The MIT Sloan Sports Analytics Conference started small. Seven years ago it was nothing more than a hundred or so people gathering on the MIT campus to discuss the latest advanced stats from around the sporting world. Fast forward to now. The conference is hosted in the biggest rooms of the Boston Convention Center. It is attended by more than 2,800 people. The two days of panels, presentations, and discussions are filled with some of the biggest names in sports and business. Sloan has truly become the end-all, be-all of sports conferences.
Aside from the distinguished people here to speak, the really incredible thing about this conference is the professional diversity among the attendees. More than half of all NFL, NBA, and MLB teams have representatives here. Only one NBA franchise is unrepresented. I won’t tell you the name of the team not in attendance, but it rhymes with bakers. There are representatives from the media ranging from bigwigs like ESPN’s John Walsh all the way to humble basketball bloggers like myself. From the business world, there are sponsorship buyers from some of the largest brands in the world including PepsiCo and Proctor & Gamble. There are also a great number of business people here to learn about, pitch, or market new digital technologies designed to make sports teams more efficient and competitive both on and off the field. It’s very much sports nerd Comic Con. Over the next few days, I will try to come up for air enough to jot down some coherent thoughts to give an insight to what this conference is like and how all that is discussed here might affect the Phoenix Suns.
A few days before this conference began, a few of the THN bloggers began discussing the best non-player assets in the NBA. The crowds in Utah and Golden State were mentioned. The lack of a state sales tax in Texas and Florida were nominated as well. But one asset in particular was believed to reign supreme across all the writers involved, regardless of what team they cover: the Phoenix Suns’ training staff. With the Steve Nash gone and the team in full rebuilding mode, the training staff is the Suns’ most recognizable asset. It’s given the team a significant competitive advantage, especially when it comes to signing older veterans looking to recover from injury. But how long can the Suns maintain this advantage over other teams. What’s to stop Mark Cuban from giving Aaron Nelson $10 million to defect to Dallas? What if another team miraculously salvages an injured player’s career and begins passing themselves off as “Phoenix 2.0”? The truth is, Phoenix’s advantage in the area is more tenuous than anyone would like to admit, especially this year when everyone in the Valley of the Sun in searching for things to be positive about. For the Suns to continue being viewed as the league-leader in training, their methods will need to evolve.
On the first day of the conference, I attended an Evolution of Sport presentation entitled “Performance Engineering.” The presenter, Dr. Phillip Skiba, trains world-class endurance athletes with a combination of technology, physiology, and math. Using computer modeling, Dr. Skiba creates a unique individual performance capacity profiles for each athlete he works with. With sufficient training sessions, his proprietary software can accurately model an athlete’s fitness and fatigue levels over the course of a season. With that information, Dr. Skiba designs training and competition schedules that ensure his athletes are able to perform their best at competitions like the Iron Man World Championships and Olympics.
This modeling software and approach to training is completely applicable to team sports yet no one is doing work of this kind in any of the four major sports at this time. There is an opportunity for the Suns to cement and increase their advantage in athletic training. I spoke with Dr. Skiba after is presentation to get his thoughts on what type of impact his training methods can have with the Suns and basketball in general.
Valley of the Suns: How prevalent are the methods you’re using in team sports as it stands right now?
Dr. Skiba: It’s almost non-existent…It surprises the heck out of me really.
VotS: Well now that you’ve presented at Sloan, the secret’s out of the bag.
Skiba: (Laughs) Well, yeah, exactly. There’s no reason this can’t be applied to any team sport. And especially if you have a staff like the Suns have, and they’re really interested in taking these guys and repackaging them and resurrecting them, this [PhysFarm software] is exactly the tool. We’ve used this to rehab people from ACL injury. So we can look at them and say when they’re gonna be back in practice shape, when are they gonna be back in play shape.
Vots: What kind of physiological measurements would you take from a team sport athlete to build your model?
Skiba: It all depends on what you want to increase as your performance output. So if your performance output is how fast can they go on the fast break, we will use speed. We’d build a model based on that. If you’re more interested in vertical leap, we’d use a model based on that. The question is, “What are the factors most important for performance?” And then we build the model from there.
VotS: I would think in basketball that what people would want to do is push a player’s fatigue threshold as high as it will go. So a player can perform better for longer.
Skiba: Yeah that’s the interesting question. So what you can really do is ask the athlete, “When you’re tired, what’s tired? Is it your legs? Is it your arms?” And then we build on that basis. We look at their strength training program and how much running they’re doing. We look at how much court work they’re doing. And we adjust that to fit them accordingly.
VotS: Obviously this works from a macro level. You can look at the 82-game schedule and decide, based on your model, when to rest players and when to practice. On a micro level, could this work for minutes management?
Skiba: That’s exactly the idea. If you’re able to watch how much the athlete is running, you can decide when to sit them. Cause you’re going to know from practice and the training done to build the model, where is their critical point of fatigue where there battery goes to zero. So once you know that, you could have a computer running next to you on the bench that could tell you it’s time to sit a guy down. And because you know how quickly they recover, you know how long to sit them for.
VotS: So coaches could develop their own model based on this information instead of just randomly think a player is appearing tired?
Skiba: Exactly and that’s the beauty of this is that you can know. You don’t have to guess.
VotS: I think that kind of certainty is something that freaks coaches out.
Skiba: In my experience, working at the Olympic level, coaches get very upset by this because they say, “Well then you’re not going to need me anymore.” No they still need you. This model tells you the “when”. It doesn’t tell you the “how”. You have to know the how to train the athlete. This is just going to tell you when to do that.
If you’re interested, you can learn more about Dr. Skiba’s company PhysFarm, on their website.
Here’s a link to the second Dispatch from Sloan piece