Former MLB Pitcher and All-Star, Lary Sorensen, sits down with Chief Technology Officer and Co-Founder of F5 Sports, Steven Cukiernik, to discuss pitchLogic's NEW feature, STUFFpL and SKILLpL.

Full Interview Transcript:

Lary Sorensen: Hi again, everybody. I'm Lary Sorensen, a former Major League pitcher, and I'm talking today with Steven Cukiernik, who's the Chief Technology Officer and a co-founder of F5 Sports. F5 Sports manufacturers the pitchLogic baseball system. Steven, why don't you tell us a little bit about what pitchLogic is and exactly what it does? Absolutely.

Steven Cukiernik: Hi, Lary. pitchLogic is a system for pitchers and for coaches, scouts as well, to measure and drive improvement for pitching. It takes the form of a baseball. It's got all the technology inside of it. It's going to pair to an app on your phone, really easy to use. And it's something that you can use to see where you are and drive improvement.

Lary: So youngsters. Look at this. College guys, high school guys, all ages can use this product.

Steven: That's absolutely right. We have users from at least nine years old, maybe even younger, I'm not sure, up through Major League players.

Lary: And how does it differ as you go from young to old?

Steven: It's interesting. A lot of the fundamentals are kind of the same. A lot of the things you want to be doing in terms of the way you spin the ball are going to be pretty similar. Obviously, the intensity of everything gets much higher as you get to the Major League, and the consistency it gets higher. But you'll see that a lot of the data really scales up kind of with your velocity. So if you throw twice as fast, you're going to pretty much have twice the spin rate.

Lary: And the key metrics to the pitchLogic system.

Steven: There's actually a lot of metrics. There's a ton of things we capture. But really, I think the most useful things for most people are Arm Slot, Spin Direction, and Spin Efficiency. You're going to get your Vertical and Horizontal Movement, or Break. And then there's some mechanical things like your Forward Extension and Backward Extension. It can also be great indicators of maintaining consistent form in your delivery.

Lary: Okay, and recently you just came through with a new breakthrough, a scoring system called STUFFpL and SKILLpL. Can you explain a little bit to us about what they are?

Steven: Absolutely. So, one of the things we found through our journey of building this product, is that pretty much every technology product for pitchers is going to give you a lot of raw data. That raw data is amazing for pro teams, for colleges that have some analysts attached to the team. But it really takes experts to dive into that, or dedicated players and coaches to learn what that data means and try to untangle really what they should be doing with it. And we really want this to be a product that is usable by everybody, so offering a lot more guidance about what's good, what's bad - that's the question we get most. Is this a good pitch? And then the follow up to that is: how do I make it better? How do I compare to my peers? We really wanted to give some guidance there by attaching a score to the pitch, but more than that, giving really targeted feedback on what can I do to make this pitch better?

Lary: How exactly do you use this when you get the baseball? Can you explain to us when you get the baseball and you throw your pitches, what shows up and how you use that?

Steven: For sure. The first thing I would recommend for everybody to do when they get started with pitchLogic is to run, basically an assessment session, what we call a STUFFpL session. So that's going to ask you to run through your pitch arsenal, and it's going to ask you to throw five of each pitch at full intensity from a mound. So what we're really trying to see is how does your stuff play or how would your stuff play in a game, and that's the basis for STUFFpL is using real game data. What pitches are successful and what pitches aren't, across hundreds of thousands of pitches, and identify what you're doing well on that basis. By doing that session what you're going to get is a score for each pitch. An average pitch is 100 points, and above 100 is good, below 100 maybe you could use some work. Although, pitches down in the 80s and 90s can still be effective, especially if you have good command over them. From there you're going to see a heat map, actually two heat maps. One is going to show you where this pitch will be most effective locating it against right-handed batters and one against left-handed batters. So you're going to really quickly see, based on the shape of my slider, it's going to play well down and away from a right-handed batter. I might have an opportunity to go up and inside with that backdoor slider, but it's really likely to get crushed if it's right down the middle. All of that is going to be really visual for you, and it's going to be specific to your exact pitch. Then finally you're going to get three recommendations of things that you could improve. And really what we want to do here is boil down. We went through all those metrics that pitchLogic delivers. It's really overwhelming for a lot of people to try to think about all of those at once, and often that's not the most effective way to use it. So what this is going to do is give you a few things that you might look at and target. Say your fastball has a 90% spin efficiency - maybe we're going to recommend trying to get up to 95. Maybe in some cases, especially with a sinker, we might actually recommend lowering that spin efficiency a little bit, which might help you get more run and sink to your pitch.

Lary: So there are actually some tips built into why you're using the pitch while you're seeing your score, that's interesting.

Steven: Absolutely. Yeah. And that's really how we want to help guide coaches and players to work together and figure out what are some adjustments I can make, what are some things I can experiment with to see if I can make this pitch even better?

Lary: Now without getting into the deep technical stuff, is there a simple way you can tell us how those scores are calculated?

Steven: It's based off of a machine learning algorithm, what some people call AI. But basically what those types of algorithms do is they look for patterns, and they can look for patterns even when they're not completely consistent all the time. So, for example, you can throw an excellent slider, and maybe it's less likely to get hit than a poor one. But every once in a while, a great pitch still gets hit for a home run. Every once in a while, a horrible pitch that you hang right over the plate is still a strike. What this is able to do is take hundreds and hundreds of thousands of pitches from real games, look at what got hit, what was a single, what was a double, and so on, and determine what types of pitches are most likely to be successful. And it's a combination of different things. We're looking at Arm Slot, we're looking at Movement, we're looking at Velo. And then we're also using pitchLogic data, all the trends from millions of pitches that we've collected and learning how those things connect to each other. So if I adjust my grip a little bit to try to adjust my Spin Direction, that's going to have an effect on my Spin Efficiency.

It's also going to change my Spin Rate a little bit. It's probably going to change my Velo a little bit. And all of those things that are going to combine together and change your Vertical Movement, change your Horizontal Movement. So we're able to untangle all of that using these really powerful algorithms and estimate how well your pitch would play in a game. Then we're able to scale that to different age groups. So if you're 14 years old, we're going to compare you to, sort of, normal baselines for your age group, instead of trying to compare you to an MLB player.

Lary: Well, that seems to make an awful lot of sense. Where did you collect the game data from?

Steven: So the game data is actually a publicly available resource from Major League games, so it's a huge pool of pitches with a really rich data set, and there are a lot of interesting things people have done with that data to try to gain an advantage, say, in their fantasy baseball league and things like that. So it's a pretty interesting resource for a lot of things.

Lary: Steven, how does this compare to what other platforms are doing and other systems that are out there now, and how did this all come about?

Steven: Absolutely. What we've really seen in consumer available technology is really people have gone down the same path that we went down up until now, which is delivering a ton of data, and then really offering pretty limited guidance as to what to do with it. So we think this is really unique in terms of something that you can put right in the hands of a player, or right in the hands of a coach who's not necessarily a Major League coach or a coach at a D1 college program, but there is some precedent for using similar technology. Most of your Major League organizations have built their own pitch scoring systems using the same basic concept, although there are a lot of different decisions to make about what you're going to prioritize and how you're going to go about it, as well as the most accessible system I know in terms of being available to people is Driveline's Stuff+ system. We actually worked a little bit with Driveline. Dan Aucoin, who is really accomplished in baseball and data science, helped build their Stuff+ system and offered us a lot of great guidance on some ideas of what we might do with our STUFFpL system, and we also worked with them to make sure that the two systems give similar feeling results.

So that score of 100 being an average pitch, that's the same way that Driveline is going to report their score. So they're sort of interoperable for people who might be using both platforms.

Lary: There's two different types. There's SKILLpL, and there's STUFFpL. STUFFpL adds velocity into it. In this day and age, when everybody's chasing mph, why is it valuable to know what the SKILL is without the velocity?

Steven: The first thing we wanted to do is really enable people to go out and experiment. Usually where people are playing with their pitches, trying new grips, trying to spin it a little bit differently and see what that does, is in catch play or flat ground. So you're not necessarily throwing at 100% effort. I'm not throwing my 95 miles an hour fastball if I'm a pro guy. I'm throwing 85 or 90, and I don't want to see that have a big impact on my score when I'm trying to play with things. Right. So the idea is with SKILLpL, we take velocity completely out of the equation. We let people experiment, and we let people try things that they think might improve the pitch and see how that affects it without having to constantly be giving 100% effort on every single throw. We showed that score on every single pitch while you're in the moment. Whereas STUFFpL is really meant to be a periodic assessment and sort of review tool - get a snapshot of where you are, set some goals for yourself, and then kind of set that aside for a little bit while you actually work on those individual goals.

Lary: And the fact of the matter is, is that most of your throwing is done not in a full bore bullpen type, "I'm airing it out every pitch". Most of the throwing that a pitcher does during the week outside of the game is done playing catch, and they can still work on things and improve their skills.

Steven: That's right. I was talking to a college pitching coach recently, and I think he said that maybe 10% of the throws that a pitcher makes during the week are from a mound.

Lary: Yeah. So you can really look at it: all it takes is a telephone or an iPad, connect by Bluetooth to the baseball, and you're ready to go. Any other set up, anything else involved with it?

Steven: There's really nothing else you need. You don't need a WiFi connection. You're able to use the camera on the phone to record video, but it's not required. So if you don't want to bother with that, you don't even have to do that. You really just need the ball, your phone. It can be either iPhone or Android, or an iPad, and you're good to go.

Lary: The other thing that's really interesting is how you can work on your Long Toss program, which is partly conditioning and partly, again, good mechanics. How does that work?

Steven: Yeah, that's been a really popular feature for us for a little while now. Being able to take Long Toss and not only do Long Toss in a normal outdoor setting, but bring it inside. You can throw the pitchLogic ball high into a net, and the app is actually going to tell you how far that would go, how far that would travel before it got back down to ground level, as well as some other stats, like what kind of hang time it would have, what angle you actually threw it on, what the velocity would be at the end point, at the catch point. And then all of the other data you'd normally get, like Spin Direction, Last Touch and what your grip was, Spin Efficiency. So you can really look for, not only how far am I throwing, but am I keeping my release consistent when I do this? Or am I just trying to heave it and cutting the ball and not having good Spin Efficiency and not keeping other things consistent when I'm throwing Long Toss.

Lary: For young guys who you want to teach to keep your arm in the same spot every time you throw the ball, and then you mature a little bit, and you start throwing multiple pitches, and you still want to keep your arm in the same kind of slot and you want to see what kind of different action you're getting on the ball: it's really useful. All the way up to the more educated and sophisticated user. They can see where their finger actually comes off the ball, and how it's turning. And you can also compare it to your different pitches in 3D.

Steven: That's right. And I have an interesting comment for you, Lary, on keeping your Arm Slot consistent. One of the things we found that's really interesting, is when you're, say you're throwing a change up and you want to get a lot of sink and run on that change up to separate it from your four seam fastball. It's really tempting to drop your arm a little bit because that's naturally going to create more of that motion. What we find in the actual data is that you can do that just a little bit. You can cheat your Arm Slot down a bit, and it doesn't hurt how effective the pitch is in game. We believe that's because you can do that a little bit before the batter starts to pick up on it. Either way, the data proves that it does work. But as you start to move even further away, all of a sudden the pitch becomes less effective. And our hypothesis, I'd say, is that batters start to be able to read that pitch and see that a change up is coming. And that's one of the unique things, actually, that's in the STUFFpL system that none of the other scoring systems that I know of - and there are some that are very secretive in the Major League organizations in particular - are able to take that into account. So we're able to actually tell you whether or not dropping your arm a little bit on that change up is going to be an effective thing for you.

Lary: So unless you're making a radical or drastic change from a high three quarters to a sidearm, hitters likely aren't to pick up on it.

Steven: Right, we stay away from making really radical changes to Arm Slots in general just because we don't think it's a good idea. Maybe a coach who is hands on and really working on your mechanics with you might be able to do that effectively, but the app will never make a huge change to Arm Slot. But there are a few degrees of separation between pitch types that you can get away with and can actually improve the pitch a little bit.

Lary: All right. And I know that we can get as scientific as we want because of that Virginia Tech background of yours as an engineer really comes into play. What's the data look like behind the scenes? I mean, you talked about the artificial intelligence and everything else. How deep can you really get into analyzing this if you wanted to take that deep, deep dive?

Steven: Well, first of all, since you mentioned Virginia Tech, I'll give a quick shout out to my Hokies about to host a Super Regional, so excited about that. But you really can get as deep into this data as you want. And what we found is there's a space in baseball, a lot of it's at the pro level, a little bit at the D1 level. Wake Forest - it's right down the street from us, really does an excellent job of using data and generating data, and they have a very sophisticated pitching laboratory where they're able to measure a lot of things and learn a lot of science about baseball. And it's fantastic that there are people doing that. I'm proud to be one of the people in baseball that's trying to learn more constantly, and I think a lot of people are, but at the end of the day, most players and coaches want to be able to do things that are going to make their game better now rather than learn and learn and learn, that's going to pay off ten years down the road. And frankly, if you're a player, that investment is a risky thing sometimes because ten years from now your career may be over. We're really trying to take that knowledge and put it into a form that's usable for everybody, rather than making each individual person go on that journey themselves.

Lary: It's always seemed to me that you can learn to improve yourself through more information and by using this kind of a product, this kind of an instrument, and waiting for the physical side to catch up. Because we aren't all born and accelerate our growth at the same pace as we go through those early years of baseball and growing. We see the six foot, 14-year-olds, and we see the 5' 3", 14-year-olds that turn into the 6' 3" stallions that they later on become, and so this is a great way to find out about yourself and how to pitch before you get the actual physicality of it.

Steven: That's right. And I mean, I would never say that a player shouldn't be working on their mechanics, shouldn't be working on how to build Velo and build strength and all those things. This is learning to spin the ball better and learning to do other things better. It is not instead of that, but it complements that when you build those things together. We do see that Velo is one of the things that will drive your STUFFpL score higher, but it's not the only thing. You can do things that might be as effective as gaining a mile an hour or 2 miles an hour that you can do much more quickly just by learning a slightly different grip or something along those lines.

Lary: The bottom line is that pitchLogic can be one of the tools that you use to get the sum that goes into being a winning ball player.

Steven: For sure. Your best athletes in every sport are never the people who just have one element of their game down. It's a complete game that really drives people to be excellent.

Lary: All right. Well, that's a lot of great information. I really appreciate the time. It's always good sitting down, talking with you. I always learn something, and it's good. Every day is a good day if you learn some baseball. Steven, thanks.

Steven: Absolutely. Thank you, Lary.