2025 Franz Edelman Award finalist: USA Cycling

Ashley K:

Welcome to the latest in our special series of resoundingly Human podcasts, highlighting the finalist teams for the 2025 Franz Aleman Award. The Nobel Prize of Analytics, which will be awarded at the upcoming 2025 INFORMS Analytics plus conference in Indianapolis this April. These finalist projects represent teams from around the world who have leveraged advanced analytics to transform their organizations, address their most significant challenges and better serve their customers and communities. Today I’m joined by members of the team representing USA Cycling to discuss their finalist project in the lead up to the Franz Aleman competition. So to start, could you briefly introduce yourselves and the finalist project you’re being recognized for in this year’s Franz Edelman Award competition?

Ryan Cooper:

Sure, I’ll start. My name’s Ryan Cooper, obviously from USA cycling, I guess I am senior data and analyst, but I can consider myself a jack of all trades in terms of everything data and performance analytics related at USA cycling. And I’ve been with USA cycling for I’d say almost two years I guess now. So came on board in 2013 right before the Glasgow World Championship. And so when I came on, we didn’t really have a data team per se, so I was pretty nacient. So I came in to kind of do a top down overhaul almost of our entire entire data structure from strategic level all the way down to kind this tactical modeling level for different disciplines.

Jim Miller:

I’m Jim Miller, chief of Sports Performance. I’ve worked at USA cycling since 2001 24 years. I took a hiatus a couple years ago, but otherwise I’ve been here for 24 years. I’m primarily responsible for overseeing all of our national teams with all of our disciplines from Olympic Games down to Talent id.

Ashley K:

So analytics is at the core of the Franz Edelman Award. Could you share a moment in the project where you had to get especially creative with your analytical methods or tools?

Ryan Cooper:

Yeah, so I’ve been doing analytics or for quite a long time. So my undergrad was from University of Texas and electrical engineering. I worked in aerospace and defense for a long time and then went back for master’s and all but dissertation and PhD or at SMU. And one of the things I learned was that as I pivoted from that kind of industry into endured sports is that when you try to apply traditional analytics and operations research into the human area, into human and physiology, there’s a lot of places where you find that humans aren’t machines, they’re not widgets, they’re not mechanical pieces of equipment that just fit nicely in a box that you can just model perfectly. And so a lot of times when you’re looking at modeling, modeling a human, it’s quite a bit different. And so when you’re kind of building out these models or building out these data analytics, that’s one thing that you have to take into consideration is just how different every person is compared to the next person. So that was one thing that was really as we’re building out this project, is the individuality of each rider and each athlete is something that we really had to focus on.

Jim Miller:

I would say that the project itself was innovative when this team pursuit came in Olympic games in 2012, we were silver in 2012 against Great Britain in 2016. We were silver against Great Britain in 2020. We were bronze against New Zealand. Great Britain was silver, and our goal had been to win the, and we hadn’t been able to crack the code. And the very first thing we tasked Ryan with outside of our strategic metal expectancy was to build a performance model that would help us win a gold medal in Paris. What it ended up doing, Ryan did a huge amount of or on this in mathematics, but what it really did for the coaches was gave them a roadmap and boxes a check to ensure that they had all the parameters in place. Once we did that, then the model ran as they were expected and as long as the athletes performed as we believed that they could perform and we had all the checks, the boxes checked, then we were pretty confident we’re going to be really competitive.

Ashley K:

So the results of this project clearly made a significant impact for the US Women’s Cycling team. Could you share some behind the scenes insight on this at perhaps the 2024 Paris Olympics?

Jim Miller:

Well, two of our team pursuit athletes team pursuit is four riders that race 4K on a viome from a standing start time is taken on the third rider. So it is absolutely a team of event. Two of our riders raced the road race two days before and one of ’em crashed. So part of this human element of modeling is things happen and you can’t account for that. You can plan the best, you can plan, but then you get these wrenches all the time. I think our biggest challenge was one, getting this athlete healthy. We had about two days to do it. She wasn’t significantly hurt, but injuries or when you hit the ground, it can take a couple days for you to feel normal or good. Again, we didn’t have a couple days. The other challenge was we didn’t have as much time training in the line or the team pursuit as you would normally want to train or you would expect to train or you would hope you would have three of the four athletes race on professional road teams. So we were grabbing time and doing training camps for two years whenever we could. And we probably only had all four riders together maybe two times in 2024. So we were really reliant on the model and what we thought we could do if we got all the parameters, we didn’t see it, we didn’t have a time with the team to really see it come together.

Ryan Cooper:

And so I remember, I think it was the morning of the qualifying round. So the first round, the way this works is the first round is the team writing solo for time on the track and you don’t really know, so you have the environmental conditions, and I’m texting Gem. I was like, okay, what’s the air pressure, the air density on the track and he’s coming back at 1.14 or 1.142 or something. And so plugging that into the model and running it and coming back and saying, okay, with the pull strategy that we’re going for, meaning how many half laps is each rider going to be in front from the beginning? So we know the lineup, we know the order, we know what the strategy’s going to be, and we say, okay, here’s the time that we’re going to expect. And so I think we came back with a 4 0 5 3 or 4 0 5 4 or something like that, and I think the reaction was still kind of a disbelief.

It was still kind like, well, I don’t know. I don’t if that that’s going to be what we’re going to hit. And the time they get out on the track and the time comes back 4 0 5 2 and you get this spark and you go, okay, they knew on paper, they knew in the model this should be a time that we can hit because the metrics are there, the numbers are there, even despite the crash, we know all the physical traits and the characteristics are there with the strategy, with the conditions on the track, with all the rider interactions, we kind of know that this is possible, but then you see it and now the sparks really gone off and then qualified with the second fastest time still New Zealand was still had a faster time at that point, but just marginally and when you compared that to where we were in the end of Glasgow, so that rewind a year before go back to 2023, there was a time where we raced head to head against New Zealand and Glasgow in the second round.

So in the second round, you race head to head against another team based on your qualifying time. There was a time in Glasgow where New Zealand was four seconds up at the halfway mark. And so when you think about a race that’s won by fractions of a second, they were up by four seconds at the halfway point and now you go fast forward a year Paris in the qualifying match and we’re down by fractions and fractions of a second in the qualifying round to New Zealand. And so now we have this confidence built up. Okay, 4 0 5 project 4 0 5 is what we set out to do. That was the number that we had in our head. Now we’re 0.2 seconds off of that number. Right. Yeah, it was super cool, super cool feeling for sure.

Ashley K:

In your experience, what makes for a great team when working on a high stakes analytics project? How did your team foster collaboration throughout this entire process?

Ryan Cooper:

Well, yeah, it’s interesting For me it’s like Jim and I met when he took a little hiatus from, he mentioned this earlier from USA cycling, he took a couple year, maybe not even a full two year hiatus, but maybe just a year and a half, two years. And he came to a company called Training Peaks, and I was working at Training Peaks at the time as a chief scientist at Training Peaks In there. He saw I think kind of the power of a lot of this data behind the scenes and what we were doing with modeling and what we could do with kind of performance, performance modeling in the background. And I think part of that really helped solidify in his mind this concept of tech and innovation and performance. And so when he went back to USA cycling, there was definitely this knowledge of this needs to be a place that we adopt when we start to bring more and more of this in.

And so I actually called him and said, Hey, you need to hire me or something like that. He tells the story better than I do, and so we kind generated this. He said, here’s what I need. And I said, okay, I think we can do it. And so I came on board and I think from the top down, you need to have this culture that is really pushing on maximizing ROI on force multipliers on, we’ll talk about this in the presentation of we are in a different funding environment than a lot of these other high performance teams from other countries. And so we have to be very tactical with how we allocate budgets. And so tech and innovation and use of analytics, use of modeling, use of strategic forecasting for where we know we can actually move the needle if we do spend resources somewhere is super important for us. And so yeah, I think this culture that comes from the top down from the chief of sports performance, CEO buys into it and then coaches buy into it and once they see one project and they see the benefit of that, then it proliferates. At least that’s from what I’ve seen from my point of view anyway, I dunno if Jim has a different opinion.

Jim Miller:

I think that’s right. I think one of the things that we insist happens in our teams is this open-mindedness to learning and adopting new strategies and tactics and innovations and technology in our sport. We have so much data that you can collect. It’s not like a lot of sports where they’re just getting into managing and understanding load. We’ve understood load on an athlete for 20 years so we can look at everything. I mean, there’s not a day that a minute that goes by in a day that if we wanted we could not be collecting data off an athlete. So we have a lot of information to work with. But as a coach, if you don’t grasp that and you’re not willing to look at something different than how you’ve been doing it or what, it’s like a recipe for failure. So I think the one thing we push in our teams all the time is you have to be open-minded to this. You have to be open-minded to technology, to innovation, and you can’t be afraid to take a chance.

Ryan Cooper:

What I thought was interesting that we did this year is during one of the training camps, we sat down with this specific model and we said, we sat down with every athlete and we showed them how the model works. We showed them all their data, we showed how we came up with what a metal capable set of data looked like in terms of these metrics that we’re trying to achieve. We showed them what a gold medal capable athlete looked like, and then we showed them the delta from where they were to where these two types of athletes were and how those it impacted the race. So you could say, okay, if we could get your numbers from X to Y, here’s what type of time difference that could mean. And so suddenly now you’re saying, not only is it just something on a piece of paper, but we’re showing you the real impact of those things.

And so now you’re giving them a blueprint or a map and says, okay, you’re here today, but if you can just work on these little things, position for aerodynamics or high-end interval work to get your energy systems where they need to be, then that can mean 10 seconds or 15 seconds to the overall team. And so now you can say, here’s how your individual work is contributing to the overall good of everybody. And so doing those individual sessions with both the athlete and or their coach and then also bringing everybody together to show that as a whole group was I think hugely powerful to bringing the team together and giving ’em that north star to go after.

Jim Miller:

Also really interesting that process was this team collectively hadn’t ridden faster than four 11. We were telling them that we thought we would be required to ride a 4 0 5 to win, and then showed them that this is how we thought could do it. And I think to some degree in the back of their minds, they were like, wow, we’re so far off. That’s not possible. This can’t be possible. However, individually, if you went to their parameters, they were missing just one or two boxes that weren’t checked. So within our roadmap, then individually we’re like, okay, how do we check those boxes? If this is the amount of power you need to generate when you’re on the front, this is the amount of power you need to have in relief. This is the amount of anaerobic power or functional reserve capacity that you have to have by race day. It gave them individual things that they could work towards to start checking the boxes, aerodynamics on coefficients of drag. We spent two and a half years on aero testing and wind tunnel and Silverstone, great Britain, and once they saw all the boxes, I think to a degree they were like, okay, it’s not possible, but we have time to start working on these individual parameters. So if we do, then maybe Ryan’s right.

Ryan Cooper:

It was nice to be right.

Jim Miller:

Being right is always great.

Ashley K:

In your opinion, how do you see analytics evolving in your industry over the next five to 10 years, and how do you plan to stay ahead of the curve?

Jim Miller:

That is a really good question. I would say in ours four, we’ve always had four scientists. We’ve had physiologists, nutritionists that are working on human forms, human performance biomechanics, but those professionals would primarily, I think, be hobbyists in engineering and technology, statistics, analytics, those disciplines I guess if you will. But I think we’re getting to a point where we’re a little bit like F1 where everything needs a real engineer. You need educated professionals to continue to drive the sport forward. And as each team adds one or is starting to embrace ai, starting to embrace these deep analytics, deep math models, you really either have to join and start adding those professionals yourself or they’ll just start getting beat. You can’t take a mechanic out of your local garage, put ’em in an F1 car team and expect that that team wins. But that team is made up of highly educated engineers and mathematicians, and that’s what’s required to win at F1. Now, cycling, as odd as it is and as dynamic as it appears to be, it’s really not as dynamic as we all want it to be, and you really have to have these professionals now to be competitive.

Ryan Cooper:

I look at it, there used to be these low hanging fruit. If even if you go back to what, 2012 or 2016, probably aerodynamics, you could go out there and you could win. If you went to the wind tunnel in certain events, you could say, okay, we went to the wind tunnel, we’ve done a few little things and we’ve optimized our aerodynamics, and so we’ve gotten 30 seconds out of that or whatever on a time trial or something like that. And so you could go in and spend a little bit of time and get huge benefits out of that over some teams that just didn’t have access or didn’t think to do these kind of things. That’s all table stakes now. So everybody across every discipline, we joked about this. We were in a meeting out in Palo Alto actually last week talking to some of these companies that may be the future of cycling.

And we were joking that even BMX race, which is a very, very short race, this is like a 32nd race, and so you’re very, very short. You’re almost no pedal strokes whatsoever. You’re coming off a ramp, you’re doing a couple little jumps, and then you’re pedaling really hard, and then you’re carrying a lot of momentum through the aerodynamic innovations in that sport, which is traditionally just not even aerodynamics isn’t really thought of as kind of this backyard sport. Everybody was doing it. It’s table stakes now. And so now going forward, we have to look at what’s next. And so this is what’s going to be next, putting all of these things together, so the physiology, the energy systems, nutrition, all of these things that are in these little kind of silos now, you’re going to be modeling all these together and kind of bring ’em into this cohesive, when is the best time, when’s the optimal time to take the right nutrition on a long climb? When’s the right time to maybe take a breakaway or try and chase something down for a road race? What tweaks do we need to make on bicycles and BMX race? What’s the right trick selection, BMX freestyle? So all of these kind of things are going to come from analytics.

Ashley K:

Before we wrap up, is there any final thought or story you’d like to share that encapsulates the spirit of your team and your journey to the Franz Aleman Award competition?

Jim Miller:

I would say it would be being a bit of an underdog. We’re significantly underfunded in our sport compared to our competitors. We definitely hit above our weight. We win medals on less money than our competitors do. And then the Franz Edelman contest where most of these companies have teams and teams and teams of data engineers. We had Ryan, we had one guy. I think what really encapsulates this project and what we accomplished was in all fronts, we were the underdog and in all fronts, we over achieved. We’ve over succeeded and punched above our weight. Brian’s a team of 10 by himself, Ben.

Ryan Cooper:

Yeah, no, I think that’s great. The ultimate underdog story.

Ashley K:

Well, thank you both so much for taking time to share this special insight into your finalist project. I wish you the very best of luck in the Franz Elman Award competition, and I look forward to meeting you in person in Indianapolis for the 2025 informs Analytics+ Conference. If you’d like to learn more about today’s episode and guest, visit resoundingly human.com and check out our show notes. The podcast is also available for streaming and download on Amazon Music, apple Podcasts, Google Podcasts, and Spotify. Wherever you listen, please be sure to leave a five star review to help others find and enjoy the podcast. Until next time, I’m Ashley K and this is Resoundingly Human.