It was an honor to be interviewed on Wharton Moneyball, which airs on Business Radio Powered by the Wharton School on Sirius XM. Wharton is the business school of the University of Pennsylvania, and has been ranked number one by US News & World Report every single year since inception. Specializing in Finance, The New York Times has deemed Wharton's undergraduate population as "the closest thing that exists to a Wall Street farm team."
Hosted by Wharton professors, their show explains how decision makers in the game can avoid the common mistakes and embrace the data. For those who missed it, here is the transcript of the interview.
WM: Well today we have Phil Wagner; he is a MD, and founder of Sparta Science out in the Silicon Valley. He was a strength and conditioning coach at UCLA and CAL and went to medical school at USC. Tell us about where this (Sparta Science) came from. You did some strength and conditioning work and you saw something that lead you to go to medical school. This is not a common direction so can you tell us about that?
PW: I started off my career after undergraduate working as a strength and conditioning coach and the nature of the industry, particularly football, was if you don’t win you're out. So after a couple of experiences at schools where we didn’t win on the football field, our staff got let go. So I started to think there has to be some way to validate what we are doing outside of wins and losses.
WM: Especially since you are not in charge of the team. Something that we believe is that an organization should be judged not strictly by outcome but by process.
PW: It takes time to develop good processes. To throw out the whole staff or process just means you are just starting over again. The pressure to develop processes and staff that are hopefully going to improve the outcome are what lead me to go to medical school, to see how the process of treating patients is handled. Everything from evidence-based treatments, even to morbidity and mortality conferences where you have to defend that outcome using research and evidence based protocols. It is very similar to how you would respond to a loss as a sports team.
WM: If only that were the case. I love this idea of mortality conference. After every win and loss, lets have the morbidity conference and see how we all do. Thinking of sports teams as patients and say we lost this year, what happened. Man that would be a sophisticated organization if that were to happen. Ok so you have the m&m conferences as you learn about treating patients, then what happens?
PW: Then the next step is you start to develop the concept, but how do you organize that into something that is tangible. So I moved back to the Silicon Valley and met a software engineer before we started a company. We then created a platform where data could be gather and provided as a solution for teams to make those decisions and have those discussions.
WM: One of the ideas you had was to be a supplier and a consultant.
PW: Yeah, to make the largest difference we have to be integrated in multiple organizations, sports, and countries. We can then take that data and start identifying those meaningful trends and patterns. Having anonymous access across those all organizations is much more powerful than choosing one or two to focus on.
WM: That makes a tone of sense. So how much can you tell us about what you do, tell us what Sparta Science does.
PW: We have a training center where we actually train individual athletes and that’s where a lot of the initial data was collected, and after that we launched our software product, working with other organizations. Much of the major processes center around force plates to identify meaningful trends and movement patterns (training exercises), followed by linking that with injury and sports performance.
WM: What’s a force place?
PW: It’s one of the most validated metric of biomechanics and measures how much force you put into the ground; left versus right, heel versus toe, and where is that pressure. Ultimately, with the exception of water polo, most injuries and performance is dictated by that ground interaction with your feet. Measuring that allows the prediction to look at both injury causes and explaining the graceful or not so graceful sport movements.
WM: So you bring a person in and put them on a force plate and have them what….. jump?
PW: Yeah, so they will do a series of jumps. Over the years developed in the protocol, we have identified the parts of the jump that are important, and which parts to throw out and that’s added to a database and compared to provide specific values, t scores, of looking where they lie within that sample. Then the numbers are meaningful rather than looking at just units (newton’s per second, N/sec,). The next question is what does that value mean, is it good or bad?
WM: So you have this Sparta science center, and there you train individual athletes, you found you could collect more organized data by introducing the force plates that would record in a very scientific way, data that’s correlated to things you want to know about the athlete. From that you built a way to calibrate measurements to tell you something meaningful. No one knows what to do with N/sec so you made a construct that allows you to relate each athlete’s performance to other athletes to tell you something.
WM: To me what it sounds like is you are collecting a bunch of predictor variables and looking at what the outcome is. What are you trying to predict with all this stuff?
PW: Well its two fold, injury prediction and performance abilities. Several NFL and NBA organizations use it as a draft analysis tool. They look at players and decide what are their physical upsides or likelihood of being successful at one position over another. Often health and performance are closed off as two separate entities but the reality is it’s all about movement efficiency, what allows you to perform better often allows you to stay healthy longer.
WM: Staying with the force plate technology, how much of it is a given within an athlete, so used as an assessment tool, and how much of it is a development thing, you can actually move someone around once you know what they are doing and actually change what they are doing? How much is fixed in an athlete and how much can you change in an athlete and develop them?
PW: There are certainly innate characteristics that are hard to change within an individual. We have also noticed strong ethnic correlations between athletes and how they perform certain movements. All the variables are malleable and plastic and what really determines how much you can improve it where you are starting at, or your training age. So if you are starting at a value, a t-score, of 40 you have a significant upside to change, but on the other side of that bell curve, t-scores in the 70’s, a lot of the time the outcome goal is to stay there for as long as you can to extend that career or that ability to stay healthy.
WM: We know you have some case studies on your website, do you have any examples that you are particularly proud of or that you think are interesting to you or teams that use this technology?
PW: Yeah, one of the things that put us on the map is a little over 3 years ago we got a call from a kid who said he is “sleeping on my parents couch in the Bay Area, the Warriors just released me, and I am trying to make it in the NBA, I went to Harvard, these guys are not really giving me a look, I am Asian, and oh yeah there is an NBA lock-out”. So we analyzed Jeremy Lin and were able to figure out what he needed, which was very different than your typical NBA player. With the NBA lock-out it was kind of a perfect storm, where he had extra months to train. When a lot of the other guys spent that time in Vegas, he got to practice basketball skills more and train in a way that was very efficient because we gave him exactly what he needed based off assessments. It ended up being a pretty big breakout year for him, and changed the trajectory of where he was going.
WM: So without giving away anything that’s too proprietary, can you give us an example of what kind of training you might have prescribed for him based on what you saw versus what you might have prescribed for a typical NBA player, because you said it was different for Linn.
PW: I think, a lot of it has to do with ethnicity in this case. African Americans tend to be more explosive but also lack a lot of the mobility to stay healthy where as what we see with the Asian population is that they tend to be more mobile but lack a lot of that explosiveness and ability to be quick and powerful.
WM: This is talking about professional athletes correct?
PW: Yes, that is a big part of the normative values, is that it is part of an elite group of college and professional athletes.
PW: So with Jeremy we did a lot of heavy lifts, which is rare for a typical NBA player. He was pretty surprised when he learned we were going to squat heavy and jump a lot, because that is generally not normal when working with NBA players. Doing that pretty consistently with Jeremy made some pretty significant changes in his physicality that were more performance rather than health based.
WM: Are any sports especially interesting to you or sports that you think your technology is especially important to?
PW: You know I think there is value to every sport and there have been interesting lessons along the way. With my background in strength and conditioning I never thought about the draft analysis side. However in the NFL where there careers are so short and the draft is so meaningful, seeing it evolve there has been very interesting.
WM: Do you envision a day, like 10 years from now, where force plates are a part of the combine?
PW: That was a big reason that we decided to partner with the Atlanta Falcons, their general manager, Thomas Dimitroff; that is his vision. He sees the large value in that, being able to provide that to the combine and the NFL in general. We never foresaw that being of large value, but it has been, and in a sport that is so career length based it has been a huge value.
WM: That is an interesting issue because teams are always looking for an edge, and the NFL draft is as competitive of a place as you can find, and to hear Thomas say he wants to make this available to everybody is interesting because you might have thought he would want to lock it up for himself.
PW: Well he did. He locked it up for himself first. Thomas is a savvy general manager because he knows what he wants to identify in movement patterns and certainly wants to give his team an advantage. But, he is also a voice for the NFL and wants to see it be used to help the game, in addition to helping the Atlanta Falcons. From my standpoint the next interesting frontier is over in the UK with the premier league soccer and the academies.
WM: It’s a shame you couldn’t find a place with a lot of money to spend.
PW: I certainly didn’t know the revenue of these places, and the emphasis they place on these academies that lead up to the big leagues.
WM: Do you see that these utilities have any use in the youth age? You could certainly see this as an early assessment diagnostic tool. So you mentioned using this in the combine, is there anything that you measure now in the combine that forecasts what you are measuring. Is there any correlation or is it truly a set of uncorrelated set of observations?
PW: With the combine one of the challenging things is to try and correlate those values, a lot of those metrics are still a little bit lower on the objectivity scale, how well you can touch a peg above you on a vertical jump really has more to do with timing and skill as apposed to physicality. So a lot of the assessments are tough to correlate because the force plate assessment we use, what we call the signature, is very physical and has a low learning effect and a low skill level. Running a 40-yard dash is a whole business in and of itself, it’s such a skill that you have to teach how to run it, so it’s hard to make a lot of those correlations because a lot of those are very skill based.
WM: So this is something that you got, but you can alter it. Essentially that’s your hypothesis, not only can you measure it, but you can alter it in the right direction. That is what you were arguing with Jeremy Lin, you took his measurements, his signature, and you realized that there is a deficiency and if he could remedy that deficiency with heavy weight training, which is essentially quite counter to what most basketball players are doing (well because they already have that power) and he didn’t. What you are arguing with NFL players is that running the 40 yard dash is more skill at doing all those pieces rather than pure physicality.
WM: So Phil you spent some time over in the UK, can you tell us what you were doing with the soccer players, what was your experience like over there, what kinds of things were you doing?
PW: A big part was aligning that development of the academy players to that first team and identifying what their challenges were. It was a big learning process for me, because much of this is business based, there is a lot of restriction placed on salary cap for a club. So one of the things a club is doing is developing players and eventually selling them, that’s how they are generating revenue to continue to improve and reinvest in their organization. Having those benchmarks along the way of which players are likely to develop more and which might need a little less emphasis or priority to develop. So the biggest value over there was to evaluate those players and develop them knowing the likelihood that they are going to be at a certain benchmark next year or the year after.
WM: What kind of response do you get from the players when you show up with your force plates and your technology and you say give me your best. How do they respond, are they curious, do they want to see how they stand?
PW: That’s actually the largest buy in, the players themselves. I think the reason why is this younger generation grew up with transparency. They grew up with the internet where if they want to know or see something they just Google it. So this technology, part of what is revolutionary about it is it doesn’t hide the results from the athlete, very much like medicine. A lot of the time in our industry, strength and conditioning values are based on what the staff member deems appropriate. This is much more like medicine, if you are assessed then you can see it all, it’s your information. The athlete really appreciates that, and I think in a way it builds trust between the coach and the organization and the athlete because a lot of the time the outcome is not good. Therefore you are vulnerable, your naked, as the staff member when you say well here is where you are at, it’s not what we would like, but here is what we are going to do next.
WM: Right, so you are able to give them one piece, being the assessment but the other piece is development. Regardless of the assessment piece you can say well here is your development plan essentially based on what we see.
PW: That’s right, very much like a physician and a patient in that everything is transparent, you are presenting the facts and you are presenting the proposed treatment plan and it’s all there. I think athletes really latch onto that ability to see everything about what they are going to be doing.
WM: But Phil there is a degree of trust because it’s not like it’s something that they readily understand on their own, these are measurements in your signature that have to be interpreted for the athlete and how do you get buy in on that.
PW: Yeah, so that’s tends to be the more challenging part. We work with coaches to become experts on it so that they can be empowered to explain it to the athlete. That’s the majority of our business, and what I was doing over in the UK, educating front office staff and coaches to explain it in a way that empowers them as the expert and then tell it to the athlete. For example here is your eccentric rate of force and explain how that relates to knee injury and what the plan is to continue to improve that or prevent it from going down in the season.
WM: Now you talk about the injury predictions, in our conversations you talk about this for baseball players and particularly for pitchers. Can you explain to all of us how force plates, which you jump on, can be connected to upper body injuries in pitchers?
PW: We actually just had a paper accepted linking the predictive ability between the vertical jump and UCL injuries (Tommy John). We have the validation behind why that is happening and that’s where the hypothesis begins. The thought is if you are unable to generate initial force through your lower body then that force generation and transference has to happen further up the chain. So if you think about a pitcher, rather than loading down, and transferring that forward toward home plate, you’re missing that loading down portion and right away transferring that momentum and force toward home plate.
WM: So if you forced a guy to throw from a chair, he is going to blow out his elbow a lot faster.
WM: So you’re saying that a pitcher that is at risk for tearing their UCL is using their elbow too much and not enough of their lower body?
PW: That’s right. We see that in a few cases, one if there is not a large background in vertical strength training to teach the athlete how to use it, we also see it in taller athletes because it is more difficult to load down the taller you are and finally we see it when guys get tired, when you get tired you stop loading down and you start using momentum and falling over after you start that throwing motion. That’s when we start to see that force plate signature change to a predictive risk for UCL injuries.
WM: That is absolutely fascinating.
We have an email question from Rob: He says that you are talking about mostly individually tailored programs, that’s the nature of sport science, that’s one of the advantages sport science offers, yet there are some situations, say high school sports where you don’t quite have the resources for individual training. Are there team-based instructions that come out of your technology insights?
PW: I think that this exists also at the college and pro level, you can’t over individualize, because you lose that camaraderie of the team training together. There has to be a little bit of that team element, whether it’s due to resources at the high school level or at the pro or college level, making sure guys are sweating together in similar paths. So it really comes down to reducing the number of paths for that certain situation, rather than having 12 different programs, maybe there are only 2 or 3. These can be identified off your needs, so there is a strength group and a flexibly group. So Jeremy Lin is going in your strength group and then other NBA guys, mainly African Americas are going in the flexibility group. It could be individualized to the n-th degree, but that’s more of a happy medium along that route.
WM: Is it the case that your training and technology suggest certain types of training that is counter to what has been suggested by other people in the previous era? Or should we all go back to the old school Olympic lift type of stuff?
PW: The reality is there is value in everything; it is just a matter of who gets what. So I think that we are not suggesting any drastic new exercises, what we are suggesting is that people need something different. While Olympic lifting is good for one, yoga is good for another. So its how you prioritize and use analytics and technology to be sure who needs what.
WM: To wrap on Rob’s question, your solution is to just come up with 2 or 3 different tracks and then use the assessment to tell each individual athlete what track they are going to be on. It’s more like clustering.
WM: We really appreciate your time Phil, hope you come back and join us down the road.
PW: Yeah I appreciate it guys, thank you.