andrew’s baseball bio Link to The Athletic profile on Andrew Link to Tom’s OpenCommand thread Link to OpenCommand account Link to OpenCommand github Link to The Athletic on OpenCommand Link to OpenCommand validation Link to OpenCommand accuracy Link to computer vision article 1 Link to computer vision article 1 Link to pitch-tipping detection info Link to Tom on miss distance Link to Tango on miss distance Link to Sam on misses Link to Ben on COMMANDf/x Link to The Athletic on Command+ Link to “Stuff” model command stats Link to BP on command 1 Link to BP on command 2 Link to team targets article Link to team targets graph Link to command value post Link to Tom’s post on baseball analytics Link to Andrew’s Substack Link to Andrew on public baseball analysis Link to Andrew on idea value Link to Ben on baseball secrets Link to Andrew on idea value Link to Andrew on draft performance Link to Hustle+ paper Link to Ben on hustle Link to Andrew’s podcast post Link to Andrew’s podcast Link to Marvel’s Wolverine clip 1 Link to Ben’s first Wolverine pod Link to Ben’s second Wolverine pod Link to previous Winquest banter Link to Gasper story Link to Paine on position-player pitchers  Sponsor Us on Patreon  Give a Gift Subscription  Email Us: [e-mail]  Effectively Wild Subreddit  Effectively Wild Wiki  Apple Podcasts Feed   Spotify Feed  YouTube Playlist  Facebook Group  Bluesky Account  Twitter Account  Get Our Merch! var SERVER_DATA = Object.assign(SERVER_DATA || {}); Source

Hey everyone, Ben here. Before we get going, a quick announcement, we are hiring. Sadly for us, Shane McKeon is moving on after three and a half years and more than 500 episodes of editing and production assistance. We are very grateful for his excellent work. We will miss him, but despite Shane's impending departure, the show must go on. And so we are searching for Shane's successor, who will be the third ever regular producer in effectively wild history. Not counting me. No particular prerequisites, but we are looking for. for someone who, preferably, has some substantial professional podcast production experience. Primarily working with audio, we have no plans to transition the podcast to video.

But video production experience is a plus just in case. You don't have to be the biggest ball knower in the world or to have listened to Effectively Wild since 2012. But some familiarity with baseball and the show would also be appreciated. We're hoping to find someone who will be conscientious and diligent and who will handle the show as if it's their own. which in some way it will be, someone who's passionate about ensuring the best possible listener experience, and availability and flexibility are both pretty important. As you know, we do three meaty episodes a week, rain or shine, summer or winter, so we need someone who has the bandwidth for that and can allow for some variation in recording days and times from week to week. We record most often in the early

to mid-afternoon Eastern time, so ideally our producer would be available a little later in the day, so we can turn around those recordings more quickly. If all of that describes you, then we want to hear from you. You can email us at wanted at fangraphs.com. Just send us your resume with some sort of preferably non-AI-generated cover letter or accompanying message. We want to move fairly quickly here, so if you're interested, don't dally. Again, wanted at fangraphs.com subject line podcast editor 2026. Thanks, and we look forward to hearing from you.

And now you can hear some more from me. Did Richard love lady have a striker, Taylor Teagot, and who had more, or Jason Kendall or Russell Martin, what if Show Here Tani's dog was also a good lawyer? What would you do if Mike Drunchers showed up in your foyer, or is it foyer? Find out Uneffectively Wild. Find out Uneffectively Wild. Find out on Effectively Wild today. Hello and welcome to episode 2532 of Effectively Wild, a baseball podcast from Fangraphs presented by our Patreon supporters. I am Ben Lindberg of the Ringer, not joined today by Meg Rally of

Fangraphs, who remains on vacation and will for one more episode after this before rejoining us next week. And so I have brought in substitutes for Meg. Later on this episode, I will be talking to a couple of former front office executives. Andrew Ball and Tom Kim will be talking about Tom's research into Open Command, and we'll also be getting into the public-private research divide and how much more teams know than we out here in the public potentially. But first, I wanted to talk to someone who has helped bridge that gap by sometimes releasing some nifty statcast stuff for public consumption. He is, of course, Mike Petriello of MLB.com. Welcome back, Mike. Welcome, Ben. Welcome to me, I guess. Thank you for having me. You can welcome me too. I feel welcomes to my own show.

Great start. Fantastic start. Hey, do you know how I know that Meg was on vacation and wouldn't be joining us? Because Davy Andrews, is one of my closest friends, got to write an entire piece about the show Reacher on Fanggras.com today. So that's how you know, like, the bosses are away. The kids will play. Meg, let Davy get away with some stuff on her watch too. That's just, you know, let Davy be Davy. I think it's just good editing in general. But this was initially supposed to fit into the rubric of the series. I've been doing while Meg has been away, where I've been reviving former baseball podcasts. And so I approached you and your former co-host, Matt Myers, of the Ballpark Dimensions podcast, because I was hoping to bring you back as a duo.

But it turned out that Matt, who is still editing and working at MLB.com, he has hung up his mic for good. I don't know if you hang up a mic or whatever you do to a mic when it's no longer in use. his podcasting career is behind him. And so he politely declined the invitation. And so you are here as a solo act. I guess it's just ballpark dimension. There's just we're one dimensional today. Yeah. And I can tell you this. For the next three days, I sit about three feet from Empire. As I have for the last decade, they are rearranging all our desks soon. So that that partnership will end. Obviously, we'll still work together. But yes, he was very grateful to get the request from you.

but he is much happy to be more of a behind-the-scenes guy right now. Well, you can carry on the legacy of ballpark dimensions today, and we can get into some stat-casty stuff, and we can talk about a few current events. One thing that maybe can be our jumping off point here, I sent you a message late last night. We're recording on Wednesday afternoon, which was prompted by yet another Junior Camerro walk-off home run.

So last time, Josie and Randy Gisarely were here, and we were talking about how maybe Yordon Alvarez is in the process of fumbling the ALMVP bag that he appeared to be in possession of. And Canamero is making a late charge here. And it helps perhaps that, well, the rays have clinched. The Astros status is still uncertain. Yordan has sort of been slumping, though he did Homer on Tuesday, but only kept pace with Camerero because Camero hit his second walkoff Homer in four games. and he took his time appreciating that he had done that. So, Yordon hit a game-winning dinger.

Camoneros was a walk-off, and he waltzed around the bases for roughly 40 seconds. I clocked it at 40.3. The official stat-cast trot time, you told me, was 40.7 seconds. And I put it to you, was this the longest trot on record? And initially, you had some doubts, but then you were able to confirm that seemingly it is with a couple of caveats. Yeah, so a couple of things are true at the same time. So the official number, I guess we go to two decimal points, 40.65, which is a credit to you.

I mean, that's pretty close. I don't know if you were hand timing or some. I was, yeah, I had my finger on the stopwatch. My scout school skills coming in the end. Pretty good, right? Like, I don't know, it makes me think of, if you remember Larry Granillo from back in the day. Yeah, to Tater Track Tracker. Exactly right.

Yeah, this was, yeah, this was from the moment of contact to plate touch. So that's how we're tracking the tape. Yes. So here's the truth. When you message me about this last night and I was probably mildly dismissive that it could be the longest one. That is totally on me because what was happening when you sent me that is I was half watching the Cleveland game, I think, while playing my guitar. And you sent me that note. I wasn't watching the race game. I had no idea what had happened. And I was like, no, there's got to be longer than that. Yeah. Because I'm sure you constantly get people saying, was that the longest? And then you find out

that it's not nearly that. That's what happened. to us with our stop last segment is people will write in and it's like, oh, this has to be unprecedented. And it turns out that no, everything is precedented in baseball. Right. And as we talked about a little bit, there's a lot of nuance to this because it's not just like, you know, you could look up the hardest tip ball of the year in about five seconds. That's, that's easy, right? But I pull up a list like this and there's a lot of nuance because there's a lot of stupidity that happens at the top here, right? There's so many of them where it's, well, we didn't know if it was a homer or if it was going to be a double and the runner stopped at second

based waiting for the umpire to make a signal or pulled his hamstring going down the line all sorts of stuff. So I message, I've a long-running group chat with Sarah Lange's and Jason Bernard, and Sarah took it upon herself to actually go through and look and exclude all of the very weird ones. And it does turn out, this is number one. And that's, you were right. You were right. And I was wrong, and I'm going to give you full credit on that. Well, thank you. Yeah, I have a good inherent sense of how long a really long run trot is. Because he took his time. I mean, this was, was standing in the vicinity of home plate, admiring this thing for quite some time, and then not exactly getting into gear and picking up the pace that much once he actually

did start driving. But the game was over. Like, it's okay. The game was done. Right. Now, the thing is that by Caminero standards, this actually wasn't that long because he famously, I think we discussed on the podcast a while back, he had a homer in Lidome where he was clocked at 56 seconds rounding the base.

And it was like he was. stopping at every base to celebrate. He was hugging people. He was, like, doing all kinds of gesticulations. I mean, it was the most roundabout route. It was, like, you know, in the cartoon family circus when you see, like, the path that the kids took around the neighborhood. And it's, like, really convoluted.

That was basically his path around the bases. And that wasn't even a walk-off, if I recall correctly. That was, like, a go-ahead game winner, but it didn't actually end the game. And so he was taking his sweet time. But, you know, I guess the lead-on. atmosphere, it's maybe a bit more permissive to ostentatious celebration than Major League Baseball has been historically. So by MLB standards, I mean, 40 seconds, if we were to do some kind of like MLE, you know, equivalency translation here, I think 56 seconds in Lidom probably maps on pretty

closely to 40 seconds in MLB. So I'm excited to see. Can he push this bar even higher? I didn't see whether there was any backlash or dirty looks or anything. that he was getting from the A's as they were waiting for him to complete the trot. But I mean, you know, look, you got to let him celebrate. It was a nice hit. Yeah, I wonder, I could probably go back and find this out. How many A's were still on the field after the 45 minutes it took him to round the bases? Probably can't be that many to have been that mad?

And can I tell you this too real quick, non-sequitur? This is somehow the second time in the last five hours that family circus has come up in a baseball-related context. Yes. Because I was looking at Sam Antanachi's round. trying to catch Brian Rokio's walkoff in the Cleveland Chicago game. And it's kind of the same idea with like the dotted lines and everything. I can tell you this, it was indirect.

It was, I can't remember the kids. It was more Billy than Tommy or whatever. Well, speaking of defense, when we brought up Camerro on that recent episode, I denigrated his. And I saw that you actually posted shortly before we started recording that maybe I need to update my priors when it comes to junior Camerro's defense. because according to fielding run value, he's actually been pretty darn decent lately. So the first two months was awful.

I think what I said, minus eight fielding run value or something like that. And essentially even, or maybe plus one since June 1, which is the second year in a row this has happened. Because I remember filling in for Adam Barry's newsletter late last year, and I wrote essentially the exact same story that got off to a bad start. Now, is that like the Julio Rodriguez of defenses? I don't know if that's a thing. There's maybe something to be said about two giant air quotes that you can't see here. New ballparks in a row for him. Yeah.

That could be something. But anytime we're talking about defensive stuff in a somewhat small sample, I want to go eyeball it and see how confident I am in it. And I can tell you this, I'm 100% confident that he was really bad for the first two months. Yeah. And throwing errors and fielding errors and misplays that weren't marked as errors. I have no compunction in saying the number in the first two months feels completely right to me. And that I didn't see as many things going down the stretch. over the last couple months. He has played solid to relatively even. And I think that's cool. And if they

can just figure out how to make this happen from day one next year, they might have a superstar. I mean, he's a superstar. They might have an all-around like superstar player. Because I think the skill is there, just kind of the mistakes that were happening in the first two months got evened out. Yeah. And you kind of worry if someone's struggling on defense that much at 23, then how far away is DH from his future? Not that he couldn't be a valuable DH with the kind of hitting he's doing. But obviously if he's playable in the field, then he's way more valuable, and he has certainly been playable at least of late. Where are you with looking at defensive splits and looking at samples of a month or two? Because in the past, of course, that used to be the third rail,

really, and we would talk about, oh, you need a full season, you need three seasons sometimes of defensive stats to get a sample that would be meaningful. Now we have statcast and we have FRV, and so there's more precision and granularity there, but there is still the question of opportunity and just what kind of batted balls come your way and all. So, I mean, I think it's good that you combined it with the eye test and then went to, you know, trust but verify. But where are you in terms of trusting, say, monthly splits in statcast defense? I remember, I don't know, by the way, if people know this, you can actually look at it on a daily basis. I'm not recommending you should, but you can if you want to. It's there. But I remember. But I remember. I

remember when we first put up the monthly splits a couple years ago and I was kind of the one who was advocating for it and there was obviously some hesitation like oh hey you know defensive metrics are they going to be useful for that and I think it's turned out to be actually pretty good because I remember writing this year on April 8th 10th something like that just looking at the first 10 days and saying okay here are the teams who have rated well in defense in the first 10 days and the teams have rated poorly and it's stuck pretty well you know so your point about opportunity is obviously key and no amount of granularity and defensive metrics will ever overcome the fact that you can only make the plays that are presented to you. You're not going to get your four plate appearances a game. You're not

going to get your six innings of a start, whatever. That can't really ever change. And while I'm certainly not going to sit here and say the current metrics are absolutely perfect and can't be questioned, I do think the fact that the technology keeps improving means you can start to get to some meaning pretty quickly. I'm guessing, I don't have it in front of me if I were to go back and just look at April, who are the best defensive players in baseball? I'm guessing PCA is probably pretty high up the list, right? Yeah. And I can give you an example because I just looked this up the other day.

We are sitting here today, I think, almost exactly one month since Stansby-Swanson got hurt for the Cups. And just looking at infield as a whole, through his injury, they were number two in baseball, since his injury number 19 in baseball. Is that a small sample thing? Maybe. Is it the effect of he's a really good shortstop? And Nico Horner was a really good second baseman, and he's a somewhat lesser shortstop and now not playing second base? Maybe it's both.

But I do think there is at least some signal there. Yeah, I was wondering about this. I didn't prep you for this, but a thought experiment because we always have to caveat with stat cast era or whatever you call it and, you know, since 2015 or maybe even 2016 in some cases. And that sample's getting bigger and bigger, obviously, but we will never have this sort of detail for the majority of baseball history, at least until centuries pass. So I wonder which stats you're more or less confident that when you say this is a statcast era record, you're also sort of saying this is an all-time record? Because I was thinking of this in Caminero's case, because, okay, we have to caveat and say this is

the longest home run trot tracked by statcast. But would there have been a longer one prior to 2015 in that era of baseball? I mean, maybe in the, you know, just like unsure whether it was a homer or someone gets hurt category, but the legitimately taking time to enjoy it, that's the sort a thing that you would have been more likely to get knocked down for doing in earlier eras. So I'm sort of skeptical that there would have been a legitimate time much longer than that. So I wonder like all the various categories because there's so many stat cast stats now. And I guess, you know, this comes into play with like PCA having a historic stat cast season for outfielders or, you know, now has the, what is it, the highest ever career run value for outfielders.

But of course, that's, you know, not entire careers for a lot of. previously talented outfielders. So that kind of stad, or when we're talking about individual bat at ball events or pitches, if you want to go back to pitch tracking era, I don't want to touch you off here by, you know, invoking all of the velocity wars that you get into on social media. Somehow you're always on the front lines of those exchanges about people who think that everyone threw as hard or harder in earlier eras. But, you know, I'm fairly confident, I guess, that the fastest tracked pitches dating back to 2008, probably that encompasses the fastest pitches ever, even if there were outliers like Nolan Ryan or whoever.

Hard for me to believe that they were muscling it up to 105 or whatever just because of the way pitch usage and pitcher usage was back then and short bursts and everything in this era. So, yeah, like looking at the span of stat cast stats, which ones are you most and least confident that, you know, what, like, we still have to apply the fine print, but I almost feel like I don't need it. Yeah, that's a great question. I think I would break the stats into two categories, measurables and comparables, right? So, measurables are this pitch was thrown this fast, you know, you hit the ball this hard, whatever. And comparables in the PCA example, everything is like compared to average. So it does sort of matter what is happening at the time you're playing. I think it was probably Jeff Sullivan, who wrote this like 10 years ago about

catcher framing, where it's like, you know, the average change. so much once Ryan Dume it stopped being able to play catcher, and that affected everyone else's grades. You know what I mean? So it's like, could I make the case that PCA is the greatest defensive center fielder of all time? Probably, but if we had these numbers for all of history, is there a case that, you know, Paul Blair or William Mays or somebody against a presumably lower average and with more balls in play could pile up more value? That's probably pretty likely, I would think, right? In terms of measurable, though, I'll give you a great example. I think that O'Neill Cruz probably hit the ball harder

than anybody who's ever lived. And I don't think that makes him the best hitter because he's obviously not. I know people want to say, oh, you know, Barry Bonds. Well, Barry Bonds was a god, but it was more for the eye, really, than like hitting the ball harder

than anybody on the planet. So that's probably the one I would point to is, I don't even remember what the record is. It's like 123 or whatever he's up to now. That to me, I cannot really conceive that anyone has hit a ball harder than that, even if it doesn't make him a superstar. Yeah.

Yeah, I was wondering about the exit speed specifically because it is hard to think like, you know, hardly anyone has been bigger or stronger than Cruz or Judge or Stanton or Otani or this current generation of hard hitters. You know, and I'm sure people who saw like Frank Howard play or Harmon Killebrew or, you know, we grew up watching Gary Sheffield, right? Like if you told me that Gary Sheffield was hitting the ball 120 at times, like, would that blow my mind? I mean, I always thought was Sheffield because he hit the ball so hard and often would just rip these balls foul.

And I believe that like when he aged and his bat speed slowed a little bit, he would actually be even better somehow because like all the balls that he was pulling foul would be fair. I don't think it quite worked out that way. Although, you know, he was a pretty productive older hitter too. But it's just, it's hard to say because that was, I mean, relative to his era, he was hitting the ball hard as anyone is today. but the pitches were probably a little slower then, and the pitch speed does contribute a little bit to exit speed. And, of course, like, the ball construction has fluctuated, and so that affects the coefficient of restitution,

which can affect batted ball speed. So that stuff is inconsistent, too. I don't know. It's, like, obviously, the league-wide numbers have changed, and there's so much more of an emphasis on hitting the ball hard, and, like, once everyone knows that bat speed is something desirable, then people really train for that. and everyone's angling to hit barrels and stuff now.

And so there's more appreciation for that and there's strength training. And it's hard to imagine that, you know, Babe Ruth or whoever or Josh Gibson or someone or, you know, like there were players who were like built, you know, back in the day. And you see photos of like Willie Mays shirtless and you think he's like a time traveler or something because he looks like he could be a current modern day athlete. But I don't know. It's just it's hard for me to imagine that. the current heavy hitters could be topped in that category.

Yeah, and I'm glad you brought up the pitch speed aspect, too, because it's not quite hard in, hard out in the way that people wanted to believe, but it's not zero. It's the math is, what, 80% batter, 20% pitcher or thereabouts? So it's like, there's something to that. And the reason I look at Cruz is because the hardest hitting guys other than him are Stanton and Judge, basically. And we don't know all the metrics and measurables of everybody ever, but as great as Babe Ruth was, as great as, you know, Ted Williams was,

they were not quite built like these guys, you know? And here you have O'Neill Cruz who is measurably hitting it harder and repeatedly, not just like a one-off, not just a, oh, maybe it was a mystery. Like, he's done it a couple of times now. It's hard for me to believe that anyone has ever hit the ball harder than that. Yeah. And we're all subject to recency bias and presentism and thinking that our time is special and unique. And I guess the counter would be, well, there was so much more baseball before statcast

that the sample is way bigger and so if you had some outliers over the first 150 years or whatever of Major League Baseball then maybe even if the league-wide average was a lot lower as I'm sure it was then maybe like the true standouts

still there would have been more time for them to accrue higher numbers but these are just sort of you know one-time events like it's you know you don't even need a big sample to say does this guy throw hard or does he hit hard, it's something you can often tell with one pitched or batted ball. So hard for me to say.

Other things I'm thinking of like speed, for instance. I mean, you know, when you had A's pinch runners who were actual sprinters or, you know, like, were those guys faster than today's 30 feet per second sprinters? I don't know because I do think we can get a little over our skis when it comes to pretending that like every previous generation of players was just, you know, out of shape and riddled with dysentery or something like that. That wasn't totally true. I mean, yes, okay. Like there was less money.

There was less training. There was less incentive. The player pool was a lot smaller. So certainly on the whole, the talent level was a lot lower. But still, like, there were incredibly athletic people who were around, you know, a century ago. It was not like the human race has dramatically evolved since then. And so I don't want to totally dump on, you know, the first 150 years of baseball and say, oh, there were no athletes who could compare to today's. No, of course not. Where I am on that is, you know, people will be like, if Ted Williams played today, he'd get killed. And I don't buy that. I think the cream of the crop from any era could still play. Ted Williams would be fine today. You know, Bob Gibson, he could play today. It's really the median guy from back of the day. You know, 1950 American League MVP, Phil Rizzuto probably doesn't get past double A today, right? That's that is the kind of guy who wouldn't be around.

Wow, shots fired at Scooter. That's tough. Yeah, come get me. He can't, I guess, sadly, RIP. But yeah, I guess I'm with you. And of course, the debate always becomes, well, are we talking about you just got dropped into the time travel machine and you're coming in cold? Or are we talking about Babe Ruth or Ted Williams born today with all the same training advantages and technology and tools and everything? That's a very different conversation. So can I say to bring it back to the beginning here? probably the most notable moment we ever had on the podcast we did was I was the one who had Adam Ottavito on when he said he would that's right ruin baybruth every time yeah which is which wasn't even the point and I felt really bad because that like totally blew up and I apologized to him afterwards and he's like well I said it I'm like yeah you did yeah no and he's probably right but again it's because babe ruth would just be so overwhelmed by seeing modern stuff but you know if somehow Babe Ruth were were raised in today's world and game

maybe he would still have the skills to hang. And also, one of my favorite pieces you did a couple years ago was your investigation of Ted Williams's red seat Homer at Fenway and whether it really went 502 feet because we're accustomed to all these, oh, yeah, that Homer went 600 feet or something. And now we have data that suggests that homeruns don't actually go that far. And so these are tall tales and exaggerations. But then you looked into it and found that, no, it probably actually did go that far, if not farther. So sometimes the reputed feats of yesteryear, they actually do stand up to scrutiny. Yeah, no, I had a lot of fun doing that. A lot of that was built on the work of Clay Nottily, who's one of our data scientists here.

And without spoiling it, I can tell you that like a squirrel hoarding away nuts for the winter, I have been trying to come up with things I might write about over this winter. And I have another one of those in mind. And I'm not going to tell you which one it is, but I kind of want to repeat the same exercise. Because it's one that comes up a lot. And it's one I think I could actually, like, do a little bit of research on. So hopefully we can get there. All right.

That's something to look forward to. Well, maybe we could do a little stat cast tool catch up because you haven't been on the show for a couple years. And the last time you were was, I think, episode 2164. And this was just when the bat cast stats debuted, the first generation of them. And a lot has changed since then. So that was sort of like bat speed and squared up rate and blasts. and swords and some stuff that probably you've deprioritized since then and fast swing rate and such. And since then, you've expanded on that with new swing metrics, swing path, and attack angle, etc.

And then you also rolled out scoop stats for first baseman this year, which was one of the remaining pieces of the defensive puzzle that hadn't yet been incorporated into stackcast. So, yeah, give us the quick roundup of the last couple years of stack cast progress. That's probably a lot to tackle, but what's been most exciting or revealing for you? Yeah, the bat tracking stuff is number one that came online in 2023. So we've had a couple of good years of fun stuff with that. Earlier this year, we did swing timing, right? So if you're late on fastballs or early on fastballs,

and a lot of the bat stuff has been specific to hitters, but I really like that one for pitchers because Tyler Rogers, you know, who throws like 82 from underneath the surface of the earth, was making badgers late on his fastball more than any. anybody else, which I thought was just the coolest thing. You'd expect it to be like Mason Miller. And no, this kind of gets to deception, which is a thing we've been talking about forever. And no one's like truly, really cracked. I love deception studies. That's my favorite. Exactly. So that was cool. And then also, you know, missed distance, right? Like, who misses the

bats by the most? And well, that is Mason Miller. That's cool. It doesn't have to be necessarily a pure value thing. Like, that's a scouting tool, you know, like, okay, well, I'm worried about his curveball because he's missing by half an inch. Not worried about that. as curveball because he's missing by six inches. Like that's cool. And it's visual and it's easy to understand in a way that a lot of things aren't necessarily. So yeah, that was our big thing. And then the first baseman scoops, because that's just something we get asked about a lot. ABS, obviously huge deal for this year.

So that was a lot of our winner last year. You know, we always have things in the pipeline, probably nothing new coming up for the end of the World Series. We're working on improving our minor league player pages because they're pretty crappy right now. But there's a lot of good minor league data. and people have prospects. And so we're getting to that. Hoping this is like the least analytical thing, but I find it entertaining, and I hope you will too.

Hoping to launch like a team uniform tracker thing, like how is our players, you know, or not players performing, but like how are the teams doing and the city connects, this or that. And then my white whale I've been thinking about for a while and I'm hopeful we'll actually get there. We don't have pitcher time to the plate. I want to know if a guy is like one three to the plate,

one six to the plate, because players think about it this way. I think it was actually Bowman, who wrote the other day, speaking to Bryson's stop, where he's like, yeah, it's a math problem. If, you know, pitcher plus catcher is under 3-3 or whatever, I'm not going to go. We should be able to show that. Yeah. Why don't you have that? I'm not getting on your case.

I'm just wondering why that wasn't earlier in the roadmap because it seems like something that compared to some of the stuff you all have done, maybe wouldn't have been quite as complex. And I guess it probably factors into you have all sorts of base running value stats. and catcher base stealing prevention stats and pitcher and such. So that seems like an important component of that. Fantastic question. When does the clock start? Yeah, that's a good point, I guess. And the reason for that is up through 2019,

StackS was powered by Trackman, which is really, really good, but it was center of mass, didn't really have, like, limb tracking and everything. And so we tried to do it then and looked into it, and you'd get some really weird clock is starting. stuff. Like, if you remember the way Kershaw pitched, it like raises hands to the sky and then bring it back down. And it's not when the clock should start, right? Now with the Hawkeye can help us do limb tracking, it's going to be, I think, hopefully based a little bit more on, well, you know, the front leg is up to its highest point or whatever it's going to end up being. Because it just

wasn't reliable before because you're right. We thought about this like a decade ago and it just didn't really support doing it cleanly enough. And for the bat tracking stuff and the swing trajectories and all that. It's cool, of course, to have another way to say that Mason Miller is good and just to show that. Obviously, we all knew that, but it's fun when the tracking stats reinforce our perception and the results. Have you seen or done any interesting analysis? I mean, maybe this sort of stuff would be most useful when it comes to projections and player development and looking at minor leaguers and all the rest of it. Or have you seen really interesting, useful applications of it when it comes to big leaguers? And I don't know, studying slumps and hot streaks and what's going on there? Or identifying maybe pitches that are working well or aren't or changes mechanical tweaks that hitters should make that kind of thing.

Yeah, I think there's a lot of untapped value there. I've seen some commentary in the baseball world recently where it's like, well, there's no more public-facing research, right? We're going to talk about that on the second segment of this episode. Yeah. And it is fair. Like all of the great people get snapped up and hired quickly. So that's like completely fair. But it's also like there's so much new data that comes out all the time.

And there's a lot of questions and ideas like that that I'm almost surprised people haven't tried to get a little bit more deeply into. Because I'd like to do some of that myself as well. But I'm only one person and writing is maybe 20% of my job these days. You know what I mean? So there's a lot of the data is there. And there is so much more unexplored territory that people could really get to. Yeah. Do you get a sense from interfacing with teams, as I assume you do regularly, of what they're working on or how they're using this information, which in most cases they had before the public or they had some form of it before it was widely rolled out?

So are you privy to, oh, teams are doing this, but this isn't happening so much in the public sphere. And if so, what is it? Spill the beans. You will be shocked to know that they are pretty buttoned up about these things. You are a public-facing figure after all. Yeah, well, I said, I have friends, you have friends. But for the most part, they are not going to, they're not going to get in too great detail about what they're trying to do, especially if they think that nobody else is actually doing it. Yeah, yeah, unfortunate for all of us, I guess.

And Tango has talked, your colleague, Tom Tango, has talked at times about a stat cast war that this is maybe all building up to and will eventually culminate in. is there anything you can divulge or tease? Because I assume that's the end game at some point. And you want to make sure that you quantify all these components and not roll it out prematurely. But it seems like that would be the natural logical extension of, hey, we're quantifying all these individual components of player value. At some point, we'll probably put them together. Yeah, this comes up a lot. Brad Doolittle wrote about this, the summer pass and wrote about it a couple years ago.

And I have found it interesting that everybody seems to think that since day one, like, that's the goal. That's where we have to get to. And will it happen at some point someday? Probably. Really, a couple of reasons it hasn't happened yet is I can't imagine it'll be that different from what Fancrafts currently has since they are currently using statcast defensive metrics anyway, you know? And obviously using some of Tom's own work to do the war framework. So there's at least some question about, hey, is it useful to have a third or fourth version of war out there that it's like, 98% consistent with one of the other wars. So there's that. But it's ever really been the main goal. You know, like what's interesting to me is quantifying stuff that hasn't really been well quantified before. The first base been receiving things, like a perfect example of this. Is it like the most valuable game-changing thing in the world?

No, I think people would like it to be, but it's not. But it's a cool thing that tells the stories with, you know, and it's something that you've been asking about but never being able to get to. And could this all be incorporated into a value metric? like war, sure, will we do it eventually, probably? But it just never ends up being the highest priority, kind of for the reasons, I already said. Yeah, the scooping turns out to be really complicated for what seems like maybe on the surface. It should be sort of simple. And there have been previous accounts of this, and I guess, you know, SIS has done some scooping stuff. But

this one, which is stack has space, it turns out you have to account for a whole lot of factors because you have to weigh whether this was a good throw and how likely it was that this would be received. And then where is the base runner? And once this came out, I don't know if you've made further refinements or will be, but I know when people were cherry-picking individual plays, there are some that look a little off

where it's like, oh, this is kind of harsh because, boy, like, I don't know how many people would have made that play or even if he had made that play, the runner would have been safe. And I think Tango said something to the effect of, well, it works out overall, but if you're looking at individual plays, it might not always stand up to scrutiny. So I don't know if you're still fine-tuning that or want to talk a little bit about what went into that and what made it challenging. Yeah, I mean, almost always with any of these metrics, you know, you'll fine-tune it internally as best as you can and find all the bugs you can. And then you put it out. And then inevitably, like, 10,000 people will look at it in the next hour and somebody will show you something, right? Yeah. So will there be an update to the model over the winter? Yes, probably there will be. Kind of the issue with some of these metrics, and this is one of them, is that they're not intended always to match the eye test in a way where if you watch a pitch and I say it's 100, it's not going to be a looping curveball. You know what I mean? Like a pitch is going to be that velocity and it's never going to look different really. And for this, you'll have stuff where it's like, well, that was a really high throw. How are you saying the first baseman should have caught it? And it's like, well, that first baseman was five foot eight. And he was metal. Wilson, it's a pretty easy play and height's kind of a skill. So it looks weird. The reason the runner

position is built into it is because if the runner is 40 feet down the line, well, in the first basement's got plenty of time to step off the bag and get back to it. There are some I'm not going to defend that they looked really janky, 100% with you. Some of that is just data quality where the data is not 100% reflecting what I think actually happened. So that's always a bit of an issue. And some of it's just model refinements like, oh, we're not accounting for a balance as well as we should be or something like that. And I think what's interesting about not just first base metrics, but really any of the defensive metrics is we try to be pretty open and transparent

about showing a lot of the plays and the values attached to them. Right. And I think that's generally a good thing. You want to be able to explain to people why stuff is happening and why it's not. And what we've learned repeatedly over the years is that bites us a lot because you'll find like one play and you don't like the measurement that is attached to it and say, well, that means everything's broken. even though it's like one play of a thousand.

And not to denigrate any other metrics because they all do great work, nobody else really presents the information in this way. So you can't even compare to that. So then people just assume that the other one, air quotes, did it correctly, which it's just a tricky situation. Yeah. And there's always a tension. We may have talked about this on a previous appearance with you,

but between the big picture full season stats and then the individual play where sometimes you want something that's going to be predictive on the whole. and it's not as descriptive of that specific play. And then sometimes that leads to confusion, which is partially user error sometimes, but also maybe the way it was presented, something like expected batting average, right? Which I've been somewhat frustrated at times because, you know, if you're talking about, let's say there was a no-hit attempt or a no-hitter or something, and you want to talk about, oh, there were X number of plays that had an expected batting average of 500 or something,

and they were really likely to be hits, but they weren't in this game. But then you can get into trouble because expected batting average typically is not taking into account. The spray angle, the horizontal angle, like the hit direction. It's more the hit hardness and the vertical launch angle. And on the whole, in a big sample, that's all you need, it turns out. And evidently it doesn't actually give you any added precision to add the hit direction. But if you're trying to tell the story of a specific game or a specific play, then that can kind of mislead you because a ball that had a high expected batting average might have been hit right at someone or the opposite of that. So I don't know what the evolution of your ongoing thinking about how to present and communicate those concepts. What's the current thinking? Yeah, we think about it a lot. Yeah. There's a couple different ways you can go about it. So we do have versions of all of these things that account for other inputs. Right. So expected batting average or hit probability or whatever.

we have a version that would include for Spray angle. We have a version that would include where the defender is actually standing, a version that would include for the way the ball carries and all the stuff. Because when you're looking at out probability for a fielder, it does. It does account for that. So we've talked at points about trying to put up a tool that would allow the user to pick those inputs that they want, and we haven't actually done it. And we may yet do it.

We just haven't prioritized it. But then there's the other aspect of that is, well, now we're going to give people, six different numbers for every batted ball. And is that just going to confuse the hell out of everybody? And someone's going to say, hey, that was a 48% chance in that play. And then you say, no, it was 22% because you're actually looking at different inputs. And eventually you try to incorporate so many things that at the end of the day, all you're doing is just describing what actually happened.

Right? So every hits 100% and every out to 0%. And that's not what you want either. So we do talk about that a lot. It gets confusing. One thing we did talk about like a year ago, we'd gotten to the point with the defensive metrics where if we wanted to. So right now, hit probability XBA and catch probability don't add up to 100%. And that confuses everybody all the time. And I totally get why, right? Because fielder position and

home runs. That's the big difference there. And we talked about, well, we have the ability with the metrics to change that if we want. So every time there's like a 70% catch probability, then it was a 30% hit probability. And that would have just required so much reeducation and changing like every number, it did not feel like it was going to be worth the effort. So I would say some of the things, if we are starting over from scratch today, 10 years later, would we have done a little bit differently? Probably. And as Tom is fond of saying, if we were all working for a team and the goal was to win, would we do things a little bit differently? Also probably, you know, but there's so many different things we want to get out there, so many tools we're trying to make. Because the goal of this is not

just to have a cool website, right? The goal of this is to support, you know, mopb.com writers, obviously, but broadcasters, the teams themselves, I know that they use this data for a lot of different stuff. And as the saying goes, which I will probably butcher some way, the last 10% is 90% of the effort more or less. Oh, yeah. The last mile problem. Exactly right. And it's like, not that that is not valuable. Of course it is. But if you're going to spend like another two months getting to that last like 5%, and that means that you can't put out other cool stuff, is that really worth the trade off? And that's kind of where we're always thinking about. Yeah. And there's a an additional layer of abstraction with some cutting edge stats now that even me, someone who people

would describe as a stathead, I feel like the Green Mile meme, I'm tired boss, when I see some of this stuff because my head is spinning because it's like a couple layers removed from the results. And really the history of Sabremetrics, I guess, has been sort of, you know, moving more and more toward the process as opposed to the results side of the scale. And so nowadays sometimes I'm seeing X-X-Wobacan. that kind of thing, where it's like not ex-Wobicon, like, expected, weighted on base, on contact, but expected, expected Woba on contact. It's like predicted ex-Wobicon, which is based on like swing decisions, basically.

It's like, how hard should you have hit it, given what you swung at, you know, using some of the swing metrics that we were using. It's just, it's really abstract. It's like with the stuff models, you're not looking at how the pitch actually fared. you're looking at how it should have fared, in theory, based on the characteristics of the pitch. And now we're getting there with swing decisions and, like, forget about the swing itself, but, like, even deciding to swing. And pretty soon it'll be, you know, people wired up with electrodes and stuff to, like, see how their synapses are firing, which is obviously a kind of testing that teams actually do, but it's not really

publicly available. But, you know, we'll be, like, analyzing brainwaves to figure out, like, the actual decision-making process before the decisions are made. So a lot of that helps with predictive power and maybe helps you assess true talent to an extent, but then it's further and further removed from what you actually saw, which makes it valuable,

but can also make it quite confusing for the average person who's like, well, this does not at all map onto the eye test. So that can be kind of tough to get people on board with. Yeah, I think that's well stated. A lot of what we have done, not all of it, certainly, because we do have some confusing stuff too, but I like to think that a lot of

what we have done is essentially just scouting tools, you know? Like, a hundred years ago, some scout would have loved to know how fast this guy was and how much movement this curveball was and how fast this guy's twang his bat. Like, that's not a new idea. It's just you couldn't measure it until recently. And that's, I think, a lot of what we do, but certainly not all of it. As you mentioned, it's hard. I really, I hate naming these things in a lot of ways. The acronyms you were just mentioning. I think my favorite one is Alex Chamberlain's got the VAA A, A, because it's like

a vertical approach above average or vertical approach angle above average. That's another A. And if you read Spencer Nussbaum had a really good piece in the athletic today about the nationals, and their internal system has a metric called PDFX plus, which

sort of makes me think like Adobe made that for acrobat. Yeah. But it gets complicated. It does. Yeah. Okay. A couple quick current events things before I release you. One, weirdly, the big story in baseball this morning was Mets fans applauding Pete Alonzo, who returned to City Field and pretty predictably hit his 300th career homer when he was facing the Mets. And Mets fans celebrated that home run. He got standing ovations. He got a curtain call. And a lot of people, some media members,

were up in arms about, oh, how can the fans be supporting a visiting player? There were some current Mets players who were taken aback by this, who were a little bit flummoxed by the vociferous way that Mets fans supported Piedelonzo, even though this was going against the current model of the Mets. And I think this controversy was sort of silly, like maybe most controversies. But what did you make of fans showing their full-throated support? for the returning Pete. I mean, I don't know what anybody else expected, right?

And I remember, the Mets are not good right now. If this was a game that really mattered to them, then I don't think they'd be cheering against their own team losing, you know? But this is Pete Alonzo. The game didn't really matter that much. I mean, it's a 300th homer. I don't mind the players being upset because they don't have the same, you know, emotional attachment to the franchise as people who've been cheering for 20, 30, 40 years.

So I know Shum and I wasn't thrilled, but whatever. It doesn't really matter. But as far as fans go, like this is someone that they really, really cared about. I can tell you, my son's best friend, his dad had gotten him in Alonzo, Jersey, like two days before he signed with the Orioles. Now, should Paul, have you been paying attention? Should you have known that was coming? Perhaps, but even still, right? Like, this is someone that they care about.

So you give him the big standing ovation. It's not going to hurt your team. And I assume any time after that, if it had happened earlier in the series, they would not be cheering the second home run. Yeah, Francisco Lindor said he wished the fans. had applauded more for Jonah Tong when he had a good outing. And then Manaya said, yeah, it's just crazy how things unfolded. I have nothing but respect for Alonzo and love him to death. As the 2026 mets, we've definitely underperformed.

But the guys in here work their ass off every single day. And to have that kind of reaction to an opposing player is crazy to me. I've just never seen anything like that before. Andy Green said that he hasn't seen that either. He said people are free to do whatever they want to do on the field. Our job is to never let anything external impact the way we go about playing the game, etc. So some of the players were more measured in their response, but some of them seemed a little bit miffed. Manaya also said, I know everybody loves him.

He's the man, but he's not here. So to see the reaction from the fans like that, I don't know, just different. It was unusual. I mean, I guess I get why it was jarring. But one thing I've learned is that calling out the fans for their behavior never seems to work out for players. or managers. Sometimes a team will be fighting for a playoff spot and there will be empty seats and people with the team will kind of call out the fans for not supporting them enough.

That always seems to backfire. People don't appreciate that. But beyond that, why not applaud the prodigal pizza? He hit most of those homers in a Mets uniform. Like, you know, I mean, and you're right. Like, Mets are out of it. They clinched a losing season. If that's something you can clinch, I guess it's not something you aspire to clinch.

and why not why not applaud him? And I think the applause was pointed in a way because it did also sort of segue into David Stern's chance. And so part of it was, I think, sending a message and a bit of a rebuke maybe to the Mets front office that Alonzo was not resigned or that the Mets didn't seem to have a ton of interest in bringing him back. So I think there was a subtext to the applause. but it also was a genuine outpouring of affection for a guy who did a lot in a Mets uniform. Yeah, one thing Stadcast does not track currently is standing ovation in a ballpark. It's true. And it wouldn't surprise me if this was the first time all season long that Mets fans really had a chance in a positive way to stand up and be happy.

So come on, give it to them. It's the middle of September. It's a lost season. Give it to them. Yeah, let people enjoy things. They haven't had that many things to enjoy. And I saw a lot of people saying like, oh, this has never happened before. a visiting player getting a curtain call and a lot of people being like, not in New York.

We don't do that here. You know, and there is precedent. This has happened elsewhere, but also in New York. I mean, it's not that uncommon. You know, I know Albert Pujol's got a curtain call after his return to Bush Stadium when he had a home run. But this has happened in Metz history, too, when Mike Piazza returned to then Shea Stadium 20 years ago, 2006, and he hit a couple homers. and he got a couple standing ovations and a curtain call, I believe. And I was not around at the time.

But I think Tom Seaver, same thing when he came back to the Mets. And maybe that was also sort of pointed and notorious trade. And oh, I wish we had kept him. So there's a lot of frustration, obviously, pent up resentment about how the Mets season is gone and how their offseason went. And obviously, Peele-Lonzo, at least for this season, has proved the doubters wrong by having arguably a career year. when people thought, oh, first baseman in his 30s, it's only going to get worse. And at least for now, it's gotten better. He has a career high WRC plus and close to his career high in war right now.

So he should get to take his victory lap. Yeah. There's no way this is unprecedented. Like, I can't go look this up. I don't know. Do you really think that San Francisco fans didn't cheer for Willie Mays when he came back? Or, you know, Eichiro and Griffey going back to Seattle after leaving for other places? Like, there's probably a hundred different examples of this.

Yeah. And I applaud the applause because sometimes you see the opposite, which is people booing because they feel like they have to. And it's always seems sort of silly to me. I guess it's one thing if a guy just like pushes his way out of town and I don't know, requests a trade or demands a trade or something or bad mouths the fan base on the way out. But if someone, A, gets traded, which is out of their control, or in Pidoanzo's case, like, clearly he got a great offer from Baltimore and it didn't seem like the Mets were that

motivated to bring him back. Like, why fault the player for that? So I always say, you know, remember the good times and, and cherish the warm, fuzzy feelings and, you know, don't punish the player. It's like, are we still adjusting to free agency after 50 years? Like, would you do any different if someone offered you a way better job at a different workplace? Wouldn't you chase the dollar? So it seems silly for me to punish players and boo. I'm much more in favor of the curtain call. Ben, there are so many bad things in the world right now. So many.

I'm not going to take something good and happy away from people cheering for someone they love. Yeah. And the other thing is that, yeah, you can pile on Stearns and the Mets. And obviously it was a pretty all-time terrible offseason in terms of results. But Alonzo aside, the decision to let the other longtime Mets leave has been pretty much vindicated. So the guys they signed, the guys they brought in have almost uniformly underperformed. But it's not like Mets fans are really. ruin Edwin Diaz's departure or, you know, Brandon Nimmo or Jeff McNeil, right? I mean, a lot of these

guys have face planted. It's just that Alonzo has made the Mets look bad. Yeah, no. And it's, it's sort of interesting the reaction. And I get there's a lot of emotion that goes in it. I'm so like, totally understood. But they're like, oh, Alonzo's been great. You screwed up. And it's like, well, no one ever thought he wasn't going to be good in 2026. Like that's not the point. The point is you didn't want him for three, four, five years down the road. And maybe that ends up being proven wrong. But it's not like, oh, well, you were wrong because he's good right now. It goes, never the issue.

Yeah. And it doesn't help that Mets First Baseman have been collectively terrible because, I mean, there were projection systems that said, oh, yeah, Mark Viantos, he might kind of fake Piedelonzo like production. And boy, they are dead last in Fangraph's War from First Baseman. and negative 1.1 wins above or in this case below replacement. So that only makes Mets fans pine for the departed Piedelonzo, even more. Last thing I'll say, we are recording in the late innings of White Sox Guardians game three here.

It's looking like a Guardian's victory in this game and also in the series taking two out of three, though it is dangerous to presuppose any outcome involving these teams or the AL Central in general, but right now it's looking good for the Guardians. This has been a fun and weird series in what has turned out to be a really emblematic pennant race, I guess, you know, where the AL Central, like the AL West, it's like no one wants to win. None of these teams is actually that good. The White Sox have been trying to squander their lead seemingly, but it's really coming

down to the wire here. And if this series does go the Guardians way, boy, I mean, it's going to be. close because the White Sox entered game three of this series, half a game up in the AL Central, one in the loss column. They would be tied in the loss column if the Guardians win this thing. And this is their last head-to-head game. So suddenly, despite the mediocrity of these teams, it has turned into a pretty exciting race. And I guess you could say the same about the wildcard race writ large. Yeah, I haven't been watching the game today because I've been a good and responsible podcast guest, Ben, focusing on you. But the first two games of the series,

were really, really fun. Like I enjoyed it a lot. Not always the cleanest baseball, but a lot more interesting than you'd expect from these teams. And I think everybody knows the American League is kind of a mess right now, but I'm not sure enough people

have really internalized to the way that the bracket is currently set up. Because if things land as they currently are set up, White Sox, Cleveland, Houston, one of those three teams is guaranteed a slot in the American League Championship series because they're all on this. Isn't that wild?

And it's like, once you get there, anything can happen, right? Who's to say that Jose Ramirez doesn't remember how to hit, and Stephen Kwan's been great, and Cleveland's got really good pitching, and all of a sudden they're in the World Series because they had the easy side of the bracket, and they took down the Yankees into Best of Seven.

Like, that's not unreasonable, even though I don't think Cleveland's like a terribly good team, and I just don't think enough people have thought about how weird that is going to be once we get to the ALCS. Yeah, this series, emblematic of how weird this series in this race is, is that Tommy Pham, our old friend Tommy Pham, talked himself into the lineup for this third game in the cleanup spot after a literal elevator

pitch where he just happened to be in an elevator with Chris Getz and talked himself. And look, I mean, if you were in a confined space with Tommy Pham and he were trying to get you to do something, who am I to say that I would not acquiesce to that too? But, you know, he was nearing a comeback off the IL, obviously a late pickup by the White Sox and evidently told Gets that he was healthy and ready to go. And not only was he penciled into the lineup, he was batting cleanup. So that's where we are with these teams. And it worked.

And it worked. Listen, Burkami, I really enjoyed him. He's going to end up hitting under 200 for the season. I know. Like all of the scouts are going to have been proven right after all the victory laps of three months. I know. I'm glad I didn't write that article because I was on the point of doing it. But I'm still rooting for him.

I hope that he can write the ship here. But yeah, Tommy fam, I guess he's got the gift of Gab. And he homered. So I guess in this case, at least he knew what he was talking about. And he accounted for all of the runs batted in for the White Sox at least so far as we speak. And then the game before there was that incredible 18 pitch plate appearance with Tristan Peters, where he worked a walk and helped spark a rally. And it looked like, oh, this is going to be like the turning point or this is going to seal things for. Chicago and then maybe not. Maybe it'll turn out to just have been kind of a cool footnote. But

I'm a sucker for extremely long plate appearances. Yeah. I'm going to leave you with this real quick. Tristan Peters, Eendris Gomez, and Brandon Lau or Lowe, I'll remember one of these days. Three guys who were on Tampa Bay last year or earlier this year who've had really good years. So the next time people are like, don't ever trade with the raise, well, they've got more talent than they know what to do with. Brandon is Lough. Nathaniel is low. And then the one I always have to check as Josh, and I believe he is also low. It's usually low, to be clear. It's only Brandon Lau who has so screwed us up because he says it like loud that every time I'm questioning myself now, if you just meet a low in the world, odds are that they're going to pronounce it low. It's

just that my perception of the pronunciation has been skewed by Brandon. Yeah. Thank you for this information that I will never remember. Well, thank you for information that we will try to remember. always a pleasure to read you and talk to you, RIP, ballpark dimensions, but glad we could get you on today. Thank you, Mike. Thank you, Ben. And of course, you can follow Mike on the socials, on Twitter, and on Blue Sky, where sometimes he takes the bait and perhaps should mute people that he is interacting with on there. But you know what? He's fighting the good fight, I guess, and getting the word out.

Thanks again. Well, the Guardians did down the White Sox. Six to three was the final. So Chicago has fallen out of first place. albeit barely. The White Sox do hold the tiebreaker over the Guardians, and they also hold the third wildcard spot with the Js, the Rangers, the O's, the Tigers, not too terribly far behind. Things are getting interesting. Hopefully this episode has already been interesting, but it will continue to be interesting after a brief break when I'll talk to two former front office analysts about what life is like

on the team side and what we're missing out on. How are you? I'm okay. We got so much to do. Delay. Break it in balls and blaking snows. And those stats won't blast themselves. Effectively wild. Well, I'm joined now by two guests who have peaked behind the team curtain and returned to the public sphere to tell us what they saw, or at least to speak in vague generalities about what they saw. because they may still be covered by NDAs. Best we can do. First is a man who's writing I have mentioned several times this season. Andrew Ball was an assistant GM for the Astros, where he was overseeing R&D and sports medicine and performance.

Recently enough that his bio is still on the team site. Don't know if you knew that, Andrew. Don't worry, he got there after the sign stealing stuff. His hands are clean. During the banging scheme era, he was the director of baseball operations for division rival, The Angels. And before that, he worked in pro scouting for the raise.

And actually, his first jobs in baseball were in Indyball with the York Revolution. And as a writer for the fondly remembered SB Nation Sabermetric site beyond the box score, where he attained the prestigious title of Assistant Managing Editor. One of us, one of us. And now he has come full circle and gotten back to blogging, or at least substacking, which is spiritually similar. and he still has one toe dipped into the baseball world. He's doing some consulting for the Pittsburgh Pirates and other companies. But happy to have you back on the internet, Andrew, and on the podcast.

Absolutely. Happy to be here, Ben. We are also joined by Tom Kim, who was until recently an associate quantitative analyst for the Phillies. And he has recently emerged on the interwebs. And he is also doing some interesting research there. He has recently debuted a new model called Open Command, which is intended to quantify, well, you guessed it, command. Don't know if people still say interwebs.

That was probably something people were saying back when Andrew was first writing here. But Tom, happy to have you. Yeah, thanks for having me. So it's fairly rare, I think, to have baseball blogging recidivists to have people who go in and then come back out again, kind of the Keith Law mold of, I was. writing and then I worked for a team and then I got back to writing again. It seems to be pretty sticky that once you're on the inside, you want to stay there or maybe you move on to an entirely different career. But I've been sort of surprised some of the people I've worked with, and this will relate to the topics that we're talking about today, some of the people I work with

whom I thought, you know what, they won't end up working for a team, not because they're not qualified, but because they seem to enjoy being minor public figures in the baseball analysis niche, and they like the interactivity. And then so many of them, one after the other, just disappeared. And you never hear from them again publicly or privately. I'm doing this series this week with Meg on vacation where I've been trying to revive former baseball podcasts. And I have approached some former baseball podcasters who now work for teams. And I didn't have high hopes that I would be able to persuade them.

to come back to the podcasting world. But they've been pretty unanimous in, nope, that time in my life is over. In fact, Jason Parks, whom I emailed, he is now the director of pro scouting for the Diamondbacks, as he has been for years. And his response, which I asked if I could quote, was, it's best that I remain in the shadows as I'm no longer built for the light, which is an extremely Jason Park's response and made me miss him all the more. But Andrew, I wonder why that is.

because there's a lot of burnout in baseball. It's long hours. It's not the highest pay compared to other fields. And so I wonder those of us who maybe get to put stuff out in the public and enjoy the interactivity and then are suddenly writing or doing analysis for an extremely small group and no one knows what work you did. Was that a difficult adjustment for you and has it been nice to come back to the public and be able to share words again? It has been nice. It's different, I think is the best way to say it very simplistically. It's just different on both sides of the coin. And so there's some really nice things about having autonomy to work on what I want to work on whenever I want to and share what I'm working on with people at my leisure. And there's also a lot of benefits to being on the team side and being competitive and having some of those experiences, you know, alongside colleagues working together towards a common goal. So there's a lure in both cases, cons in both cases. I'm glad I've been able to do both of them at various times.

Yeah. And I have to applaud you in person because I'm pretty sure I praised you for this previously on the podcast. But there are so many ways you could have gone with the name of your substack, with Ball being your last name, just so much low-hanging fruit that was sitting there for you. And you just passed up all of it. And the substack is just Andrew Ball or Andrew Ball Notes. substack.com. You passed up all the easy layup wordplay. It's probably more of a lack of creativity than anything, but maybe. Yeah, if Meg and I had consulted for you, you would have

ended up with something very silly that probably would have tarnished your reputation. So I know you're both still maybe involved with teams in some capacity or aspire to be at some point in the future. But Tom, maybe you could talk about this because you have emerged or reemerged recently you had done some public research prior to this, and then you disappeared for a time, and then you come back out swinging with Open Command, which has generated a pretty robust response. And so what are the virtues of putting something like this out in the world? I don't know whether this was at all analogous to things you were working on for the Phillies, but being able to broadcast this publicly as opposed to within the Waldgarden of an MLB team.

Yeah, I don't know if doing Open Command is necessarily me trying to be like a hero of some sort. I owe a lot to Fangraphs and a lot of baseball savant and these websites because that's kind of how I grew my little dream of doing baseball analytics. I think it goes all the way back to something like 2016, 2017 when I was reading Fangraphy, FanGras article all day long. And I think around COVID, I spent so much time on baseball savant that you could give me a name of a pitcher. And I could tell you how much vertical horizontal movement he has on his fastball. Yeah, baseball savant, you know, the querying, the statcast search, that takes some learning. There's a learning curve there because there are so many fields and boxes and filters. And I don't know if I have my masters in baseball savant, but I feel like my savant.

foo is pretty strong where I can generally end up with what I want, but it can be a bit intimidating, but not to someone who spent as much time there as it sounds like you have. Yeah, it's crazy how much data the public has. I don't think any sport comes even close to baseball for all this, all this data that's released to the public. Yeah. But if you had done something similar to what you've done with Open Command, I guess you couldn't have called it open unless it was actually open.

But if you had done closed command for a front office, the feedback that you would have gotten would have been extremely limited. And so there maybe wouldn't have been just putting myself in that place, just the response that you've gotten from the public and people writing articles about it and also people chiming in to ask questions and provide feedback and make recommendations. all of that dialogue, that conversation, I imagine it would have been much more muted. And also, if you care about the ego aspect of it, no one would have known that you did that, except for your colleagues and direct supervisors. That's definitely pretty interesting, because if I were to have done Open Command for a team, it would have been reviewed by other people that I maybe would have had a meeting with because I asked them to, or my manager and then maybe the director.

But when I was posting Open Command on Twitter, there were a lot of people commenting. And some of these people were telling me, like, an edge case that I just never thought about. And maybe no one would have thought about if there were only three people thinking about it. I think the biggest one was the Shoda and Naga, having his catcher glove up really high at all times. I haven't been paying attention to the Cubs baseball. So I never thought about that. That was pretty big. I wonder if had I posted Open Command something like 10 years ago,

maybe there would have been more people thinking about it and giving more insights. But I was, what, 10 years old back then? They're going to make me and Andrew feel ancient. So thanks for that. I'm sorry about that. I have been interested in attempts to quantify command, well, since you were 10 years old, I guess, unfortunately for me. But there have been other efforts out there, and nothing that was really as open source, I guess, as what you've put out here with the GitHub code and all the rest of it. But some more primitive attempts or early attempts based on other data sources, there was Sport Vision's Command FX, which was kind of a compliment to pitch FX and hit FX and ultimately Field FX.

And that was just tracking the catcher's glove and trying to see whether the ball came close to it. And Stats LLC had a command plus metric, which was trying to divine the pitcher's intent to an extent. And then more recently, the stuff models have had command components like Command Plus or Location Plus, rather, and Baseball Perspectus has had its command metrics as well. So why don't you talk a little bit about why this is important for one thing, why it's worth trying to crack this code, and why it's difficult and how you went about it with open. in command. I'm not sure how much more I could say about intuitively how important command is, because it's almost, it's almost too common sense, you know, as a pitcher you want to throw the

ball to where you want to. And if you don't throw it to where you want to, then bad things will happen. I've always known that there's these, there's attempts in the past or, you know, there's these companies already providing command services. But as a person who was always on baseball, Safon, I could never have a, I could never go to a page. that says, you know, average misdistance on his fastball rankings. And I was really getting into pitcher tactics, how the pitcher approaches hitters. And it would depend a lot on how good the pitcher's command is. If a pitcher has bad command, then he doesn't really have an option, right?

He just has to throw it down the middle at the ball spray. But if the pitcher has Kyle Hendricks, oh, Kyle Hendricks is huge because I studied Kyle Hendricks so much. He was really one of the first pitchers to throw him. throw the sinker up and in against opposite-handed hitters. And when I was analyzing him and found he was doing that, that was one of the, that was like an aha moment almost. Yeah, that's a good check. It's if Kyle Hendricks rates well according to your command metric, then maybe you're on the right track.

Yeah, he does rate well. I mean, it's pretty obvious Kyle Hendricks has one of the best command of the game. So I guess that was a good sanity check. So obviously, I'm a student. so I don't have the money to employ tens of hundreds of scouts to look at individual pitches. Or am I like a reporter who can go up to players and ask them, hey, where did you throw that pitch on this pitch and that pitch? So my only option was computer vision, which I took some computer vision classes in school. So I knew how to, I had a decent grasp on.

Yeah, that's where it all got started. I think the main difference between my model, the real difference. the reason why my model has so much more accuracy compared to a something like command effects model is how it doesn't just look at how much the catcher's glove moves or how much the pitch lands away from where the catcher set is glove but i also take into pitcher tendencies where if uh if the pitcher always throws his curve ball you know on average 12 inch under the glove then we should reasonably expect that he's trying to throw a 12 inch under the the glove. And some of the earlier approaches or command effects were just tracking missed distance from the

glove and presuming that the pitcher was actually trying to put the ball directly in the catcher's glove, whereas often it's just sort of a suggestion or a general area. And so you might be unfairly penalizing some pitchers if you assume that they were trying to hit that target precisely. I want to tag you in in a second intro, but Tom, you mentioned open commands accuracy, how do you assess that? Because we're not mind readers and we don't actually know where a pitcher was intending to throw a pitch. And I guess the ground truth model would be someone goes up to the pitcher and says, hey, where were you trying to throw this one on this particular pitch? And you just walk through every pitch they threw, which would be actually kind of a cool

exercise that maybe I should do sometime. But failing that for every pitcher in every pitch, how do you actually assess whether one command model is better than another? That's actually maybe almost an impossible question to answer. And I spent the last couple months really thinking about it hard. What's a good way to have a solid estimate on how far open command is from ground truth? And right now, in the last Twitter thread I put out, my current set of reasoning is if we take a look at just the naive raw glove position, like a command effects approach where you just assume the pitchers throwing at the glove, how far does it land?

And if you do an adjustment based on his typical offsets, so Open Command went through two version upgrades. And the first one was, as I said, doing an offset from the glove. So how much does he typically miss the glove by in a certain direction? And then the second one was glove dependence. So you actually hinted at it like a minute ago where the catcher's glove is not necessarily where he wants it to throw. It's kind of a suggestion. Yeah. And so for some pictures, the catcher's glove almost doesn't really matter.

They already have a predefined location they want to throw to. We throw the pitch calm, right? And so the catcher would just put this glove somewhere in the vicinity, almost sometimes six inches away. And the pitcher still throws it to the same location. So I quantified that through glove dependence, which is how much does the pitch location actually move, depending on how much the glove moves. We can get to that a little bit more in detail later. But so through these two versions, we saw that the average miss distance drop both times by a big margin, which is, of course, the more we can accurately measure where he was trying to throw. The less miss we're going to expect because we have a better idea where he wanted to throw.

And so from that, we had two jumps in accuracy or two drops in average miss. And so I said, okay, let's be a bit progressive with our estimates and say we have two more remaining jumps left to get to ground truth. Because if you look at the target maps of the first one, second one, and the third version, they're wildly different. And so I have a feeling that it's probably not a whole two jumps away, but for the sake of safety, let's say we have two jumps away. And so if we have two jumps away, then our expected miss, average miss is this. And so that means on average our current glove is off from our true target by, I think it was something like four inches every, every pitch. So that was my reasoning. Yeah, because some of the early metrics,

Command Effects would say that the average missed distance was, I don't know, 13 inches or 11 inches or something like that. I've seen Tom Tango cite 8 inches more recently, but some of the early models were kind of confusing because home plate is 17 inches wide, and we're saying we're missing by that much. Aren't these guys good at baseball? I mean, it's hard to put it exactly where you want it,

but that seemed like a lot in the players at the time we're saying, no, this seems like too much. And so it's often the case, I guess, with the first wave of analysis will produce some response that seems a little out of whack with what baseball people and players will say. And then sometimes it turns out that the baseball people and the players were wrong. Sometimes it turns out that the initial round of analysis was off in some way. And then you get more precision, much like with catcher framing, for instance, where some of the early studies suggested, no, this isn't a thing. And then we got more granularity. And, and some of the suddenly it really seemed like an important thing. And I think I've seen some graphics that you may have shared based on Open Command, Tom, where some teams, if you track just the typical target, there are some teams, and we've talked about this, and it's been the rays and the Orioles and some other teams that have been known for this, where they just set up in the same place. The catcher is just always kind of going middle, middle, often, and then they trust

pitchers to let the stuff take the pitches where they will. And so they're hardly even tailoring it to a specific pitch, which makes your job harder, I guess. But Andrew, to the extent that you can say, and if you need to be kind of cagey about it, that's okay. How well would you say Tom's approach here and open command mirrors what you're aware of teams doing or have teams made similar efforts or even more advanced efforts to try to quantify command? I think it's fairly similar in some way. I mean, it's evolved over time. I guess I would start there. I mean, you both mentioned some of the vendors that provide data on this.

And so for one, teams have been purchasing command-type metrics or command data from outside sources for a number of years. Some of them choose solely to farm it out, while others are kind of looking at what third parties are putting together and trying to develop their own tools internally as well. I think the idea of using the average missed distance and measuring against that is really clever and not something I'm 100% aware of the teams, at least where I've been, of people kind of factoring that in in the same way. But it makes a lot of sense, and I think it would improve upon this. But I've seen fairly similar approaches. And I think something you just mentioned, Ben, for a little while there, it seemed like at least the places I was at, every team acknowledged command is really important. but because it's such a hard problem to actually measure what it is, or because some of the early data, as you mentioned,

was saying the average miss distance is 13 inches. Some of the conclusion was essentially, command is important, but we don't have a great way to measure it and or teach it. And so we're just really going to focus on stuff. And even that approach of setting up, you know, catcher setting up their targets in the middle of the plate was, by and large, just the value of a strike is so large. It's so important to just throw pitches in the strike zone to work into advantageous counts

that we want to develop the best stuff that we can, the best velocity, the most movement, and we want to get that in the strike zone. And if pitchers can also hit a smaller target than that, that's great. And I think as teams have tried more sophisticated approaches, they're starting to maybe walk that back a little bit, maybe change the approach. I mean, even one thing that you just said, if you are just focusing on player development, In theory, you could have minor league coaches every day or catchers, you know, having meetings with your pitchers and talking about exactly where they are trying to aim their pitches and reviewing games after the fact. So on the development side, you could get a lot more information on what the intended target is and then use that to help give feedback.

But as far as a tool that's used for scouting, for evaluation, for acquisitions, I think Tom's approach is really good and probably on par with the best you can do. to some degree at, you know, at the teams on, at the team level. This approach of computer vision, maybe one of you can sort of summarize what that entails. And this is effectively wild. You can get into the weeds if you want or you can do the lay level explanation. But this is something that it seems like teams are getting a ton of use out of whether it's for something like this or whether it's, say, checking for pitch tipping, right? Or, you know, guys giving away what pitch they're about to throw, something like.

that, you can just sort of ingest the video and process it and analyze it in a way that would be difficult to do with the eye test alone. So, Andrew, is this a pretty pervasive approach in the game? I think it has become one. A lot of the public sources that we talked about, not all of them or data vendors for not necessarily public, you know, the old approaches of a baseball info solutions inside edge was to just have a team of employees and interns kind of charting this stuff manually. So watching games and watching video to document where the catcher's glove was in this instance or like you said, doing that for a variety of other things. And I think as teams have hired people or become more aware of computer vision, you can start to kind of replace that work,

make it more systematic, make it, you know, more widely applicable and just turn it loose on a lot our problems. And I guess this applies in part to a lot of teams, including the Astros years ago, kind of on the vanguard of this trend, downsizing their pro-scouting staffs, at least at some levels and some locations in favor of video scouts essentially are people who are working remotely with video and data and trying to replicate or improve upon the kind of data that you'd gather in person. Not that nothing is lost in that process, but of course, the old charting and eye test methods were subject to their own biases and flaws. So I can see why there are certain improvements here over and above any kind of cost-cutting measures. Tom, were there any

insights that you've taken away from Open Command thus far, whether it's in terms of particular pictures who stood out to you more than you would have thought, or just. just particular approaches to pitching that, you know, I wonder, I'd love to say that, okay, if we can precisely quantify command in the way that we could be for stuff, then suddenly there will be an emphasis on finesse pitchers and we will just be churning out a generation of Kyle Hendricks's, right? That would be, that'd be nice. And maybe that would have some other excellent byproducts, but do you think that's that's on the horizon or will this always be sort of subordinate to how hard can you throw and how much does your stuff move?

I guess it could be on the horizon as long as I keep working on it. Yeah, it's really interesting stuff. There are, I've been finding some takeaways that are really, really interesting, but it's really all still getting, just getting started. Open command was released, what, four weeks ago now? So all the, all the research is, I guess, still just barely getting started. for as far as I can tell it seems like command is about a third of an importance as stuff which might be a fundamental part about stuff versus command or it could be that open command is still not 100% accurate somewhere in the middle probably but yeah that was a that was a very interesting finding I've heard the idea that from people getting these third party vendors that um stuff is border ported than command.

But I think a couple takeaways you can get for that, or a couple investigations I've done for that is, first of all, almost everyone at the Moby level is pretty good command. You rarely see a pitcher implode walking five guys in a row because

everyone up there is already demonstrated that they can throw strikes in the minors. So there's a pretty hard selection effect. If you just do a correlation between, I mean, if you actually just do a regular correlation between command and results, I think there's actually a negative correlation because of survivor's surprise. Because the pitchers with worst command get away with by having better stuff.

Yeah. Also, I think this approach of throwing down the middle is more effective than people think. If you have like a 10, 11 inch miss, you throw down the middle. you're going to have a third of your pitch at the edge of the zone and you won't not even the pitcher would know which edge that's going to land at so it's it's extra unpredictable compared to a super good command guy intentionally throwing it up and away and so if you have a really good stuff and you have a below average command then throwing down the middle let the ball spray that that does a really good job it seems like and so the even if you have a worse command you actually do you there's a viable approach that works. Unlike, if you have poor stuff, it's not like you can, there's not like a way of getting around it besides command and maybe pitching IQ. I think actually, if you have a good command,

maybe the requirement is also having good intelligence, which it's still just at my hypothesis stage, but if you have a good command, you have seven inch mistence on his fastball and you throw down the middle, it's going to be down the middle and so you can't actually aim down the middle. That strategy actually doesn't work if you have a good command.

So maybe if Aaron Judge is up to bat and he's weak up and in corner on a fast pitch and weak down or away at the slow pitch then you throw it up in it but if you have another hitter maybe you throw it up in a way

and maybe by doing that if you're a rookie pitcher and you see the scouting report and if that's what you do then from Aaron Judge's perspective, you're too predictable. So maybe for a high command pitcher, they actually would take more time getting that experience and increasing their baseball IQ to reach their ceiling, which would be higher than a pitcher with lower or worse command. But I'm not sure.

There are very interesting thoughts. Yeah. I've wondered about that. We've talked about that on the podcast because it seems like the trend more and more is I'm going to go with my strengths as opposed to going with my opponent's weakness. So not that you're not looking at this scouting report, but if you have a great pitch or you excel in a certain part of the zone or something, you might just kind of go with that approach regardless of whom you're facing. I'm glad you said what you said about pitchers being impressive according to this metric because you do often hear, oh, pitchers today, they're just throwers, not pitchers. It's all about velocity.

And, you know, there's probably some truth to that. I guess all else being equal, maybe the same pitcher would have better command if he was dialing it down a little bit than if he was throwing max effort at all times. But even so, even given the ridiculous stuff, these guys are not terrible at hitting their spots. Obviously, you can't computer vision your way to open command metrics for pre-video or at least pre-publicly accessible video. pictures, so it's hard to compare across eras, though that would be really interesting. But one reason I wanted to have you guys on together is that you had recent posts that on the surface almost seemed as if they were in opposition to each other, but maybe are just more like in conversation, because there's this perception that it's diminishing returns and

a lot of the low-hanging fruit is gone. You have someone like Bill James who will often still say, oh, there's so much we don't know and we'll never figure out baseball. And then other people will say, well, relatively speaking, it seems like baseball is almost a solve science at this point. And the gains people are making are marginal. And in the public sphere, maybe they're imperceptible. So you had a tweet, Tom, earlier this month, that started with baseball analytics discourse died two years ago when people stopped feeling like they could contribute to the frontier. When the plus metrics, for example, Stuff Plus came out. machine learning became the bare minimum, and it was basically over.

And it was a lengthy tweet. So you expounded on that idea. It generated a lot of responses. But I thought of that when Andrew had a post about 10 days later entitled How Public Analysts Keep Pace with Professional Sports Teams. And you aren't arguing that they are equal and that there's perfect parity there, but still that there's something that public analysts can contribute. So maybe you can each sort of explain.

what you were getting at here. Tom, what prompted that tweet and what did you mean exactly and what were some of the common responses? Yeah. So I guess back in 2020, 2021 when I first signed, or when I was, when I first signed up on Twitter, I definitely remember there were many more people in this year and there were conversations going on. And after my job at the Phillies, I came back and I remember posting and there was almost nobody on on the page anymore. It was a little strange. It felt like I was speaking into the void. I was actually going online.

One day I was actually searching trying to find every person who is doing baseball research publicly, unpaid baseball research publicly. And there were, I don't know, I'm not as sure if there were 10 of people doing that. And that was, that almost felt strange to me. Yeah. So Twitter specifically. being a bit of a ghost town for baseball discourse. That might be Elon's fault in part. But beyond that, I think you're hitting on something there. I know Tom Tango has talked about this too. I mean,

there used to be a really thriving comment section on his site where he would be posting stuff. And then some of the smartest minds in Sabremetrics would be weighing in. And now it's quiet. You know, there are tumbleweeds blowing by there. And a lot of that, I guess, has to do. with the fact that teams keep snapping up people like both of you, right? So, but you were also, I guess, getting at the fact not just that there are fewer public analysts who are kind of on the baseball bleeding edge, but also that the barrier to entry is higher in some respects, right? I mean, maybe it's lower when it comes to certain technological tools and data that's

available, but the expertise required, someone with my background in English major, you know, it's kind of tough to be cutting edge at this point. I know you worked with some English majors in the front office, Andrew, so I don't want to denigrate my fellow English majors. I'm just saying, you know, if you have to understand machine learning and a lot of these metrics are kind of black box and it's hard to say exactly why they are saying what they're saying if they're mixed modeling. And so is that what you were getting at?

Because I guess that applies not just to the science of baseball, but to any science, right? I mean, you don't just sort of have civilian scientists who are, just making breakthroughs because those breakthroughs haven't been made now. And so you need a lot of expertise and specialized knowledge in order to move the ball to mix my sports metaphors. Yeah. There's definitely a lot of, a lot more background knowledge required to do a lot of baseball research. Maybe, I'm not sure if I don't, I don't even know if I believe this when I say it. It does feel like there's a lot more background knowledge needed. back when people were inventing things like FIP or Sierra,

maybe all you had to know was how to calculate batting average and maybe something like strikeout rates and home run rates to calculate FIP because it's just the one-line formula. Yeah, or even to develop it. Maybe you need a linear regression or something, but it's not super advanced. Yeah, and if, And let's say you want to work on the frontiers of stuff model, then you would, since almost nobody publicly releases their stuff model, then you would have to make your own stuff model, which is very hard.

To make a stuff model, that's as good as anyone in the public is really hard. I wanted to make a stuff model since 2020, and I had to go to school and do four years of computer. science get to a point where I can make a stuff model as good as everyone else. Yeah, I skipped that part. So, well, Andrew, then, come to the defense of the public sector analysts. Why do we or they still have something to offer? I would say, like, you kind of frame this at the beginning, I don't think we're saying altogether different things.

And I do think a lot of what Tom is saying is right. I think even from when I was writing the first time around, I mean, you mentioned. I wrote it Beyond the Box Score, which does not exist anymore. You cannot. In some ways, I'm thankful that you can't go read the articles. I wrote it Beyond the Box Score once upon a time. But there were plenty of people that did really good work and really good research at that site for no money, being in their free time, either because they wanted to work in baseball

or just because they were fans of the game and it was a thriving community that no longer exists. And I think there are probably fewer outlets or fewer places doing this, which is a larger conversation about attention and consumption and all of that. And also, you know, embedded in what Tom is saying is I do think just the bar has raised to some degree if you take something like just projection systems. I think Tom Tango always talks about this, but the abs of the floor is something like the Marcel projection system, which is very well understood. But if you're going to put something out that you feel good about, it has to clear that bar.

Well, now the bar has been raised to Zips and Steamer and Pagot. and, you know, it does make it a little bit more challenging to put something out that is interesting because you just have a higher bar to meet. But what I essentially wrote about is, you know, like you said, I'm not necessarily saying public analysis is better than teams. In fact, I'm not even really totally that interested in which one is better. I'm interested in the fact that teams have every advantage and reason in the world for the analysis to be way ahead of the public. and for almost all of the ideas to emerge from their first, and yet that's not what has historically happened.

And I think there are reasons for that, that that's what's more interesting to me, is how does the public kind of catch up when they don't have as much information, when they don't have access to players, when they don't have full-time jobs doing baseball analysis, and their livelihood isn't, you know, on the decisions that they're making.

It's just interesting to me that you continue to see these ideas like an open command emerge from the public and kind of gain notoriety and influence the way that we understand the game and eventually influence the way the media talks about it, how awards are voted on, and even how teams operate in a lot of instances. So what are the structural advantages that the public has then in terms of driving interesting research? I mean, for one thing, when an idea comes out that has some merit, I don't want to suggest it's a perfect meritocracy, but if a good idea comes along that has some utility,

then it will circulate. So Tom's work on Open Command made its way to me pretty quickly being in these baseball circles and has been adopted and built on already. Whereas Tom, if you had done this for the Phillies, and I don't know exactly what you did for the Phillies, but it may or may not have caught on. Maybe it just would have been buried if someone with the team wasn't that interested in it. And also, maybe you wouldn't have even had the chance to work on it because one nice thing, if you're blogging or tweeting, you can kind of follow your passions and your interests, not that you don't sometimes have assignments too, but if you're just really interested in something

like command, then you can, to your heart's content, just follow that wherever it leads. Whereas if you're an analyst for a team, most of the time, it's probably not just, hey, do whatever you want, tell us what you come up with. Right. Like there's probably some project that gets handed to you that you may or may not be passionate about personally. So I'm sure you've seen that with the several teams you've been with, Andrew. I mean, some good ideas probably just don't really circulate or get adopted or even get generated just because of the way a front office works. Yeah. Frankly, I think you covered a lot of it. But I think, you know, there's a few things at play here that come to mind for me. I mean, at the first one,

radical difference is just the accountability that is there in terms of having a byline on an article that you write or a Twitter account where you're putting something out. There's negative things that come with that. And there are certain times that anonymity is much better and, you know, doesn't introduce biases. But I think it forces you to spend a little bit more time refining things, you know, make sure that you're proud of what you're putting out there. And then it allows people to essentially, you know, find you and comment to you and talk to you about that and give you feedback and allow you to learn much faster. Like you said, Open Command was released not that long ago. And Tom's mentioned just the outpouring of ideas he's gotten from other people.

With teams, you know, there is much more of a process. And you create something and then it goes through an approval process of a few people. And at any one of those various steps, it could sit there for a while and not be reviewed because people just don't have the bandwidth or the time or the interest. It could get shut down. It could get that you're moved off of that for something else. So, you know, ideas just don't flow in the same way. And also that, you know, accountability is you have to kind of earn attention. If you're putting something out there in the public, nobody has to click on that. You have to make it interesting. You have to write it up and make sure that it's well articulated so it can be easily understood.

and adopted if people want to, which, you know, at least in the places I've been, sometimes that happens and sometimes it doesn't. Sometimes a model or a tool is created, and it's almost expected of the people who are going to use that to figure out what it is and how it works or set up a meeting with the person that designed it. So there's already built-in mechanisms in the public space when you're putting out work for those ideas to be more easily understood and adopted. Yeah, and there's part of it, this may be.

says something about me, but when I went from being a baseball prospectus intern and fledgling writer to then being a baseball ops intern, I really missed that interactivity, just hearing from readers, receiving compliments. Again, maybe I'm just thirsty for praise and that's all it is, but it was just like, you know, I might send a memo out or something and no one would really read it or be aware of it. And it just felt like, boy, I sure miss just putting something out there and have people respond to it positively, negatively, at least is generating some sort of response. This is nice.

And it just felt so quiet suddenly. And I guess there's a little less direct benefit to you. If you're working for a team and you do something, it's not like you own it. It's not necessarily associated with you. People working for other teams don't really know that you did that, right? whereas if you have a public byline on something, it's just, it's better known. I mean, you know, people are probably aware that Tom did this open command work and are probably pretty impressed by it in a way where if he had done this as a quant for the Phillies, no one other than the Phillies would have been aware that he had done this. And so there is sort of a self-aggrandizing aspect to it, too, I guess.

But also it does help the work when you get that kind of feedback on it. And I would guess, Andrew, you tell me, but, you know, probably there are short-term priorities. People get assigned to say, okay, what's the answer to this particular question? We have to figure out this thing. And are there people who are just sort of assigned to, hey, brainstorm, come up with something interesting? Or is it more just really directed to here's the thing that we have to do right now because it's this part of the year? It's probably more the short-term assignments are doled out. I think there are individual managers or individual teams that do a good job of creating space

for people to follow their creativity and their passions. But I can say myself, I'm guilty of this. I'm somebody that helped prioritize R&D projects in past experiences. And if you asked me without a list in front of me of all the things we could work on, of how much time I think our group should be spending on just new, innovative, moonshot-type ideas, I would probably say somewhere between 15 to 20%. As soon as you actually put the list of all of the things that we could work on in front of me, I would use up 100% of that time before we really had any time for these types of experiences.

So I do think it's a hard problem to solve, and it takes being really intentional and being okay with the fact that sometimes you're going to work on things, or a lot of times you're going to work on things that nothing is going to come of them, that you're going to follow the thread and there's going to be nothing at the end, but all of that is hopefully worth it to, you know, potentially find, you know, that big competitive advantage or just frankly to keep people engaged and keep them not feeling like they're just on the hamster wheel all of the time. Yeah. Yeah. And I'm sure, you know, this was the way I felt was just the way I'm wired and probably the fact that I didn't have a whole lot to offer a baseball team relative to just the way I'm wired temperamentally, but also just skills-wise what I was suited to do.

I think I was more cut out for podcasting or writing or editing that I was for actually contributing anything to baseball ops just because of my background and skill set or lack thereof. And also I just like wearing sweatpants and pajamas and staying up all night. So that's a little easier when you're a writer. But I would guess that even though there is a public private gap, Tom, you said that you kind of just devoured fan graphs and baseball savant and all these public sites. when you were on the inside, were you still doing that to an extent? I mean, teams are still scouring these sites, whether it's for the writing or the tools, even though every team has some sort of database and some sort of front-end and great UI with data that isn't available publicly.

If a public site is well designed and can give you a good snapshot glimpse of some stuff, it might be still advantageous to look it up there. So, I mean, you know, I'm not looking for you to praise fan graphs on a fan graphs podcast. I'm just wondering whether there was still some use for you, even having access to whatever you had access to, to the stuff that I have to rely on, for instance. Yeah, well, I'm about to praise fan graphs because that's what everyone looks at. Yeah, yeah. I mean, people look at fan graphs all the time, even if you're in the team. It's a, it's a good website. I mean, this is so funny because the other day I was thinking, hey, what's, what's Mason Miller's command?

And I was like, can I find a website that has this? I'm the guy that made, I'm the guy that made command, the open command. And so, you know, there is no website where I can just search of Mason Miller's fastball command. You know what? Yeah, maybe I should go make a command website right now because a lot of people might go on. I think there's a reason why websites like fan graphs and baseball safvon still has millions of visits every every month. Yeah, I guess you can go to his player page and find his location plus or something. So there's something there for you if you want it. But yeah, I remember when I was interning, this was

more than 15 years ago now. This was pre-statcast, but even so, we were privy to some stuff that wasn't available publicly. And at the time, it was really just proprietary models and video, a lot of angles that weren't available publicly, that's still the case. So definitely when I was cut off from that and went back to blogging, it felt like a phantom win. It was like, oh, I want to go back to this interface and look up this thing that I had access to and now I don't. But there was still so much of that that was replicated publicly and there's now more than there was then. So Andrew, if you can give us an inkling, where are the biggest separators or what are the sources of the greatest gaps in public and private knowledge or research?

Well, first, I just wanted to say, because something you said there made me think of it. I think that's another one of the benefits, sometimes hidden benefit, in this case, to public analysis, is that lack of data. I think it makes people a little bit more creative in some cases when you lose access to something. I don't know if you had this experience, but, you know, if you had something and then it was taken away from you, you know, you could complain and lament. that you don't have access that information anymore,

or you could figure out a way to solve the problem, you know, more creatively in a different way. And it just, it challenges you. So I actually think that's one of the benefits. It doesn't always feel like that, but to working in the public is sometimes having one of your hands tied behind your back makes you take an approach that teams wouldn't think to. As far as where the biggest gaps exist,

I mean, I just do think like the models are better because there is better data. You know, there's better inputs. therefore there's better outputs in some cases. And there's things that are available to you. There's time to work on these. There's access to players. You get more feedback in that sense. You get more feedback from other experts.

You get more feedback from the people actually going out and doing things on a daily basis. But outside of that, I don't know that there's that much of a difference. It still really comes down to who can generate interesting questions and the right questions in certain cases to make sure you're not. being biased in some way or you're not chasing the wrong things and you know we haven't talked about this at all and i'm not necessarily saying i want to go down this rabbit hole but i do think you know with the influx and as ai kind of comes up and changes the technical barriers to doing some of this work i think it's going to become only more important that people are good at those components of the job of asking really good questions and then making sure that they can figure out how to

apply and implement the things that they're taking away from data and information And Andrew, you did a couple pieces. This was in June and July that are maybe relevant to this. I've written a little bit about how teams try to protect secrets. And you wrote a post entitled Your Ideas Aren't That Valuable, which suggested that maybe secrets are a little overrated. You also did a post called Our MLB Teams Getting Better at Drafting. And the conclusion, spoiler, was yes, but maybe a little less than you would think, given the strides that have happened. when it comes to the availability of data.

And maybe part of that is just there's inherent randomness and difficulty in the draft. But can you explain that a little bit why just having the best proprietary model and then slapping an NDA on it so that no one else knows about it actually pays fewer dividends than you might think or why the gains that teams have made don't scale perfectly along with, hey, we have better data than we used to. Yeah. To the first part, I think the biggest reason there, the biggest factor that sometimes keeping this information internal and keeping it a secret is overrated is some of the things we've already talked about as far as learning and collaboration. And so, you know, we've talked about when you put your ideas out there, when you post open command publicly, you generate a ton of feedback right away.

People take interest and they tell you things that you might be missing. you learn a lot more than if you keep that to a smaller group. And I think there are certainly things that teams are developing that are competitive edges that they should try to hold on to as long as they can. But for the most part, those edges are pretty fleeting. And if your success is going to be defined by this one edge that's going to last for a short amount of time, you're probably not going to succeed in a long run.

I think it's more about how quickly can you put ideas out there, learn from them, take advantage of them, and then move on to something else and have a new idea. And I was thinking about this a little bit as we were just talking about how much teams, you know, follow the work that's being done in the public space. I would say teams are very aware of the work that's being done at Fangraphs, baseball prospectus, you know, anywhere out there of substacks. Like if people are doing good work, it's finding its ways into teams because somebody's reading it and they're sharing it with their colleagues, far more frequently, I guess I would say,

are they actually reaching out to that person and asking questions and, you know, building upon those ideas or having any sort of idea exchange? So to some degree, like, they are learning from those ideas, but they're not learning as much as they could if they were more willing to engage with them in a way. But, you know, sometimes they're worried that even a question that they ask might, you know, tip off people to what they're thinking about and what they're doing. So it's, I don't know, it's just gotten to be a little bit much for me. And in terms of why I think they haven't made some of the strides in the article I wrote about the draft, I do think it comes back to some of the implementation. I wrote in one of my most recent article, something that I've said a lot of times before, that I think if you took the average effectively wild listener and you showed them what was happening in Major League Baseball front offices as far as analysis, I do think they'd be incredibly impressed.

There are really talented people doing high-level work all of the time. And I think if you also kept them there and showed them how much that work was, you know, influencing any given decision, they'd be a little bit more distraught because it's not always making the impact that you would think it should. And there's a lot of reasons for that. I think it's sometimes because the analysts doing the work aren't explaining the work well enough and giving people a reason to adopt it. I think it's because it goes through so many checks and balances to get approval. There's a lot of reasons for it. But it's, There's high-level work that's being done that just isn't necessarily incorporating all of the information and influencing decisions as well as it necessarily could be. Yeah, I'm remembering back to my intern days when we'd spend an entire day putting together binders full of scouting insights for the upcoming series, and then they would just sit and gather dust in the corner of the dugout for the next few days, which was kind of deflating. That sort of thing doesn't happen quite as much as it did back then. And there are probably Yankees fans who would say that they're paying too close attention to the binders or I guess the iPads these days. But, yeah, tends to be front offices and field staff more in sync than they once were.

And some of that work is not wasted. But even so, there is a gap between an insight and the successful implementation of it. Okay, maybe last thing you can take us out with this, Tom. I did want to ask you about some research that you publicized previously. on Hustle Plus. This was a paper you published last year. This was something I wanted to look into at one point, and I did, but it didn't really pay off for me because, again, I neglected to go to school for four years for computer

science. But you were trying to quantify who is really running hard and whether that's worthwhile. Because I always question, well, look, you know, sometimes discretion is the better part of valor. You could not sprint and not pull a hammy or something. And maybe you'll miss out on the odd base hit, but you won't go on the IL. You'll get a bunch of base hits that you wouldn't have gotten otherwise. But it does still frustrate people, of course, when someone seems to lolly gag.

So how did you go about quantifying hustle? Of course, there's injury risk and all that. But I'd say the most annoying thing when I watched baseball is this guy jogging to first face. And then the second basement lobs it and he just barely gets out. And maybe he had just ran as hard as he can. he could have just gotten, you know, one out of 10 of them as safe. And I wanted to quantify that. I'm not sure how well of a job I did because the data set I happened to work with was,

I didn't get a video, I just got tracking data with anonymized players. So hard to sanity check for that. But essentially the process was measure how fast the guy can run and measure how fast does he run. And then the difference is how much he hustles. I was limiting that to just running from home to first, but we all know hustle is everywhere. There's hustle and fielding. There's hustle and running bases. There's hustle and fielding, throwing everything.

So hopefully one day, you know, maybe the public gets access to it, and then we can all see how much everyone hustles and be mad about the right players. Yeah. And you've both been hustling, and we will continue to. But I appreciate your effort on this episode. And, you know, since you invoked the specter of AI and LLMs and the rest of it, Andrew, are we in a world where teams are all vibe coding that maybe they're looking for different backgrounds because people can fake their way to coding something more easily? Do you think this is something that teams are trying to replace people with?

And could that backfire? Or is it an additive tool? And it's just something that will save people labor. I guess how AI brained our baseball teams these days. It's probably all of the above, if you're talking about the entire landscape of the 30 baseball teams. I think it depends team to team how much they are implementing things. And obviously, AIA is progressing every day. But I think there are some cases where people are changing the requirements for what it means to be in certain roles because of how you can maybe utilize the tools available to you to lower the barrier to entry and focus on different traits in terms of who you're bringing in or who you're putting.

in certain roles. I think there are places where the technology could, if not replace people, replace processes or improve processes, something like computer vision or using this technology to do something that was done manually or was done less efficiently in the past is certainly another application. So I think there's a lot of good ideas right now in terms of how the technology can be leveraged and implemented. At least in my experience, I haven't seen a ton of full-on adoption. I think people are more skeptical and there's a lot of inertia and tradition in baseball. So we're probably not on the cutting edge as far as as far as AI adoption goes at this point. Well, unless and until we are replaced or exterminated by AI, we will keep pumping out podcasts.

And I hope that you two will keep writing and researching. And really, I'm shooting myself in the foot here because I'm only bringing more attention to Andrew's work and Tom's by Highland. highlighting open command. And there are, strangely, front office folks who listen to effectively wild. Apologies to all of those people. But just highlighting Tom's work, I don't know what your aspirations are, but I'm sure you have heard from many a front office member about this research once you put it out there. And if you were to apply for another baseball job, it's helpful to have this sort of thing out there, not just as a private project, but as something that has been stress tested publicly, but happy that you have at least moonlighted again as public baseball

analysts and writers. And for our sakes, I hope that will continue. But everyone should check out Tom's research, which I will link to, and also Andrew's substack, give it the effectively wild bump. I will link to that as well. Andrew and Tom, thanks so much for joining me. And thanks, Ben. Cheers. Okay, by the way, if you're interested in hearing more from Andrew, he texted me this after we finished recording, I meant to mention before recording that I recently started a podcast, because the world needs more pods, with two of my former Ray's colleagues.

It's a longer format show where we talk about championship sports teams and the decisions that built them. It's called Winners, and we released our first two episodes about two weeks ago. Elsewhere, he has described it as a podcast telling the stories of championship sports teams through the decisions, strategies, performances, and lucky bounces required to win it all. The first couple episodes were about the 2016 Cubs and the 2004 Pistons. I'll link to it, check it out.

Speaking of podcasts, I do one about video games. It's called ButtonMash. You can find it on the Ring Reverse feed. And last week, I did two episodes about Marvel's Wolverine, a new big budget high-profile video game. It's not a great game. However, it is a baseball game. In some of the Tokyo levels, you see big baseball billboards.

But beyond that, in one of the earlier stages, Wolverine comes across a pinned-up newspaper page from the Daily Bugle with the headline, Maelstroms win the pennant, and a photo of a baseball player celebrating winning the pennant. And if you interact with it, Wolverine, played by Liam McIntyre, says this. Not a good sport, baseball. Well, hockey too. Baseball and hockey. Hey, those are my two favorite sports, too. Good call, Logan. A couple other follow-ups. We talked about Mickey Gasper, improbably being the best hitter in baseball for a few weeks beginning in mid-August. Listener Nicholas notes something that we didn't point out. He says that

Like Hannah Kaiser, he had never heard of Mickey Gasper either, until I saw a note somewhere online that the Red Sox were sending the guy who'd been the best hitter in MLB over the previous couple weeks to AAA when Trevor Story returned from the IL. He was only there for one day, as Wilson Contreras suffered an injury and Gasper was called back up and homered in each of the next two games. So yes, he was actually optioned for a day in the middle of the hot streak you were referring to. That only enhances his story.

Also got a response to some Cade Winquest banter from episode 2465 back in mid-April. Rufus writes, I was listening to a back catalog episode from earlier in the year and was reminded of what Cade Winquest went through, with never actually debuting with the Yankees after multiple chances. He was at that point a phantom major leaguer. Ben and Meg spoke to hoping Winquest saw better days, and upon checking, I was happy to see that's the case.

I don't recall hearing an update on the podcast, so I thought I'd mention that Cade Winquest finally had his debut with the Cardinals, and it appears to have gone well. Beyond his debut, he seems to be off to a solid start of his pro career. Good point. He was selected by the Yankees in the Rule 5 draft in December. He was returned to the Cardinals on April 13th, and he did make his Major League debut on August 29th. And now has several games under his belt with an unsightly ERA, but a respectable FIP. And hey, he was credited with a win on September 4th.

So WinQuest's WinQuest is complete. And that concludes this Cade WinQuest Inquest. Also, I'd say the most common response we got to our top 24 players' 24 or younger draft on episode 2530 was, Where was Jordan Walker? Patriot supporter Casey Joe wrote to say, I was listening to the latest episode with Michael Bowman and Bobby Wagner and realized that no one drafted Jordan Walker. Was he eligible? He's 24, but I didn't know if he was disqualified by baseball age. No, he wasn't. He was eligible. Listener Bobby wrote in to say, I'm a little curious about why nobody drafted Walker.

The Zip's rest of career chart projects him for eight war the rest of the rest of the way, which makes it seem like Zips doesn't really believe he figured something out this year. Yeah, in fact, it's worse than that. According to the spreadsheet I linked to, Zips had him with 5.5 war for the rest of his career, which certainly seems low to me. We probably should have mentioned him as a snub, at least. I guess not being mentioned as a snub is the ultimate snub. Obviously, he's taken a big, if belated step forward this year. He was an all-star. He won the home run derby. But as I said to Bobby, it's mostly a matter of his defense already grading out really poorly, even with his belated breakout with the bat this year. And yeah, you could call it that.

He's amassed only about three fan graphs wore. He's also on the older side of eligible players, and if he doesn't have an even higher gear than his ceiling from an overall value standpoint, is a little low for a draft like this, where we were taking only eight players apiece. So, sorry, Jordan. And finally, I talked with Hannah and Zach about the proliferation of position player pitchers and also how bad they've been. Neil Payne took an interesting look at that at his substack, which I will link to. He noted that position players are now allowing almost 1% of all runs, 0.9% to be precise. And it used to be that the percentage of all runs they allowed tracked pretty closely with the percentage of innings they pitched. But now that's all out of whack because

they've been so bad. It's more like 0.4% of innings and 0.9% of runs. I thought this was a compelling way to look at it. Neil noted that there have been more blowouts. The average margin of victory is pretty high, and especially the share of games decided by 10-plus runs, 15-plus runs, 20 runs, has been high. Neil writes, these two factors have now formed a feedback loop of sorts, where margins are more likely to get out of hand, causing teams to put non-pitchers on the mound, which then causes the score to get even more out of hand and so forth. When the score margin was in double digits at a given moment, ERA used to be higher than normal, but not too much so. From 1974 to 2017, the average league-wide ERA was 4.11, which rose to 4.50 with a 10-plus run

margin at the time of the pitch, and 4.55 with a 15-plus run margin. OPS was affected even less, 732 under all situations, 741 with a margin of 10-plus runs, 742 when it got to 15-plus. But since 2018, as we've seen more and more position player pitchers, those figures have ballooned massively. From a 4.2 to 1ERA and 726 OPS across all situations, to 5.18 and 792 when the margin is 10 plus runs, and 6.37 and 864 when it gets to 15 plus. So essentially these days when it rains runs, it pours runs, because position player pitchers pour fuel on the fire. That'll do it for today. Thanks, as always, for listening.

We hope you'll continue to listen, and you'll have something to listen to. As long as people continue to support the podcast on Patreon, which you can do by going to patreon.com slash effectively wild and signing up to pled some monthly or yearly amount to help keep the podcast going, help us stay ad-free, and get yourself access to some perks, as have the following five listeners, Joe Har, Jude Sweeney, John Maynard, Greg Thomas, and Adam Persky, thanks to all of you. Patreon perks include access to a third unrestricted episode each week. That'll be coming up next time, unabridged for the Patreon supporters in the crowd. You can also get yourself access to monthly bonus episodes, membership in our Discord group for patrons only.

exclusive live streams, prioritized email answers, personalized messages, shoutouts at the end of episodes, potential podcast appearances, fan graphs memberships, and more. Check out all the offerings at patreon.com slash effectively wild. If you are a Patreon supporter, you can message us to the Patreon site. You can also send your questions, comments, intro, and outro themes to podcast at fangraphs.com. You can rate, review, and subscribe to Effectively Wild on Apple Podcasts, Spotify, YouTube music, and other podcast platforms. You can join our Facebook group at Facebook.com slash group slash effectively wild.

find the Effectively Wild subreddit at R slash Effectively Wild, and you can check the show notes in the podcast post at Fangraphs or Patreon or the episode description in your podcast app for links to the stories and stats we cited today. Thanks, not yet for the final time to Shane McKeon for his editing and production assistance. I'll be back with one more episode before the end of the week, which means I will talk to you soon. song to death but the shore to make you smile This is effective in wine This is effective in wild This is effective in wild This is effective in wine