Effectively Wild: A FanGraphs Baseball Podcast - Effectively Wild Episode 281: The Soft-Tossing Starter Draft/Harry Pavlidis Explains All PITCHf/x

Episode Date: September 6, 2013

Ben, Sam, and guest Harry Pavlidis draft slow-throwing starters, and Harry explains the intricacies of classifying pitches and scouting pitchers from PITCHf/x.

How are we going to know who won? I don't know. Nobody wins in this. Yeah, I don't think anyone wins this draft. Good morning and welcome to episode 281 of Effectively Wild, the daily podcast from Baseball Prospectus. I am Ben Lindberg, joined by Sam Miller. And today for our last show of the week, we are joined by Harry Pavlidis, who is the new director of technology of Baseball Prospectus.

Congratulations on becoming that. Well, thank you and hello. What does that entail exactly? Tell us and also the people listening what that job description is what that the job description is uh depends on the day and the time of the day i really think the job is about ensuring good user experience you know basically we deliver all our products virtually through technology you know sometimes there's actual printed books or our humans show up in places and talk to people. Try to avoid human contact.

Well, you know. It's inevitable. It's part of the job. So for us, it's technology that supports those things. Primarily our end user, our subscribers, our readers, your listeners. Also you guys. Yeah. also you guys yeah because you guys are the editorial side and that that's a user community that i have to take care of as well so i've really been doing a lot of stuff already for the past

couple months but now it's just the role is becoming more more of my time and broader responsibilities so there's going to be some changes with our website in the future that I think we'll all be looking forward to. Well, uh, we are happy to have you and you're gonna, you're gonna get sick of emails from me if you are not already. I was going to say,

I have a frequent emailer of, of tech requests and other requests. So, yes. Uh, okay. frequent emailer of tech requests and other requests. Yes. Okay, so we're going to do two things today. The second thing we're going to do is talk to Harry about his manual classifications of PitchFX and pitch types, which is something I am fascinated by. But first, we're going to do a draft.

Yay! It's going to do a draft. Yay! It's going to be a short draft, I guess. And it comes from a suggestion from Zachary from Connecticut, who is both a BP reader and a podcast listener. And he initially asked me this question in a chat at BP last week, and I said that it would probably work as a draft. So we're going to try it. The question is, if you had to build a rotation of pitchers

who don't average 90 miles per hour with their fastball, who are you taking? So we're going to each draft a full rotation, five starters, of guys whose average fastball velocity is under 90 miles per hour and the winner is going to be uh the the guy whose rotation gets the most warp the most wins above replacement player over the next 365 days um so someone out there hopefully, will keep track of this and report back next September 6th. Does anybody remember about, I don't know, maybe six years ago or so, Posnanski wrote a thing on the Posnanski blog about how the Royals,

basically anything the Royals ever did that made sense was dumb because the Royals had such a losing hand that they should only be doing things that seemed outlandish on the off chance that they work. And one of his ideas was that they basically commit to having a rotation that's only guys who throw in the mid to low 80s, like guys who don't even get signed because they only throw 83. Because who knows? Maybe that's a – you never know what the next crazy overlooked commodity is and maybe it'll be that and i've always thought i i think about that that posnanski column more than i think about almost any posnanski column especially when i hear people criticize teams that like suck already and they do something that makes them

suck still and people are like ripping them. And it's like, yeah, but pretty much anything they did was going to make them suck, you know, continually like, you know, when you're dealt a losing hand, there's not a lot of ways to play it. And so, uh, you know, I don't know, pulling out a knife and like, I don't know, it's like stabbing the guy next to you. It doesn't seem like it makes sense, but it's as good as any other idea. So, uh, so that's when, when this question got asked.

That was the first thing I thought about. Yeah. Isn't that what they call a high variance strategy? Yeah. So there's a buzzword for that. Well, yeah, we're definitely all going to be dealt a losing hand out of this deck because there are fewer of these guys than i probably

would have expected um to to come up with the pool of draftees here we went to the the pitch fx leaderboards at baseball prospectus which are based on harry's classifications uh and there are only there are 41 guys this season who have thrown at least 200 four seamers in starts oh oh yeah right right i lowered it to 100 because the level was so small yeah uh so 100 which is just you know a few starts um and have an average velo of 90 or below so we're gonna draft almost half of the the potential people that we could draft and and it's gonna get to get ugly towards the end. Like there's a, there's a real chance. The end of the first round,

there's a real chance that the first round is going to get dicey. So is it really, is it that bad? It doesn't feel that bad. It feels to me like you're going to like that. We'll all almost have a league average rotations. It feels to me. So maybe I'm someone might,

maybe I'm wrong. Someone might draft Bruce Chen here in a minute. Why wouldn't you draft Bruce Chen? We should, we should just start drafting. Okay. I'm wrong. Someone might draft Bruce Chen here in a minute. Why wouldn't you draft Bruce Chen? He's kind of good. We should just start drafting. I'm drafting Bruce Chen. All right. So Harry did a random draft order.

I ended up with the first pick. So with my first pick, I will take Doug Pfister. Of course. Really? That's an of course? There were only really two choices, I feel like. I feel like two, but I thought that maybe the other one.

I have two choices at this pick, but I'm going to go with Jared Weaver. Okay. Yeah, so I would have personally gone with Jared Weaver or Bruce Chen. So that's your first pick then, Bruce Chen?

No, I'll take R.A. Dickey. Okay, yeah, those were my top three. All right, so with my second pick, because we are trying to accumulate innings here, I am going to go with Mark Burley, who never misses a start. You're not trying to accumulate innings. Well, good innings. Uh-huh. And his innings are pretty good.

So I'll take Mark Burley. Sure. I'm very happy to take, with the fifth pick, Travis Wood. Hmm. I'm very happy to take with the fifth pick Travis Wood that would have been my probably my fifth pick too but I don't know

yeah I don't know it seems fair I'll take I have no idea what I'm doing as everybody knows I'm looking at this list for the first time right now. So I'll take Bronson Arroyo because it's at the top of my list alphabetically. Yeah, Bronson Arroyo is kind of a decent pick in this. When you said the innings thing, when you started to say the innings, I thought for sure you were going to get Arroyo.

Surprise me. I sorted this pool of 41 pitchers by whiff rate uh and he was at the very bottom his his whiff per swing is 4.3 percent which correct me if i mean that means that he gets a whiff every 25 pitches but that's every 25 fast but that's on the fastball okay right yeah uh remember who's got the other stuff so dick's why Dickey's a good pick, because he's not about the fastball. So he's kind of snuck in here. Yeah, what's interesting is I also tried, rather than look at all 41 names, I tried to just sort and figure out what the key was.

And I was surprised because, yeah, the whiff per swing rate doesn't really turn up any kind of method to it, it seems to me. I mean, I'm looking, like, you know, a lot of the high whiff rates are guys we're not going to get to. And I thought velocity, I thought, well, if the whole point of this is that velocity is better and, you know, these guys all suck because they throw under 90, well, the 89's got to be better than the 88's and the 88's got to be better than the 87's. But that's not really at all the case. So it's interesting that there's not really a clear column to sort.

Okay, I'm not thrilled about any of the picks from here on out, but I'm going to take Dan Heron. I have him 10th. He's kind of been good second half, so sure. Okay. good second half uh so sure okay it's a scary group of people i could i barely got to 15 pitchers on my list before i beyond a certain point there's just no one i can name and feel good about myself. I'll take Dylan G. Yeah. Yeah. I had, uh,

Dylan G was, uh, number eight. He was also eight on my list. I'll take, um, I feel like I, I'll take AJ Griffin.

That was my next one. Um, okay. I, man. Uh, all right. I, man. Uh, all right. I'm going to take, uh, all right. I'm going to, I'm going to just take a risk here.

Uh, who is the sadist who gave you guys this question? Zachary in Connecticut caused this. Zachary in Connecticut. Okay, this could be a massive mistake. This could cost me the whole competition. I'm going to take Jaime Garcia. Oh, wow. I didn't bother to, I didn't rank him.

I left him off. It's a risk. Yeah, I was really thinking about that. It's a risk. He's throwing. He's back on the mound, I think, on a mound. He's fragile risk. He's, he's, he's throwing. He's, he's back on the mound. I think on a mound. He's fragile.

He is very, yes, but he's pure talent level. He, he'd probably be in the top five here. Certainly. Yeah. Probably fourth.

Yeah. So I'm going to hope for good health. Yeah. John Danks. Wow. I did not have him ranked. I had him, he was 16th for me. He was the second to last

player I bothered to rank. I'll take Jason Vargas. Okay. It's the last round. Okay. Am I picking already? Your last pick. Okay.

I guess I'll go with Marco Estrada. Okay. This is evil. That guy might not come back this is difficult there's another tantalizing injury guy there's a tantalizing possible retirement guy I'm going to go

with kind of a crazy I'm going to take Tyler Skaggs I was going to take Tyler Skaggs. Oh! I was going to take Tyler Skaggs. Tyler Skaggs really sticks out on this list because he's the only guy who you see his name and you think, oh, he's on here. That's surprising. Yeah, I did not realize that he threw quite that slowly. Yeah, strange thing this year with him. I didn't expect him to be.

I knew he was slow, but I was surprised to see him here. I figured he might get better. I don't know if he'll get faster. I had him seventh on my list, and I would have taken him, except I thought that he would last to me with the last pick. So that's kind of a rumor. Do I have the last pick? This is it.

Yeah, you have the honor of picking. The honorary Bruce Chen slot. I'm actually not going to take Bruce Chen. Partly because if I did more research, I would probably know that the guy I'm going to pick is out for the next nine months or something. But I'm going to take Wandi Rodriguez. That's who I was about to say. I was going to guess

it's who we're taking. I don't know if he's healthy. His recent visit with Dr. James. Elbow arthritis. Yes. He saw Dr. Andrews and it was, quote, good news. Yeah.

If I've got this right, Ben has Pfister, Burley, Heron, Jaime Garcia, and Marco Estrada. That's right. I got Weaver, Travis Wood,

Dylan Gee, John Danks, and Ricky Skaggs, not Tyler. I was trying to add a little bit of bluegrass into mine. Sam took Dickey, Arroyo, Griffin, Vargas, and Lundy. So I'll just quickly go through the names that we didn't pick so that people can know. Okay, so the names that did not get drafted are Andy Pettit,

Paul Mahalem, Bruce Chen, Tom Malone. Mahalem was an interesting one I thought might be pickable. Pettit is too. I'm not sure what he's going to do. Yeah, not sure he'd play. Chen, I thought, would get picked in this. Yeah, I love Malone, but only for what he is. Carlos Villanueva, I bothered to rank him.

I did bother to rank him. He was my 16th or 17th guy. Yeah, he was my 17th. And then Freddy Garcia, Dylan Axelrod, Christopher Rusin, bother to rank him my 16th or 17th guy yeah he was my 17th and then uh freddie garcia dylan axelrod christopher russin brett myers uh john garland jose alvarez ted lily wade leblanc kevin slowy hiram borges bergus burgos michael fires andrew albers scott diamond pedro hernandez Scott Diamond, Pedro Hernandez, Vidal Nuno, Sean Markham, Tyler Cloyd, John Lannan, Eric Stoltz, Yusmero Petit, and Barry Zito. It's almost like throwing hard is beneficial.

Seems like it. That's the conclusion that I would reach. Are these, do you, this is an interesting list too, because with a couple of exceptions, it's almost all guys who basically always threw soft. I expected to see a lot more guys like Freddy Garcia, who are just old now, used to throw fairly hard. Brett Myers is one of those guys.

But it's not. It's guys like Lily and Zito, who even when they were good were living in the high 80s, basically. Heron threw hard at one time. Heron did. Yeah, Heron did throw hard. Yeah, that's a good point. But you're right. For the most part, it's guys who have always had to live at this level.

All right. So there you go, Zachary. All right. So there you go, Zachary. Someone keep track and we'll see if anyone breaks. I don't even know what a rotation of these guys would be expected to accrue over a 365-day period. I think it'd be a nice achievement if one of us had a cumulative ERA plus of 100 or better. I think it's cool. I'll put these into my baseball prospectus team tracker. I will be watching every day

the next 365 how this competition proceeds. Even over the winter. Even over the winter, just in case. Okay, so the other thing that we had you on to do was ask you questions about your manual pitch tagging. who has seen every pitch thrown in the PitchFX era, not in video, but in a plot of some sort or a spreadsheet of some sort, you have seen and classified every pitch that has been thrown since, what, the end of 2006?

Well, yeah, but it's just every pitch that's been tracked. Right, okay. Full tracking since 2008. Plus minor league data Behind the scenes stuff, but it's something like I didn't add it up I should have but I know I'm probably close to six million pitches there So yeah, so that's it's a it's a herculean task or as Sisyphean task or something Something that a Greek a Greek guy would do

Or something that a Greek guy would do. Hey, man. Don't go ethnic on me. So the first question, I guess, for people who might not know is, why do you do this? It's why did I start doing it is one question. Why did I continue doing it is the second. And the third is, why do I still do it? So the reason I started doing it was because it didn't exist you had to classify pitches yourself they didn't start publishing them through the BAM feed since 2008 so the you know in 07 you had to do you had to do

it yourself so that was just you know step one step two was just um there's a mountain over there and I'm going to climb it. There was no real rational reason. I had time or made time. So I just kept going. And then after a while, it became a realization that there was enough quality in it that people found it very useful. That's at the point where I said, okay, I'm going to keep doing this. And so here I am. And what, what would you say the state of the, the automatic classifications are at this point? They keep getting better. I think, I mean, they're pretty good, but they're, I mean, I use the term agreement. I mean, I can't really say who's right, but you know, depending on how you group their, they, they'll put a lot of different classifications I use the term agreement. I mean, I can't really say who's right. But, you know, depending on how you group, they'll put a lot of different classifications around fastballs and stuff,

which is cool. But sometimes that makes things a little more complicated for counting. So between 80% to 86% agreement on their, between what they put out in real time with mine. But that's only if you kind of do some benefit-out grouping. So it could actually be you'll get more pitch types per pitcher. I'll have a more constrained set of pitch types for each pitcher. So you don't have to regroup things in any way. So can you describe the process that you take?

I know a little bit about it, and I saw your presentation at the Sabre seminar. And yeah, I guess just from the moment you wake up, this is a thing that you're doing because I'm often emailing you to ask when something will be done because I'm writing something about someone who pitched the previous night. So describe how this happens. Well, basically, once I have the data into my database, I do a couple quick things to make it easier for me to work with slash better. So I use some of Alan Nathan's stuff to take drag out of the movement of the ball and get a little better idea of spin. the movement of the ball and get a little better idea of spin and put i plot it then with gravity added back in over a longer period of time with a longer flight so it's easier to see the movement

in the pitch and then i look at it with that right alongside a view of how fast the pitch was against the axis it was spinning around so it's basically the movement left and right and the speed and spin axis so the two views that i have in front of me which you saw um i think yeah either half or both so i think you saw both sides of that in saber seminar for a few guys so you know i have a spreadsheet and that's got a lot of macros and stuff behind it. And I have what's, I guess sold as a gaming keyboard that is custom mapped to, you know, run macros, tag pitches, things like that. So sorting and grouping and bundling the stuff happens with keystrokes. And so it goes kind of fast for me because I've been doing it a lot

and I built a system to make it go fast. So as I do it, those colors that you saw in the Saber seminar, they start appearing. So I'm getting immediate feedback on the accuracy of my tags, both the movement and the speed. So that's pretty much all I need to know if it's good. And then Dan Rosenson checks everything possessively

and tells me to fix things that are not right. So in a typical night where there's a full schedule, how many pitches are thrown approximately? 4,500, say. And that takes you how long, roughly, to classify everything? I don't know, probably a couple hours. I don't know. I usually don't do it all at once. It's hard to say. Yeah, a couple hours. I don't know. I usually don't do it all at once. It's hard to say.

Yeah, a couple hours. So when I see those plots, it looks like they're, for most guys, they're pretty clearly identifiable blobs of pitches of one type, but then there are always the borderline pitches that kind of overlap.

And those are the ones that I am curious about. How do you differentiate the ones that are right on top of each other, but it's a different type? As best I can is the first answer. The second is by awareness of arm slot and arm path and idiosyncrasies of pitch fx and how spin and movement can be revealing in confusing ways like i'm able to pick out 92 mile an hour darvish splitters that that those have got to be sinkers but now you go look at the video you actually the catcher actually called for splitter.

You can see that Greppi threw a splitter. So I recognize that because the movement was a little different and the spin was a little different. The rate was different due to an artifact in PitchFX, I think. But it's kind of hard to explain. I probably can't explain. It's a bit of a savant problem, I think. I can tag pitches, but I really don't know how to describe it.

But yeah, I try to figure out where the arm slot, arm path would be. I will break things down to different arm, you know, different kind of group the arm slots. Like if I think the guy's inconsistent with his release point, I'll separate out pitches by release point in my visual and and try to get it that way so like ubaldo jimenez you have to do that bronson arroyo is actually i think it needs to be reviewed again because i don't think he's throwing two different fastballs i was gonna i was gonna ask bronson arroyo is like the he's like the kobayashi maru

of well the best day ever right he's like, I only throw a curveball. I do not throw any sliders. And then I was like, it's just all arm angle. And I was like, yay. So everything on that side is a curveball now. It's made my life a lot easier. Guys like Freddy Garcia are pretty fun. You know, there's a lot of guys.

Actually, a lot of guys on this list that we were looking at. Burley can be tricky. Wood, Arroyo. Jaime Garcia's fastball is a strange pitch. It's almost really a cutter. You know, see, Heron's difficult. He just changed his splitter. Read more about that on Friday in the Washington Post baseball analytics blog I'll be writing.

And, yeah, a lot of these guys. It's funny. All these guys who throw slow are the guys who tend to be blobs more likely than not with their pitches because they rely on changing speed, changing angles, taking a little bit of movement on or off, speed on or off. So, yeah, it's kind of funny that we had this draft and you're talking about those guys. So two questions. One, how often do you have to go to video? Rarely. Maybe a couple times a week.

Okay, good. Dan will do it when he doesn't trust something, but he often confirms what I found. Video is really helpful for guys who throw splitters and cutters. often confirms what I found. Video is really helpful for guys who throw splitters and cutters because some guys use their splits, like Samarja their split will move like a cutter or slider. So sometimes, sometimes for those guys,

you have to look at video. Sometimes there's no point in looking at video also because it may be off the pitch that will be signed differently. So, but yeah, Dan Rosenson likes to do that. He'll check out video and confirm stuff. He likes to keep track of what types of signs guys use.

So if we do have a discrepancy, we do video research and check it out. So you mentioned Arroyo where they're all curveballs, but they do different things. They're essentially different pitches, even if they all fall into the same two-letter abbreviation. And then you have sort of other instances where, you know, like a slider and a curveball, a pitcher slider and curveball, or maybe his slider and cutter kind of exist on a spectrum and they sort of bleed into each other. I just wonder, do you think that the language that we use for these pitches is too limited? I mean... Well mean yes and no i mean i intentionally

keep a narrower band of labels because i think it's helpful and i think the feedback i get from my work both from the public and from my private clients is support that that's that's a better way of doing it um for somebody like a royal like those look like three different curveballs they're just three different arm angles so it's really if you want to ask like well does he throw the three different curves it's like well when he drops his arm it does this i mean it's so what what you end up having is this battle between we want to call it what it looks like we want to call it with the pitcher and his catcher and his pitching coach call it and the the old way oldest way of thinking was just call it

it conventionally looks like but it becomes very confusing we really want to know what a guy calls his pitch we really really really really want to know what the catcher's putting down and if we you know i go through a lot of lengths sometimes like with granky recently trying to find out what is he really throwing because he went from cutter and slider to something kind of in between that got less cutterish and more sliderish over time. And it turns out, you know, it's slider. So it's like, okay, but when did he actually make the change? And even people in the know don't really remember or know. Like, no, he's only thrown one type of pitch all year. It's like, no, I can show you the data. He clearly had two different pitches in like May or June, he changed. And then by july it was something else you know so the challenge

is where do i draw that line where do i change that um so you could so you want to know what the guy says he's throwing but sometimes you have to be a little less nuanced than that because it's going to be confusing because guys use strange names for their pitches or they don't call it something that's really accurate you know and useful so what we're trying to you know compile is secondary tags which aren't so much tags but descriptors that may clarify a grip or what type of cutter it is so for like ted lily he was like this is a this is my cutter i mean i remember when he's at the cubs he'd have a game in the post game he'd be like my cutter was really good today and his catcher like his slider was really good today they're talking about the same pitch and they call it different things and there's i've run into that with a um the reds there was a guy his

there he was calling the guy's change of a splitter and it certainly wasn't a splitter just the catcher just called it that way because it acted like a splitter. So you have to figure out what the best is, and usually there's no satisfactory answer to what everybody – you know, when's a guy's fastball labeled a cutter? Like, is it all the way over to Mariano Rivera land? What about Rafael Soriano? He says that's not a cutter. It's a fastball.

It just cuts sometimes. So I left a fastball. It just cuts sometimes. So I left it fastball. But a lot of other guys, it's the same type of thing where you get some information on the guy. It's like, no, it's just forcing fastball, but it naturally cuts. It's like, well, then we should, if it naturally cuts a lot, we should call it a cutter. So you'll see things change. I think I just a couple months ago finally did Coleman Dirt that way. David Robertson was last year where he finally made his a cutter. Kenley Jansen is a guy who they say,

not just a regular four-seam fastball. It's like, no, that's a cutter. We're going to keep that a cutter. So it's how do you decide? I don't know. You kind of pick your battles and try to please as many audiences. But the secondary tags, you might be able to give that mix of both. This is a cutter that's like a Roy Halladay cutter or a Mariano Rivera cutter.

Because some guys' cutters look just like a normal forcing fastball. So I guess does your brain then in a way kind of work like the automated automated system in that the more the more pitches you you feed it from a certain pitcher the better you're able to classify those pitches or absolutely yeah yeah i mean i try to watch i mean when you say i how often you go to the video well i mean check something like twice twice a week maybe but how often do i watch pitchers pitch you know like a lot you know watch a lot of baseball, and I talk about baseball a lot. And it's, you know, you start to understand what the differences between the pitches are and when you have to look for these kind of crazy nuances.

But it's really hard. I mean, some guys, it's really ambiguous what they're throwing. And you can go back and forth and back and forth on what they're really throwing. And, you know, I've even asked pitchers themselves, and I think they've lied. So it's like, it's what do you, you know, at some point you're just going to have to say, this is what we're calling it. It's a consistent label, and we know what it is. And if it's something that's really wrong or misleading, somebody always pops up and says,

that's not the right thing to call that pitch. The usual feedback that comes, it's like, oh, I understand why you're calling it that. That's a cutter or a slider. Yeah, he calls it a cutter. Like Bumgarner, what is it? Zito, what is it? So you can call them either one. So that was going to be my next question because I feel like I've seen you tweeting with, I don't know, maybe Brandon McCarthy or other people.

I don't know, maybe Brandon McCarthy or other people. What are some instances when you've had a question and just gone directly to a pitcher, either yourself via Twitter or via a beat writer or someone who asked a question for you? Not all the time. I mean, just tonight, just a few hours ago, I asked one of my colleagues at the Washington Post about Dan Heron.

So I probably do that more than I look at video for clarifying things. So yeah, that's not unusual at all. McCarthy is somebody direct. Trevor Bauer has answered questions. There's a few teams that I've managed to get either through beat writers or through just people who work there

and who can talk to the pitching coach for me. Then of course there's the clients who I know I've got good information from them, sometimes real good insight on what different pitches a guy throws that may seem surprising. So sometimes there's that. But really, for the most part, we look at things, try to figure it out, sometimes just ask other people. But going to beat writers is a really good weapon because they can get straight to the source and you know find out things and

they're very helpful there's some really good people who are like almost anxious to help help me do that which is really really i i can't tell you how grateful i am to people like that so yeah so how how feasible is it to scout pitchers through pitch fx i when i i mean when a guy comes up and you look at data from one start or something and you can say that his you know one pitch type of his looks like a pitch that some other pitcher throws and it's been effective or not effective we've talked on the on the show before about teams that have kind of abandoned uh in-person advanced scouting and have gone to pitch effects and video how so how how effective can you be at at scouting a pitcher or evaluating how effective he will be purely by looking at his pitch effects data do you think well i think it's

two separate things. There's the advanced scouting one and the more evaluation scouting. So the advanced scouting, I think, is fantastic for sequencing and things like that out. I mean, you can definitely reduce costs. I don't know why teams don't do it in the minor leagues with this stuff, to be honest. And when guys get called up, you can know what he can do by looking at his minor league data. So I think for advanced scouting, I said this, I think, in my first PitchFX Summit talk in 2008.

And it was on one of my slides. I'm like, you will save money with advanced scouting with PitchFX. And the baseball operations side of the room. People like Keith Walner and a group of people were there. And I kind of had this like blank expression on their face. So then a couple of years later I was like, see, they're doing it.

So on the player evaluation side. Yeah. I mean, I can tell you a lot from one. It's hard from one start, unless I'm familiar with the park. So if you give me like, if you don't tell me what the ballpark is or someplace that I'm not familiar

with the, you know, the idiosyncrasies of that particular pitch FX installation at that point in time, it's harder to do. But otherwise I can definitely scout off of a, off of pitch FX data. Now there's things missing from that, but what I can get, the feedback I have received from scouts is very positive about it. It's like, that's pretty much, yes, you saved someone an airplane trip, except you don't know their deception, their slot, their extension. You don't know some of those things. What's their poise?

This is a whole array of things that you can't do with pitch effects. But to me, it can actually give you the ability to cast a pretty wide net at a very low cost. So if you have data on amateurs, minor leaguers, or major leaguers, and you're looking for guys who have good stuff, you can find that out pretty quickly with this. The trick is getting to deception and effectiveness and command and pitchability. But there are, I think, ways of doing that, combining all the different types of rich data that we have and scouting knowledge into building ways to model those things. So that's what I like to do.

That's fun stuff. But it's really kind of pushing into new territory. So that's where I think pitch FX is going now. I think it plays a very critical role. I think it will play a critical role in talent evaluation and player development. It's just shaping that into actual practice is going to be both technically challenging and culturally challenging.

practice is going to be both technically challenging and culturally challenging. And one question that I get a lot, and I don't know if you know the answer offhand, but what are the pitch effect systems that tend to be hot or cold? It's funny. I don't really catch too much of, even though I'm looking at everything all every day i'm going by park so it's like i'm taking all the guys in that park that day and so i don't kansas city usually runs high i think la i think i think uh us cellular runs hot um the cold ones are i don know. That's hard to say. Those are the ones that generally... But things are pretty much in a narrow band.

I mean, it's pretty consistent. And the ones that are hot are maybe one or two miles an hour at most. I mean, it's not a huge thing anymore. Movement's a different story. I mean, how much the difference is in the movement of clusters and pitches varies from part to part. Where the whole general shifting, both in two and three three dimensions which i won't even try and explain uh is is is different from park to park and so you have to kind of adapt to that as well so like places like minute made and uh tropicana are both shifted over in one direction a very similar degree and i always

think it's i have like conspiracy theories because that's where, you know, two former, you know, baseball, you know, prospectus or Hardball Times writers who moved into that field who are physicists, know everything about pitch effects, and they happen to work for the clubs that have parks with misaligned systems. Like, hmm.

What does that mean? I guess that was going to be my last question, is that it seems like most of the sort of first generation of PitchFX pioneers are now working for teams somewhere. It seems like you are one of the only members of that group who is still out telling us things instead of behind the wall doing it for a team, which is wonderful for us. I don't know, has that been your desire to kind of stay in the public and provide a resource and write and share this information?

Or is it something that's just sort of happened this way? It's the former. I've never pursued anything other than doing business and consulting and providing data. So, I mean, I just, I work for a few teams. There's a way to look at it um but yeah i mean my my thing is definitely having that independence and public interaction that that's a big big thing for me that that's part of what i enjoy about this and why i got into it so it's never been my intention to work for a team so if they have projects and stuff like that, great, sure.

But I've never, I didn't set out into this to go work for a club. I actually was maybe, you know, earlier on, I was like, definitely don't want to have like work for the man. You know, I don't feel like that anymore. But I definitely enjoy the independence that this type of work that I do and the type of relationships I have gives me. That said, if a team gave me a good, good offer and just like showed up with it, I would, you know, I would, I would listen, but you know, that that's, that's not, you know, that's not on my playbook. Right.

Well, we're, we're all happy that you feel that way i think because we we benefit from from all of the information that you have distributed uh so so thank you for your your tireless efforts every day at tagging blobs of pitches and spreadsheets why people enjoy it it's fun so all right uh so you can check out har's PitchFX classifications at brooksbaseball.net and also at Baseball Perspectives. You can follow him on Twitter at Harry Pav and read about him occasionally at Baseball Perspectives and sometimes in the Washington Post and anywhere else. Anything else we should mention? I think that's it. Okay. So we're done for the week. End of week spiel. If you want to send us emails for next week's email show,

you can do so at podcast at baseballprospectus.com. Rate and review us on iTunes if you listen to us there. You can join our Facebook group at facebook.com slash group slash effectively wild and we hope that you have a wonderful weekend and we will be back on monday