What Pitch Counts Miss About Pitcher Fatigue
Ask about Knowledge at Wharton
Episode Description
ABOUT THE EPISODE What if baseball’s most trusted pitching limit is more habit than science?
Travis Sawchik, baseball analytics writer and author of Big Data Baseball and The MVP Machine, explains why 100 pitches is not the full story, how workload units and acute-to-chronic ratios can better measure pitcher fatigue, and what teams may be missing when they focus only on game-day pitch totals.
Also discussed: MLB trade deadline strategy, the Dodgers’ pursuit of another title, Tarik Skubal’s impact, the Brewers’ missed opportunity, when teams should chase variance instead of average gains, Baker Mayfield’s recent production compared with Patrick Mahomes, the Ravens’ coaching changes, Saquon Barkley’s mileage, and college football playoff storylines.
ABOUT THE PODCAST Sports is a game of numbers. Wharton experts Eric Bradlow, Shane Jensen, Cade Massey, and Adi Wyner team up to tackle the world of sports, from current events to longstanding issues such as: What sports streaks are the most impressive? How do you rank the best players? Can athletes be compared across sports? Moneyball Highlights explains how decision-makers in the game can avoid the common mistakes and embrace the data. Episodes are recorded at the Wharton School.
----- The Wharton School of the University of Pennsylvania is committed to sharing its intellectual capital through Knowledge at Wharton, the school’s online business analysis journal.
Launched in 1999, Knowledge at Wharton offers free access to:
- Articles, podcasts, and videos highlighting Wharton faculty research and analysis of current business trends - Interviews with book authors from Wharton School Press - In-depth series of curated content like Wharton Executive Education’s Nano Tools for Leaders - A searchable database of more than 10,000 articles covering all aspects of business - A weekly newsletter that delivers Knowledge at Wharton insights directly to your inbox
Visit the Knowledge at Wharton homepage: https://whr.tn/3qUpSBE Subscribe to the Knowledge at Wharton YouTube channel: https://whr.tn/3DEAgAU
Read the full transcriptHide the full transcript
- Welcome, welcome to Wharton Money Ball. Welcome to another hour of sports analytics here
- on the Wharton Podcast Network. This is Cade Massey hosting today with the whole team.
- We have Eric Bradlow in here. We have Shane Jensen in here. We have Audie Weiner, and we will have
- a guest. We have a guest already. We'll bring him in momentarily. We're gonna go with our usual format.
- We have a guest first half hour. We'll do open lines in the second half hour. We are recording on a Tuesday afternoon as
- we typically do. The show will go up on Wednesday. It is the first week of August.
- The sports calendar is turning. Soon, we will begin diving into all things football, but we've got baseball heating up as
- well, and we will for the next couple of months, and there's always lots to talk
- about on the baseball front. To help us with that, longtime friend of the show, Travis Sochick.
- If you guys have been listening, you know Travis. If you pay attention to baseball, especially baseball
- analytics, you know Travis. He's had a couple of big books, including Big Data Baseball.
- He came out of the world covering the Pirates, and he saw firsthand the data revolution
- with those guys, and then also the MVP machine, which I think it was more Astros,
- but it wasn't as focused on a single team. They just happened to be one of the leaders in that world.
- But you can find Travis's writing. Over the years, he's been at 538, Fangraphs, The Athletic, Baseball America.
- He's done some work with Driveline. I'm gonna ask him about, you'll see him in lots of places.
- Travis, we always love having you. Thanks for making time for us. Yeah, great to be back with you guys,
- and it's always a pleasure to be with you. I'm in the middle of nowhere with modest
- Wi-Fi. If things get too bad, Eric Bravo can jump in here and take lead. Y'all just give me a heads up
- if that happens. What I'm most interested in, I'm the least baseball guy out here, though I'll shag flies
- at least as long as y'all will. I'm most interested in some stuff, Travis, earlier
- this year on pitching loads. Here's my, not attempt to capture it, but just some of the key concepts I'd like
- to hear more about. I think the whole group would like to hear more about. You guys were talking about better ways to
- understand the load on a pitcher, and there was some sense that teams weren't getting it
- right. And in particular, they were pulling some guys when they didn't need to pull guys.
- And you were using a couple things that really catch my attention. Pitching units, I think, you had some measure
- of normalized way of understanding what the actual wear is on a guy from an outing.
- And then you were playing with, I think of it more from the world of sports science and performance, this acute chronic workload that
- is such an important part of managing, as you gear up for a marathon or a half marathon, you've got to worry about, you
- can only go up so fast. You're managing the acute chronic ratio. So y'all were applying that to pitchers.
- And I thought, our audience would love that. I would love that. My guys would love that.
- Let's hear from Travis about that. Yeah, so I think the impetus was, I was tired of seeing guys get pulled with
- no hitters in perfect games, like 80 pitches. And pitch counts really aren't science.
- They're just, it's kind of a number. It's just a very rough, generalized number that
- hopes to measure fatigue. But I think it's really more of a cover our ass measure in baseball.
- So if something goes wrong, it can point to the guardrails. Adi, give us your understanding of pitch counts.
- How well are they used in baseball? So they're used almost like magic numbers. And Travis is about to tell us that
- there isn't really science behind that. That doesn't make managers and commentators and people
- all around baseball think that there's something special about say a hundred pitches.
- And they will pull pitchers when they hit that mark. Or even if they're just shy of it,
- if they give up a hit, it's just sort of this, it's being used as a totem almost.
- And I'm not sure there's much behind it. We know historically that pitchers went way past
- that number. I mean, 150 was absolutely common for a starting pitcher. Finishing the game was common.
- They didn't have to throw a Maddox, which is like an incredibly low pitch count, nine
- inning game. So there has been a jump. Now, obviously they throw a lot harder than
- they used to. So things have changed to make a comparison for the yesteryear is not appropriate today, I
- guess. And people have noticed that the Tommy John incidences and the injuries are higher, but they're
- higher than ever. And they continue to get higher despite the fact that we are imposing these limits.
- And that I think sets up the quandary. That we're in right now, which is if they're doing such great things to measure fatigue
- and are acting on it, why do injury rates stay just as high or even get higher? Yeah, and just to follow up on what
- Adi said real quick, I think there is science behind the more you pitch, the more
- chance it will tear on your body. It's just the kind of, I think like a pitch count of 100 or a pitch
- count, like I think we read more science into these kind of numbers because they sound
- very precise or whatever, but they, you know, I mean, as Travis was telling us, it's
- just, you know, threshold, guesswork threshold. So we'll pitch it back to Travis who's
- been able to get back with us. Travis, you should know that Adi and then Shane just gave us a little bit of
- background saying these numbers seem to have emerged. They're very different from the workload the guys
- used to do years ago. Yes, they save the wear and tear some, but Adi's saying, look, it seems to still,
- we still have lots of wear and tear. So it's not doing the perfect trick. So back to you, Travis.
- Yeah, I think what we're trying to do at Driveline is just understand total fatigue and
- recovery better. And part of it is, and we're lucky to have someone like Josh Hedgka, who's a
- great data scientist, researcher, and he's also a minor league pitcher. So he's been wearing wearable devices and tracking
- how all his throws, his catch play, his bullpens, his game work. And he's seen how his body recovers after.
- So we've come up with, he's played a role in creating these fatigue units, which is
- tries to capture every throw you make in a week and the intensity of the throw. And we use our Pulse technology.
- But yeah, just the idea that we need more information, more data, and pitch counts aren't
- capturing at all. And so we use Pulse to help create this workload unit that measures the actual stress
- in the arm in real time. And we can estimate, even if a pitcher doesn't have Pulse on, we can better estimate
- just by the count of their throw types during a week. And then we applied acute chronic workload ratio.
- We didn't invent that. That's been a long way to measure- Travis, let me stop you.
- Let me- Olympic sports. Okay. Let me stop you before you go to acute chronic ratio, because you're working, you've created
- this new data unit, this workload unit of some kind, and you're saying you built it
- up with real-time sensors on actual athletes so you can differentiate the kinds of pitches.
- And once you have it though, you could estimate that same thing. You can kind of approximate, let's say, even
- if the pitcher's not wearing that unit, if you know what they're doing. So can you tell us more about that?
- Because that's what you're, that's the building block of what you're, where you're going from, from
- here. So tell us more about that unit. Like what moves it around? Also, you guys, he mentioned, they're not just
- tracking end game workload. They're tracking the days between as well. So they need all of that to do
- the acute chronic thing he's going to do in a minute. But again, back to this basic unit.
- Tell us more about that basic unit. It's fascinating. You're saying basically all pitches aren't created equal,
- I think. That's right. And there's a big difference between catch play, you know, a day or two after a
- start, you know, just throwing the outfield or with another teammate. There's a difference between a bullpen throw and
- there's a difference between a game intensity throw. So not all throws during the week are
- the same, but we want to add them all up because all those values mean something.
- And what a lot of athletes have realized in wearing the Pulse technology is that they
- were overworking on what were supposed to be lower intensity days. They were doing too much.
- And if you're not tracking that, then you're raising your overall wear and tear in the
- body, the stress on the body before you even go to a game performance. So one of the big lessons for most
- guys is they do too much throwing or intensity on days when they're supposed to be
- taking it easier. That's one of the big insights. But taking that, and again, this all derives
- from the insights we've learned from putting this on actual athletes and seeing what their week
- looks like wearing what was Modus is now Pulse. What does a game-like throw mean in
- terms of, you know, torque on an elbow, shoulder versus, you know, a bullpen or catch
- play? So it's all derived from that, and that we can extrapolate out. Like we modeled how much stress did Yamamoto
- really put on his body in pitching game six and seven in the World Series? How much stress did he really, how dangerous
- was that? So we can extrapolate out and kind of use what athletes like Hedgcock have done as
- proxies for what other athletes are doing in real time. So again, the problem with pitch counts is
- that might be the only thing a lot of organizations, especially amateur organizations, are measuring in
- the course of a week, but that's not the only throws that matter. So we want to have a better idea
- of the total volume of work and how that can inform what an athlete is prepared to take on during a game day, during
- a total week. And even Hedgcock, who I think deserves a lot of credit for doing this research, he
- changed how he operated in a week as a relief pitcher, and he had a better understanding of what his body could withstand and
- when he could go, you know, what would it take to go multiple innings and prepare
- for that kind of workload. So there's all sorts of insights to be gained from this, but it's essentially a smarter
- pitch count, and it's better than just using a hundred pitches as a kind of cover
- your ass, one size fits all metric for a starting pitcher at the pro game. Well, take us that next step to the
- acute chronic workload, and I'll try to get us a little bit of a headstart by saying, look, this is all training, essentially all
- sports science. It says you can have a more acute spike in your workload if you have a
- bigger chronic capacity. You build up your capacity, your chronic workload, and that gives you more room.
- It's almost a percent. When you're training up for long runs, it's like you don't want to increase your weekly
- distance more than 10% of the time or something like that. And so you're taking this general framework and
- adopting it to pitching workload. And so give us some of the insights you get from that.
- Yeah. I mean, the root of drive-on is basically just borrowing, stealing, the sport science at
- Olympic sports had built up over time, and they're much better at training athletes.
- They have a longer history, more data. So we basically borrowed a lot of those processes, and we applied it to baseball.
- And the general rule is you don't want to, that ratio, you want to keep it under 1.2 or 20%.
- You want to have a 20% spike, I think. It is, you don't want the acute to jump the chronic by 20%, I believe.
- I haven't revisited my March piece in a while, but I believe it's 1.2, the 20% bump.
- So if you can keep your rolling kind of average within that range, we believe you're
- generally in a safe range. There's no, you can, we're never going to prevent all injuries, but we believe there's this
- band where if you don't go over that mark, you're probably in a green light territory.
- And we had some fun with it because we wondered, you know, how did Nolan Ryan build up for a season?
- How did he withstand these 160 pitch counts in a game when he was throwing mid 90s, you have to assume.
- How did, some people have done it. So what did his workload look like? And it was fun, you know, just in
- that Nolan Ryan piece, I found an old Sports Illustrated article where he's talking, he's starting
- to ramp up like high intensity throwing in January. So he was building up that base load
- well before spring training. And he operated, I think, a lot differently than a lot of pitchers did at that
- time. So that's just an anecdotal example, but yeah, to your point, it's basically borrowing from Olympic
- sports science and sports with a longer history and better ways of training and adapting athletes
- and applying that to baseball. And that's a lot of the root of driveline is taking a lot of those, you
- know, weight, under load, overload training, all these things that have existed and all these other
- sports for so long and applying it to baseball and measuring everything. If you can measure it, you can improve
- it. That's at the root of a lot of what we do. And I think fatigue work, we gotta do,
- I mean, baseball as an industry has to do a better job of keeping pitchers on the mound, even with all the, yes, there's
- more stress than ever on bodies, but we have to measure more. We have to, and if we can measure
- more, then we can learn and improve upon that, hopefully. So workload units, acute chronical ratio, that's, you
- know, these are just kind of early efforts at a better understanding of what fatigue is,
- what it looks like and how to better manage pitchers and athletes in the course of
- a given week in pitching. Travis, is it safe to say that the focus of like a hundred pitch count and
- maybe more generally has been almost exclusively on the acute workload and they haven't thought about
- the need to manage the chronic? You wanna manage the acute down, but you wanna manage the chronic up.
- Is that a simple way to think about this? I think so. I mean, it leads to the paradox where
- by trying to manage by overly focusing on the acute, they paradoxically reduce the chronic and
- so they reduce the capacity for the acute. Yeah, and the problem is they're only measuring
- the acute. They have, you know, most organizations, at least for a long period of time, have had
- no idea what an athlete's actual chronic workload was. I think that's probably changing to some degree.
- You know, this is proprietary stuff for a lot of organizations, so who knows what.
- I have to think like the Yankees, Dodgers, some clubs are probably actively monitoring a lot
- of this at this point. But yeah, I think the problem with pitch counts for the longest time is you're just
- measuring the acute aspect and you're flying blind with the chronic. And all the ramp up and build ups
- are kind of the same template, same cloth for everyone. And again, when we've had athletes start to
- wear Pulse and monitor this, they're often surprised. I've talked to Nick Martinez about this and
- other athletes are like, wow, I had no idea I was doing so much in my off days. And so I have to think that's probably
- a common thing across baseball. Pitchers don't wanna feel like they're not doing
- anything. You know, they're guilt, oh, I gotta do something today. But it might be detrimental to your long
- -term health or even in-game performance if you're doing too much in your off days.
- So yes, to your point, Kate, we've measured acute pretty well. There's a decent database on that going back
- to, I think pitch counts and baseball reference go back to 1988. But chronic, we don't have any database publicly
- on that. And I do think something like workload units could be very helpful. One last question for me, and then I'll
- open it up to the guys. But in the article that I read, or articles I read, you had some examples.
- Just to make it kind of, give us an example. It seemed to me that a more complete
- modeling, at least in the illustrations you gave, showed the capacity for some pretty big spikes
- beyond the 100 pitch count. Like you were saying, you know, it looks to us like this guy could have thrown,
- I don't remember the exact numbers, but like 130, 140 and not been too risky. Is it really that extreme a difference from
- what we're seeing actually managed on the field which is what might be possible?
- Yeah, I mean, it has happened. You go back and you look at some of these pitch counts.
- You know, we use the Nolan Ryan example, of course. But even Yamamoto, we modeled out what his
- World Series workload looked like. And, you know, it was the craziest thing to see him come back out for game
- seven. So he modeled how crazy was this. And you can, you have to adapt your, you can't keep doing that, or you can't
- repeat that consecutive times within a couple of weeks. But you can smooth out that rolling average
- if you ease off the next outing or that week. And it drives me nuts because we could
- be seeing more no hitters and perfect games if we just kind of accepted that. Pitchers probably do have the capacity for more
- work within a start. And you think about some of these really efficient outings where you're not throwing as many
- pitches per inning or you're having more rest per inning. So even every pitch in a pitch count
- isn't equal. And I think that's another thing, kind of the next stage of workload units, fatigue units,
- would be, there'd be a time element to it within a performance and probably weather conditions
- and all sorts of other things. So pitch counts are just so- Interesting. Such a, it's not a, it's a dumb
- number that doesn't take into account any other factors. Too coarse. Yeah. Yeah.
- Eric. Yeah, so Travis, I have a ton of questions, but I'll just ask two, but it's
- really one question and a moderator of that. So how non-linear is the relationship between,
- let's call it strain or stress from a pitch and the actual miles per hour? And then secondly, how much heterogeneity is there
- across pitchers and what moderates that heterogeneity? Like, can you flatten your curve, as you
- said, by more training? Can you flatten it by throwing certain types of pitches?
- Can you flatten it by rest in between days? So one, how non-linear are the curves
- and what actually do we know drives the shape of those curves and how much can you do about it?
- Yeah, I think your second question first, when we think about like the Nolan Ryan example,
- if you start building up your chronic workload earlier, longer than you're building up your base
- and you can withstand these acute spikes earlier, and we do, it would be interesting to
- learn more. I mean, I think the breaking ball, fastball, change up pitch, the different pitch types certainly
- can be studied out more too. I think I'd have to ask Josh about how that actually factors into some of this,
- but I think we had more of a kind of a general measurement on that and the linear relationship, I don't know if I
- can give you the best answer on that right now, but that's something we should dive
- into more and come back on another show, give you a better answer. All right, well, Travis has some technical challenges,
- so we won't keep him too long, but before we let him go, what other baseball questions do you have for Travis?
- Team, we can pick up the workload thing after we let Travis go, but opening up more broadly to the entire MLB right now,
- what do you, Shane just wants to talk about the Red Sox, Travis, that's what he wants to know.
- No, this is actually, well, I mean, I guess it came up, I think it's a general team question, is that how much kind
- of you can kind of measure, because you're talking about for a specific pitcher kind of
- managing their workload week to week and stuff like that, but an entire, you can think
- about an entire staff. Can you kind of measure whether certain staffs are overtaxed or not?
- I mean, again, we've got sort of weak measure, kind of weak measures, I guess, for
- that based on outcome, but I kind of feel like, I mean, people are talking like,
- oh, well, my team didn't pick up a bunch of relievers, our relievers are all gonna
- be gassed by August. Is there some sense, could you kind of almost throw up a rankings of how gassed
- different pitching staffs are at this point in the season based on how they're usage up
- till now? Yeah, it would be interesting to have team -wide measurements of this too, and I know
- we have the bullpen grids and consecutive days used and pitch counts per day, and certainly
- you could apply this to team-wide, and maybe to Eric's question too, the linear, we
- know that if you're throwing harder, there's gonna be more stress per pitch, but if you've
- built up a chronic workload, and are some guys better able to handle that? So yeah, there's definitely team-wide applications of
- this. We just need, and we don't know team to team how much proprietary data is being
- kept in this, you know, in the workload silo, and it'd be great to have more public information on this, so much of it.
- We're a little flying blind on the outside too, and the amateur game is a whole nother, we really need more kids and more
- data on kids too, that's a different topic for a different day, but yeah, to your
- point, we should have a better idea of bullpens, are they burnt out? So there's so many applications for just thinking
- about fatigue differently, and just having more, we just need more data in the system.
- Travis, we're just clear of the trade deadline, what storyline or two are you most interested
- in following between now and the end of the season? I mean, I guess, the Dodgers, that's how
- anyone wants to talk about, you either love them or hate them, but I do think - Well, I didn't bring it up.
- A quest, well, if I was a Dodgers executive, I would want to have a, be the third team to try to chase down
- a third consecutive World Series title, that's, I think that's true dynasty status, so going, winning
- the Scooble sweepstakes makes a lot of sense, there's, the interesting thing about the public World
- Series models don't even capture it, that ace starting pitchers, the best pitchers take on more
- innings in the postseason, so I think, even like the fan graphs, World Series chances are
- probably underselling the Dodgers' likelihood of winning again, because Scooble and Yamamoto and their best pitchers,
- Edwin Diaz now being back, are gonna take on more innings, and so I think that's
- compelling just for a historical standpoint, and what that means for, if the Dodgers win a
- third title, the public's gonna look at it as a very unfair game, and what does that mean for labor negotiations, so that's a
- fascinating storyline. Red Sox, they gave up an awful lot to upgrade Ketcher, and Curtis Meade, that was
- unfortunate, I like what the Guardians did, the deadline, I like what the Rays did, I
- thought the Pirates, you know, they upgraded their bullpen, and they did, so those are my
- three deadlines. Travis, do you think what the Tigers gave up, I mean, acquired from the Dodgers for
- Scooble was worth it, it seemed like an incredibly small haul, and that other teams should
- have been able to cough up a lot more, I almost felt like it was an insider job.
- I was a little, I was surprised that there wasn't more, a team like the Brewers,
- I understand they're rational operators, and they don't want to trade too much of future value
- for today value, because they view each season as kind of, you know, a slot machine
- pool, but I'm surprised another team didn't outbid that, if that's indeed the case, because imagine
- Scooble, Misalski, staff, you're not gonna have this many opportunities, if you're the Brewers, to really
- be a potential, have a World Series caliber pitching staff, and team to make that thrust
- through October, so I am surprised there was not a better offer, assuming that this was
- the best offer out there, and of course, teams' evaluation of talent can differ from public
- top 100 lists and prospect rankings, but yeah, I was surprised, it seemed a little light,
- because this is arguably the best pitcher ever traded at the deadline, one of them, anyway,
- the best left-handed pitcher in baseball that's created the trade deadline, and there's not a
- super elite prospect involved, and for the Dodgers, I know people have joked, it's not a
- rental, it's simply, they're just onboarding Scooble now, so it's just, they're one of the few
- teams who could extend him as well, so. I think that's a big part of it, I'm sure teams like the Brewers, or maybe
- a little gun-shy on this Scooble, because the Dodgers can think, oh, this is prospects
- we're giving up for a long-term thing, if you give up a bunch of prospects, and he doesn't resign with you, then you
- kind of, that's gonna make, I think, some of the smaller market teams gun-shy, that's
- kind of my interpretation of some of this. Unless you have a ring in the bank, is it, any of us could answer this,
- but Travis, if you could get your thoughts back to what you were talking about with
- the Brewers, I think there's four or five teams in baseball, that if you, let's take
- our Phillies here in Philadelphia, you partner him with Christopher Sanchez, how are they not, the
- two of them going one-two, him and Mizrowski, or put him in, I don't know, with Chris Sale, put him with, even in
- the Yankees, I don't know if Garrett Cole's the Garrett Cole of old, but you have
- two top pitchers going in the postseason, you're gonna be the favorite in most series, so
- I'm just surprised that, and according to what I saw, Adi may have read differently, the
- Dodgers didn't even give up any of the top five of their farm system talent, so they didn't even give up, they gave up
- three potentially good players, but not even the top five, so how does this make sense?
- It doesn't, just to add to that, Eric, apparently, and I don't know if this is true, that Yankees tried to negotiate with them,
- and they wanted Cam Schlittler, and they wanted Rice, and these are the two best players
- in the Yankees, currently, and what is this, that's why I asked the question, so it
- was flippant, but nevertheless, what's going on, this just doesn't seem right. Yeah, those are the only two players from
- the Yankees I wouldn't trade, the rest of them is, well, the rest of them can all go, but the rest of them can
- all go. Okay, Travis, what do you got? Yeah, I would love to know the exact offers that the Tigers received in this deal,
- so I just have to assume the Tigers operated rationally, and they thought this was the
- best return package, I have no idea if that's true or not. The Dodgers, and I know people have said
- that there's a little bit of a tangent, that the Dodger, anyone could have outbid the
- Dodgers, but the Dodgers don't really need, they have a great developmental system, they're at the
- cutting edge of sports science and baseball, but they don't need their prospects, they can fill
- all their voids externally, they can package prospects, go and go move them for premier talent,
- like they've done for Mookie Betts, for Tyler Glassnow, and now for Scooble, because they can
- leverage your system, because they're financial, they're local media resources, they're, what do they have, $800
- million a year in revenue, no one else can match that, so even though there's this
- narrative that anyone could match the Dodgers, it's not really true, because they can just leverage
- resources another way, and maybe even they can leverage that in a way that even the
- Phillies can't. But yes, Scooble, I am surprised there weren't more offers, because I think about the 2001
- Diamondbacks, and Schilling, and Randy Johnson. Why do we have to, why does this
- come up in my life? Like, yeah. They combined for 36% of regular season innings, but I think it was like 53
- % of post-season innings. So, it just, with the amount of days off, added urgency, someone like Scooble's taken on
- a much higher percentage of post-season innings, and it just really warps the outcomes.
- I saw that the Cowleshy, I know FanGraph says the Dodgers at like 31% World Series odds, but I think the public markets
- have them at 41%, and I think that's probably more accurate, given how this deal, and
- just the elite pitchers they have, and usage ramp up in post-season, and so, it's
- gonna be tough. I think more teams should have been serious about this, and I know clubs like the
- Brewers wanna act rationally, but to win a World Series, you have to be a little
- irrational, and I do think you have to push a few more chips in. So, I'd love to know what their actual
- offer was, because Mizrowski and Scooble could have made a real impact, and how many years
- has it been since the Brewers won a World Series? It's been a long time, and I find
- it amusing. I don't even think it's rational, you're rational. Like, the Brewers, using them as an example,
- how baseball's fine, and the system's fine, is amusing to me, because they haven't put in
- a World Series in like four decades. So, this seemed like a year to maybe push some more chips in.
- So, we would jump in there with you, and say it's not necessarily irrational, it's considering
- optimal variance. Like, you can't get to the top of the pile, especially if you're under-resourced, unless
- you dial up variance. And so, there's a time and place to push in more chips, not irrationally, but because
- you know you need more variance. And man, oh man, I hate to hear you say that, because I'm sitting up here
- in Wisconsin, 45 minutes from the Brewers, with family connections, and it would be fun to
- have them be a little bit more serious. They keep on outperforming expectations, but then that's
- only a regular season, they don't advance. Travis, we should let you go. You've been very generous with your time.
- It's delightful, as always, to talk with you. Wish you the best with what you're working
- on. Keep on throwing it our way, too. We're always interested in what you're doing.
- Yeah, thanks guys. Thanks for dealing with my technical issues here. I think it worked out, you know, reasonably
- well. All right, Travis Socek, and that has been the first half of Wharton Moneyball.
- We still have a half to go. Come back and join us after the break. Welcome to the second half of this week's
- show. Just off the line with Travis Socek, baseball analytics guru and journalist writer, Travis Socek.
- We talked pitching load, among some other things. Some technical challenges, for sure.
- We hope the gist of it came across. Audie was a little quiet, and one thing I've learned, guys, just a second, just to
- name something that I've learned, slowly, over the 12 years we've been doing this show.
- Audie has these zingers at the end of it. We'll be trying to wrap up. I'm trying to bring the whole thing down,
- trying to get away, and he'll have one more question, and it will be, like, the perfect question.
- And so what Audie is thinking about after 30 minutes of conversation, even if he's been
- quiet, is often worthwhile. And I don't have any idea what he's about to say, but I'm pretty sure it's
- gonna be interesting. Audie, what'd you think about that? Let's see your channeling Columbo.
- Yeah, well, actually, I had used the little chat, I mean, the raised hand icon, and
- you didn't pick it up, so that was one of them. You know, it's hard sometimes to get in.
- But Eric actually answered, there are two points that I wanted to make. Eric actually brought it up in more mathematical
- language, but it was an old point of Rick Peterson's that he- Remind us who Rick Peterson was.
- So Rick Peterson was one of our earlier kind of regular features on our show. He was a pitching coach with the Oakland
- A's, and then later with the Mets, and - And the Orioles. And the Orioles, and so he's before, he
- predates all this sort of high-tech Olympic training type stuff that Travis was talking about.
- But one of the things that he taught us that I guess they knew implicitly without
- real data was that a pitcher can slide back a little bit and be a little less 100%, maybe, and this is what Eric
- was calling nonlinear, and the gains on fatigue side would grow much larger than the loss
- of velocity, which he translated out to loss of performance. And so basically what the observation is is
- that you can be a little less, throw a little less hard, and yet, and you can last a lot longer at a very
- small cost in performance. And I think that's a message that has been lost in the current application of pitching
- on the action field. And I think unfortunately those small differences make a big difference for certain pitchers on the,
- you know, I think some pitchers, you know, the ones that are on the margins, they
- need, probably in their mind, they need every advantage they can give. So they're gonna throw as hard as they
- can. So yeah, I mean, I think probably, it may even be common knowledge within the industry
- that you can kind of dial back, and that actually is much better for your own health.
- But the advantage structure. Let me just finish, because I wanna say two things.
- First of all, and I like your point, Shane, some of them are right in the bubble.
- They gotta be as excellent as they can or they're not gonna stay. But take a pitcher like Garrett Cole or
- Tex Google. The guys, imagine getting them eight, nine innings at only 99% of their value, instead
- of six innings of their hundred, I would take the eight, nine. Yeah, or Mr. Rousk just dialed back to
- 99. Yeah. And then like- Stuck to a 105. Also, you've talked about like crucial points in
- a game. Like maybe you, if you're Garrett Cole, who's, you know, he's not as young as he
- used to be. Maybe you say to yourself, all right, look, I'm gonna have 15 pitches that are key
- moments in this game. You know what? Maybe I don't have to throw 97 with nobody on and an O2 count in the
- first inning. Because you know what? This isn't a high leverage situation. So you could easily imagine, easily, if one
- understood the shapes of these curves and one understood something as Travis was talking about, about
- the endurance of these pitchers, this is an OR problem. And it's a solvable OR problem where in
- some sense, you, you know, don't waste effort in low leverage situations. I'm just surprised that teams don't approach it
- this way. Or maybe they do. I don't even know the answer to it. Let me just, the thing that really hit
- me, and there isn't every day that you get a new insight. And Travis gave us one.
- He basically said that in some level, they're missing the training. So what they spend way too much time
- trying to train acutely and they don't build up an endurance base. And that's a really fascinating thing.
- So in other words, to be a great starter, you have to have endurance. So I'm wondering whether or not, and he
- said, he actually said, it's not the acuteness that kills them. You know, it's not the eight innings, nine
- innings of 140 pitches one time. It's lacking the baseline to keep going. Well, to be careful, it was that and
- that's the biggest thing but he was also talking about not keeping track of the off
- day throwing. The quantity. A lot of guys are wearing themselves out because it's too much on off days.
- And that's a little bit at odds with not having the base, but they're not counting
- it and they're not thinking about building the base. Those are two different but related issues.
- So, which always leads me to the question is when you have people young, maybe college,
- maybe even earlier, you got to start experimenting. Take young pitchers and have some of them
- almost, they work on endurance. You do not let them throw hard very often. They work on throwing 95%, a lot, 90
- % and see what happens to their longevity and let's do an experiment. Nobody wants to do experiments.
- And one thing that we don't truly learn cause is from experiment. But the one way in which Major League
- Baseball, I feel like has kind of developed some aspect of this is that young pitchers,
- when you're like kind of getting them up to sort of regular inning load, they kind
- of do, again, they're training the acute part. They're kind of limiting the max as opposed
- to the actual average. But they're progressively trying to increase that match over several starts, whereas maybe they shouldn't be
- progressively trying to increase that max. They should be trying to increase some rolling
- window or like an average. So, I appear to be on an OR bet today. Might as well continue down.
- So, real quick clarification for those who aren't coding it. Operations research.
- Operations research, yeah. So, to follow up on Shane's point, couple things. What is, number one, what is the objective
- function you're trying to maximize when you suggest possibly dialing down and building up endurance?
- That's kind of, number, that's always the question. What are you trying to optimize against?
- And the second thing we don't, as far as I say, we don't really have an understanding of again is how much, is it
- consecutive pitches? Is it rest in between? What does this function look like that actually,
- and it's probably pretty more complicated than we think. It's probably heterogeneous.
- And also when Adi talks about an experiment, Adi, here's the problem. Even if you ran that experiment, I think
- you'd agree, this is a low power setting, which means the experiment you run, are you
- gonna learn about necessarily enough about Adi Weiner's, I'll call it, or maybe not, but I
- might be able to learn something in aggregate or something that drives that heterogeneity.
- I'd be skeptical of being able to learn enough for an individual pitcher with the amount
- of data that you could collect. I mean, basically the question, as I would design something would be pretty crude.
- You have two groups, one that works probably the standard way, whatever they're doing now, and
- another would be heavily on the endurance side. So you really cut back on their acute
- pitching. You still do the strength training stuff, so you don't want to lose that.
- But you really do, because I mean, listen, I haven't played too many sports, but pitching
- is the one thing I used to do at a competitive level, not at a high competitive level, but at a competitive level.
- And every time I got out there, I wanted to throw as hard as I could, because that's just, that's what, while everybody watching
- was watching that one thing, how hard can you throw? And that's it. And there was no process in place to
- train the arm at the levels that are certainly that I was pitching, which was high
- school, middle school, college club baseball, but it was all about hard and no program to
- strengthen the arm other than the maximum velocity, no long-term better. We just was not something that was thought
- of. Obviously, they think about it more now, but I would imagine creating an endurance program and
- seeing what that does to a group of pitchers. Obviously individuals you couldn't get the power on,
- but to a group, maybe you'd see something. Average treatment effect, they call it.
- Guys, let me ask you about another thing that came up in our conversation with him.
- He was talking about, and he's written about this recently, surprised that the brewers didn't push
- some chips in, and in our terms, dial up the variants. And he talked about they want to be
- smart. They're just going to do it. They're going to always follow the numbers. And it just raises this question for me
- of, this is a general question about analytics. Where else in the analytics world are we
- possibly not considering variants? And where might we miss an important decision,
- get it wrong, if we're only worrying about increasing mean edges? Let me give an example.
- Eric's got an example. I've got an example I want to ask you. Let me try one first, Eric.
- Okay, sure. Okay, you're running an NFL draft room, and you've got a model, and you're a good
- Bayesian, or not even a Bayesian, you're a good quant, and you realize, well, there's a
- consensus board. The consensus boards have gotten really good, and they're big, a lot of people, a lot
- of inputs. And so I can look historically at how much our board adds versus what the consensus
- board has. And this is, by the way, this is what sports cameras do. So when Rufus and I were working on
- football rankings, we'd have our own model, and then we'd have the line, and we'd blend
- these together, basically. You regress your model to the line. So a draft, a team, ideally, should run
- the data, and they should shrink their board to the consensus. So at least that works by analogy.
- But here's my question to you. We can all estimate the optimal shrinkage for minimizing squared error.
- That's what we do. Some of the squared errors, Cade. Some of the squared errors.
- Okay, I'm sorry. Minimizing the sum of the squared errors. It's different, because it's- MSE, okay.
- But hopefully you understand what I'm trying to get at. It's like, to what extent would you prescribe,
- if you're running an NFL team, to simply shrink optimally according to minimizing the sum of
- the squared error? Versus you want some optimal variation from the board. You want variance that you're willing to trade
- off a little bit of mean in order to get something different than everybody else is
- doing. And is that not an example of what we're talking about? I thought we were talking about the more,
- if you want to win the World Series, or the- That's a simpler, that's a simpler example.
- You need to pay more for that extra win and take a little bit more chances at the expense, right?
- So- Yes. So that's more like a different, you're looking at a different question, but it's still high
- variance. It's all about buying variance. But also remember, the advantage of doing it
- in baseball, this is the huge advantage in baseball. Remember, let's take the Brewers, because Shane brought
- it up during the discussion with Travis. They're not doing it in a situation, if
- they were to have done this. They know they can win it. Like this isn't 10 games into the season,
- this is over a hundred games in the season. So they're buying variance in a situation where
- they know they have enough mean. And so to me, that's the big difference. I want to get variance when I have
- information, and you have it in the baseball trade deadline. They should have absolutely gone all in.
- I don't know whether they'd win the World Series with them, but you can't tell me.
- I mean, they have the best record in baseball. You can't tell me they don't have enough
- mean with one additional pitcher to potentially win. I mean, I'll just counter argue for the
- Milwaukee Brewers specifically, if you're running a model of like, what gives them the most extra
- probability, it's not more pitching. They got pitching down. They have no hitting.
- So it's like, you know, you could certainly argue that, yes, you invest in variance, but
- not in the part where you're already in the tail. I mean, I understand- Seems reasonable.
- Adding him on top of nothing improves- Who's their second? I don't even know, Shane.
- I'm just asking this honestly. Who's their second best pitcher and how great is he?
- Because that pitcher is going to have to pitch in the post-season too, as many times probably as Mizrowski as well.
- So I'm saying, is it like, is the gap between potentially Skoobel and their second best
- pitcher, or imagine Mizrowski becomes the number two starter. He's now going against the second best pitcher
- on another team. And again, if you think that winning the World Series is just entirely based on who
- you got number one and who you got number two in your starting rotation, yes, but
- you do have to hit as well and they have no- But it's harder to move that needle.
- But Shane, it's harder to move that needle with a single player than it is to move the pitching needle with a single player,
- right? If you're going to add one player- I agree, but we're talking about moving a
- needle off of what will change your win probability in the post-season. If they've kind of, I guess I'm making,
- and we've all got a probability model, I kind of feel like they've maxed out their
- pitching probability. And so all of them- I'm looking right now- You can't move the needle
- on the offensive side. I have to say, Shane, I'm looking at it right now and they have the best
- pitcher in baseball, but their next best on their starting side is number 36. It's their bullpen that's winning them.
- They seem to be in need of a starter, if you ask me. Now, Kyle Harris- Okay, hold on.
- Let's not stay with the Brewers too long. No, I mean, I'm only- I think Shane should have made the statement.
- You guys are acting like it's completely obvious that they should have made it. Well, it reminds me, we've had this discussion
- a little bit, I forget where we had this conversation, but it reminds me of the Cubs' decision their World Series year to go
- grab that reliever for they rented the reliever for the rest of the season at great
- cost. And that in their models, they said- Yeah, Chapman. I believe I've heard that their models moved
- from 25% chance of winning the World Series to 30. It's a massive jump with a single player,
- but that's very- They just took a chance. And this is, Eric, I've heard a number
- of sophisticated operators, running teams, say, all you can do is you want to be competitive
- and they want to get lucky. And what we're saying is you don't have to leave that last step just to the
- gods. You want to be competitive and then where you can, if you can, dial up the
- variance. And, but I'm also asking the more general question is, if we think the Brewers are
- really sharp, and my God, they clearly are, they're outperforming expectations every year for 10 years,
- where is the risk of analytics in pursuit of mean edges, missing the bigger picture that
- there's some other parameters to play with here? Where else might that happen? I just wonder where else analytics might get
- that. Analytics is- I just so you know, in the rundown, I gave an example from football.
- Maybe it's not exactly your case, Shane, Cade, but maybe it is. I was just wondering, you know, I keep
- reading article after article about this quarterback battle for the Browns, and I'm ignoring his personality
- and stuff off the field. It's hard for me to argue that potential upside of Shador Sanders isn't higher than that
- of Deshaun Watson. Look, I think we know what Deshaun Watson is. I don't, my, I'm not saying Deshaun Watson
- isn't better than Shador Sanders. I don't know. All I'm saying is, I don't know.
- So you, I don't think Watson is the, so I was, that was my whole comment. Remember, I wrote in the rundown, it's, I'd
- not, my hands are here. I have not typed it in since you asked your question. I said, play the variance.
- But Eric, I don't, I don't, is that even fair? Shane will have a more informed opinion than
- me, but it feels to me like there is a big question over what the current Deshaun Watson's upside is.
- We know what it was in his first couple of years, but we genuinely don't know whether he still has that in him or
- not. You know, I mean, like he, I mean, you brought it up, I think, almost, was,
- should, I think Deshaun Watson is the invested, if you believe in investing in variance, Deshaun
- Watson is actually the selection, because we know, he's all the way from being on unplayable
- to being a near MVP level, like that we've actually observed, right? And so we've seen the sample of the
- population of Deshaun Watson performance as has incredibly high variance. Oh, I was closer to 50 than 60
- when he was that. That's a long time ago. He doesn't have a- No, no, no, that's right, that's right.
- We're not talking about- Non-stationarity, it's a long time ago. Shadur Sanders has not shown that kind of
- range of performance. Yeah, there's open question on what Shadur's, no, my, Eric would say, look, he's a younger
- player. Ambiguity, like we really don't know. Shane's saying we know the support on Deshaun,
- it's really wide, and Eric's going, well, we don't even know what the support is on
- Shadur, he's too young to know that. And I was more- Well, maybe we agree that the Browns have two kind of
- high variance options. We can agree with that. There's no high mean option, certainly, I feel
- like. What else, while we're in the football world, what else is, camps open? Guys have been going out for about a
- week, at least in most camps. I think we have some games coming up pretty soon. Well, obviously I have a question about the
- Bucs, the Buccaneers. So, there's been lots of stats that have gone around, whether it's completion percentage, touchdowns, yards
- after the catch, quarterback efficiency, that the last three years, Baker Mayfield has been better than
- Patrick Mahomes. Now those, it's a fact, it's an empirical fact. No, I'm just saying, statistically, Shane, it's a
- fact. So, my question becomes then, why aren't the Buccaneers willing to invest more than two years
- in a quarterback who has shown, he's a, look, I'll give him, he's not Josh Allen,
- come on. He's not any of those people, but is he the 10th best quarterback in the league,
- plus or minus three? Sure, yeah, he's that good. And I was just surprised that someone who
- has produced as well as he has, it hasn't been for three games, for the last three years, is it, and he's not that
- old, he's 30, with a longer term contract, given what his performance has been. I'm not saying his next three years are
- gonna be better than Patrick Mahomes. I'm not saying that. All I'm saying is his last three were
- on par with Patrick Mahomes. It would be, I mean, it would be interesting. I mean, what those stats are.
- I mean, maybe this is not a very analytics kind of response, but we've both watched
- football intensely over the last three years. Who's been a better quarter, who's a better
- quarterback? Well, good question. No, no, it's a good question. He doesn't want you to think about it,
- he wants you to give the answer. Who won a Super Bowl in that span, by the way? Who would you rather have had as your
- coach quarterback, than Baker or Baker Mayfield? I would say the last, if you'd asked
- me three years ago, not close. If you asked me two years ago, I would have said closer.
- If you asked me the last season, I'm not sure who the better quarterback was last
- season. Forget the injury, I'm not even talking about that. I don't think Patrick Mahomes has played that
- well over the last 12 to 18 months. But if you're asking me three years ago, not even close.
- Two years ago, getting closer. Last year, so maybe if I integrate over those three years and I want to put
- some trajectory on it, we'll see what happens this year. And maybe part of the answer here, Eric,
- is kind of returning back to our kind of variance argument before is that I kind
- of feel like there's like a, like with Baker, I kind of feel like maybe the hesitation to kind of give him essentially sort
- of the elite franchise quarterback money is, you know, there's, I mean, he's, there's not the
- evidence of the kind of like, this is a person who can kind of take us on a deep run.
- And I mean, you often don't get to observe that because of teams surrounding it.
- It's not entirely Baker Mayfield's fault that they maybe haven't experienced a deep run while he
- has been leading them. But I kind of feel like it's sort of maybe, maybe to the extent that teams
- are, when they're signing that big money, are looking at kind of the ceiling of these
- players. Obviously, that's really where you kind of see the difference between- I know these are
- just summary stats, just to tell you, I know these are just summary stats and this
- isn't everything, but over the last three seasons, Patrick Mahomes has 75 touchdowns and 36 interceptions.
- Baker Mayfield has 95 touchdowns and 37 interceptions. He has a higher completion rate.
- He's thrown for more yards. And so I'm not saying he's a better player. All I'm saying is let's not make it
- seem that at least on aggregate statistics, he isn't comparable on traditional measures of performance.
- That's all I'm saying. And you could argue better on some. Could it be just because he's irritating?
- Is that possible? Well, speaking of irritating, Eric's using my precious online time and making me defend Patrick Mahomes,
- of all people. It's cruel, but- Okay, I have a football question for y'all. The general question is, what do you expect?
- How do you think about the Ravens this year when they have a lot of the same roster, not the same roster, but a
- lot of the same roster, but an entirely different coaching staff? And in particular, their coaching staff is probably
- on average 10 years younger than the previous coaching staff. And you might think about this from a
- modeling perspective. Coaching is kind of outside of most models. How would you tweak up your expectations for
- the Ravens this year, given that kind of changeover among coaches? And in particular, the youth.
- Is that net positive, net negative? How do you think about it? I mean, I think the youth probably is.
- I mean, I think John Harbaugh, I've got nothing but respect for his incredibly long tenure
- there, but probably the youth, it seemed like they did need some sort of change.
- I just don't know how to kind of, I don't know how I'm going to be able to, we'll deconvolve kind of whatever boost
- maybe a new coaching staff has. It's going to be very confounded with, I think they kind of just underperformed.
- The play, you know, had a lot of underperformance last year. And again, you could, it's circular, because maybe
- because of coaching, it was underperformance. But like, you know, if they didn't change
- the coaches, I feel like last year I would almost expect a regression up from that
- team anyway. I see. And so, you know, now you add the kind of coaching that makes, I think those
- are going to be kind of compounded together. I mean, I think it's going to move in both the same direction.
- I expect the Ravens to be much better this year than they were last year. But- And the only thing I would
- do, Kate, to try to tie the last two topics you've brought up together is, we have to agree that, I think under the
- Harbaugh coaching style, which again, extremely, no one's saying he's not a very accomplished coach, been
- very successful in this league. Maybe the matchup between him and Lamar Jackson in that style is not optimal.
- And so why don't we add variants, bring in another coach who's got a totally different
- style potentially of coaching the offensive side of the ball. And maybe, maybe, we don't know, maybe this
- is much better for Lamar Jackson. Now it's hard to argue it would be much better because the guy's what, two-time
- African or three-time MVP? Two-time MVP. But- I think he's already second highest
- running at like rushing yards for a quarterback. No, no, no, I know. I'm just commenting.
- Eric, what in particular are you thinking about possible mismatch between Harbaugh's offense and Lamar?
- Was there a specific idea? Yeah, so this, well, I mean, obviously for a long period of time, he had King
- Henry, right? And so that led to, maybe, maybe, Lamar Jackson should be taking more chances and throwing
- the ball deeper. Like I've always felt that Harbaugh, John Harbaugh that is, had a more conservative coaching style.
- And maybe you have to let Lamar inject some variance. Maybe he's gonna have more mistakes, but maybe
- he's also gonna have more big plays. And we know that big plays are very much correlated with success in the NFL.
- So maybe somebody who's willing to let Lamar Jackson open up the field more, maybe throw
- down the field more, maybe run the ball a little bit less and throw it a little bit more.
- Those are the things I have in mind, but there may be others. Well, this raises the question of what role
- coaches, head coaches who aren't the coordinator, aren't the traditional offensive guy side of the ball,
- like what influence they have on the game plans, on big plays, big moments, because certainly
- they had offensive coordinators making a lot of those choices, but of course he hired the
- offensive coordinators. Okay, fellas, in the last couple of minutes, anything else around the world of sports, particularly
- heavyweights? I have one more for football. So if I asked you right now to project whether Saquon Barkley's 2026 will be more
- like his all-time great year of 2024 or his extraordinarily average year of 2025, he's
- getting towards the ripe old age. He's gonna turn 30 this season, he's 29, but he's played for eight years.
- Matter of fact, we were talking with Travis about mileage on pitchers. How about mileage on running backs?
- Which is a topic we've talked about over the last 12 years, but if you had to bet right now, is his year more
- like the 5.8 a carry he had in 2024 or the 4.1 a carry he had in 2025? Which one are you going with?
- I mean, athletically, he's obviously closer to his 2025 version than his 2024 version.
- That said, I mean, we were just talking about, you were just talking about the Ravens
- and Lamar and the potential for a scheme to play a big role. The Eagles' offense last year was disastrously coached.
- Oh, who's that? And so I think maybe this is just hope or optimism, but I do think you
- could imagine maybe a bounce back year on a part of the law, the offensive players,
- if they can kind of get together. I mean, they don't have the same type of talent.
- They don't have the exact same talent this year compared to last year as well, but
- I'm at least cautious that Saquon can get back a little bit towards that 2024 version
- when they have a little bit more kind of, I guess, of a system in there that actually moves the ball.
- Well, I want to acknowledge that if we were gonna really do, Eric was giving us homework or a hackathon, there's lots that we
- would go get. Shane and I could go build a model. There's a lot of advanced stats.
- We'd do homework on the new system versus the old system. So there's lots more sophisticated ways to answer
- this question, but I'm gonna go with the simplest way, which is he's more or less
- giving us an over-under of 4.95 yards per carry, and I'm hands down, no question, going under.
- Like, there's no question, especially with a two -year later. Two-year later, I'll go under on that
- as well. Yep. All right, it's fun that we're talking football. Guys, on the college front, fall camp's about
- to start as well. So pros underway, college is underway real soon, and we are solidly back.
- Pads are coming on in a vocation. There's no change to the playoff this year in college football.
- Is it the same as last year? There's a couple of very minor changes, but nothing, still 12 teams.
- Notre Dame has, you know, got to walk into the playoffs. There's something about, I think, the conference champions
- are no longer guaranteed the top four spots. Not ever, yes. I remember that. It's just the top four rated, but otherwise,
- it's more or less the same. The big difference, Eric, that the rhetoric that everyone's focused on is that SEC has just
- gone from eight conference games to nine conference games, and the SEC was already complaining about
- their strength of schedule and not being sufficiently considered. Like a nine and three team might be
- better than a 10 and two team because they played a tougher schedule. That rhetoric is already through the ceiling because
- of the additional game they're playing, the additional SEC game. So get ready now.
- I mean, you're gonna be tired of it. I mean, most of us are tired of it already who've been paying attention to college
- football this summer. It's gonna get so old. Even if you're sympathetic, it's gonna get old,
- this conversation about strength of schedule between SEC and Big 10 in particular.
- Oh, well, also, I mean, ACC and Big 12, that's a whole different thing. That's comparing refrigerators to ovens.
- And I'll just say to wrap up my thoughts for the day, I'm shocked that given the Red Sox performance in the last 26
- games, Shane is not wearing his Red Sox hat. Why is he wearing his Hayes hat? Because I'm gonna tell you something.
- This is, I mean, they raked the Dodgers. Yeah, yeah. 23 and three in the last 26.
- No, no, no. It's absurd. It's completely absurd. They've played two. I mean, they're 23 and three.
- They're not just back in playoff contention. There's no reason to believe they're not playing
- as well as anybody in the AL right now. Why would you? I'm not saying to ignore the first half
- of the season, but I'm willing to ignore a lot of the first half of the season, given the way they're playing now.
- Isn't their acquisition at the trade deadline one of the bigger stories as well?
- If we weren't all talking about the Dodgers, we're talking about the catcher they got from
- the Orioles? Yes. Yeah, no, I mean, no, I completely agree. And I mean, in defense of me wearing
- a Philly's cap, for those of you watching on YouTube or wherever, it was a very humid weekend this past weekend in Philadelphia, and
- I was wearing my Red Sox. So my Red Sox cap is in the wash, but it's giving a lot of you,
- I mean, they've been so fantastic. What a fun team to watch. Enjoy that. Enjoy that, Shane.
- Why don't we wrap it there then? On behalf of the whole team, Eric Bradlow, Shane Jensen, who've been through the whole show
- here, Audie Weiner, I had to slip away, take care of some business, Travis Socek, our
- guest this week, Jacob Grodnick, our producer, Deep Patel, the boss lady. Many thanks to all you guys.
- Thank you, the listeners, as well. Come back and join us next time. Between now and then, enjoy your sports.
Brief Summary
Travis Sochick discusses pitcher workload management and the limitations of traditional pitch counts in baseball analytics.
Key points
- Understanding Pitch Counts. Travis Sochick explains how pitch counts are more about tradition than science.
- The Capacity of Pitchers. Travis argues that pitchers may have the capacity for more work than allowed.
- Travis Socek on MLB Workload Management. Travis discusses the importance of measuring pitcher workload and the need for better data.
- Dodgers' Pursuit of Dynasty Status. Travis analyzes the Dodgers' chances of winning a third consecutive World Series title.
- Surprising Trade Deadline Moves. Travis expresses surprise at the lack of offers for top pitchers like Scooble.
- The Importance of Variants. A discussion on how analytics might overlook crucial variants in decision-making.
Episode highlights
- Understanding Pitch CountsTravis Sochick explains how pitch counts are more about tradition than science.“Pitch counts really aren't science; they're just a very rough, generalized number.”3:14Aug 5, 2026
- The Capacity of PitchersTravis argues that pitchers may have the capacity for more work than allowed.“We could be seeing more no-hitters and perfect games if we accepted that.”16:54Aug 5, 2026
- Travis Socek on MLB Workload ManagementTravis discusses the importance of measuring pitcher workload and the need for better data.“We just need more data in the system.”21:40Aug 5, 2026
- Dodgers' Pursuit of Dynasty StatusTravis analyzes the Dodgers' chances of winning a third consecutive World Series title.“That's true dynasty status.”22:20Aug 5, 2026
- Surprising Trade Deadline MovesTravis expresses surprise at the lack of offers for top pitchers like Scooble.“This is arguably the best pitcher ever traded at the deadline.”24:46Aug 5, 2026
- The Importance of VariantsA discussion on how analytics might overlook crucial variants in decision-making.“Where else in the analytics world are we possibly not considering variants?”39:49Aug 5, 2026
- Pitching and WinningDebating whether the Brewers have enough pitching to win the World Series.“You can't tell me they don't have enough mean with one additional pitcher to potentially win.”42:48Aug 5, 2026
- Baker Mayfield vs. Patrick MahomesComparing the performance of Baker Mayfield and Patrick Mahomes over the last three years.“All I'm saying is let's not make it seem that at least on aggregate statistics, he isn't comparable.”52:20Aug 5, 2026





