Why F1 Mathematicians Should Be Paid More Than Drivers

On the Formula One circuit, tenths and even hundredths of seconds can be the difference between podium glory and being out of a job. Races are won and lost thanks to the skills of the drivers and the strength of the car, it is true. But as the sheer volume of data available to F1 teams increases, another group of individuals have become key contributors to a team’s success: data analysts. Analysts run countless simulations, incorporating every possible variable, to inform their drivers’ race strategy on Sunday and achieve maximum performance for the driver and the race car. But while salaries and sponsorships can push a driver’s annual income into the eight-figure range, the mathematical brains in the background make just a fraction of that. If the person analyzing the numbers and making decisions about race strategy is just as important as the person steering the wheel, shouldn’t they also be reaping the financial windfall? James Allen certainly thinks so. The president of Motorsport Network, James has covered Formula One as a journalist for over 30 years and has seen firsthand the sport’s data-driven revolution. We asked James to defend his position.

Transcript

Ben Shields: The man was known as the “Rain Master,” or the “Regenmeister” in his native German. Formula One driver Michael Schumacher excelled in all weather conditions, but it was his ability to race in the rain that turned him into a legend. In 1996, coming off back-to-back driver’s championships for the Benetton team, Schumacher made a stunning move to Ferrari. But early in the season, he was stalling, having been forced to drop out of three of the first six races. A torrential downpour at the Spanish Grand Prix wouldn’t seem like the venue for most drivers to break out of a slump, but Schumi managed to outpace the field by 45 seconds, while only five other cars even finished the race. It was that kind of skill that led Schumacher to five more driver’s championships, a record number of wins, and over a billion dollars earned in his racing career.

Paul Michelman: But was it all skill? The Rain Master may well have never reached those lofty heights without the help of two men working behind the scenes: Ross Brawn and Rory Byrne. With Brawn as technical director and Byrne designing the cars for Benetton and then for Ferrari, Schumacher was not the only beneficiary. Teammates like Johnny Herbert, Eddie Irvine, and Rubens Barrichello also had fast cars to drive and clever strategies to execute. The result was six Constructors’ Championships for overall team performance for Ferrari, not to mention one for Benetton. Brawn and Byrne might not have achieved the fortune and glory of the Rain Master, but their reputation as the masterminds behind his success wasn’t a bad consolation. I’m Paul Michelman.

Ben Shields: I’m Ben Shields, and this is Counterpoints, the sports analytics podcast from MIT Sloan Management Review. In this episode, just how valuable are Formula One drivers? And could their worth be plummeting in the age of analytics?

Paul Michelman: In Counterpoints, we look beyond the data in search of what the data reveals, or supposedly reveals, about what’s actually happening both on the field and off. In each episode, we put one analytics-based hypothesis or question to the test and see how well it stands up.

Ben Shields: Today’s hypothesis: In Formula One, mathematicians should be paid more than drivers.

Paul Michelman: On the circuit, tenths, even hundredths of seconds can be the difference between podium glory and being out of a job. Races are won and lost thanks to the skills of the drivers and the strengths of the car, yes. But as the sheer volume of data available to F1 teams increases, another group of individuals have become key contributors to a team’s success: data analysts. Analysts run countless simulations, incorporating every possible variable, to inform their drivers’ race strategy on a Sunday and achieve maximum performance for the driver and the race car. But while salaries and sponsorships can push a driver’s annual income into the eight-figure range, the mathematical brains in the background make just a fraction of that. If the person analyzing the numbers and making decisions about race strategy is just as important as the person steering the wheel, shouldn’t they also be reaping the financial windfall?

Ben Shields: James Allen certainly thinks so. The president of Motor Sport Network, James has covered Formula One as a journalist for over 30 years and has seen firsthand the sport’s data-driven revolution. I asked James to defend his position. All right, it’s my pleasure to welcome on to the show James Allen. We have a true expert on the line with us today. James, thanks so much for joining us.

James Allen: It’s a great pleasure. Nice to be here. Thanks for having me on.

Ben Shields: We’re going to get into your fascinating thesis that we’re going to discuss here in a little bit, but first I want to try and set some context. And we definitely will have some Formula One fans as part of our audience, but we may have some audience members that don’t know too much about the sport. So I was wondering at the start could you help us get a sense of what data analysts actually do for Formula One teams?

James Allen: So, it’s an interesting story. I mean, I guess it’s like the world at large that we inhabit. Twenty years ago, the amount of data that was being processed was a lot less than it is today. It started to come into Formula One around the time of Ayrton Senna and Alain Prost, and those great races that we had in the late ’80s and early ’90s where they were able to start measuring the performance of the car out on the circuit. They had more sensors coming onto the cars, measuring things like the temperatures and wheel speeds and pressures.

And then as with the outside world in general — obviously it’s a technology-driven sport — the level of data that is available has just got greater and greater coming from the cars. They can measure it in real time. Initially, it was just in the pit garages at the side of the racetrack. But as the years have gone on, using telemetry — very much similar to what I guess was sending back information from the Apollo space missions back to Houston — but a much, much more sophisticated version of it, measuring everything in real time back in the race factory. So a team could be racing out on the track in Melbourne, Australia, and it’s being monitored in real time — all the parameters of the car’s performance, thousands and thousands of miles away back in Europe.

What also they began to explore more [were] the variables, using the lap times that the cars were setting — 20 cars out on the racetrack, all doing various different speeds with different kinds of tire compounds on the cars and different performance variables, and they could start to model those things to actually try and work out what was going to happen in the race: who was going to come out on top; when people might want to make a pit stop for fresh tires; what kind of performance boost that will give them, etc. And it just got more and more sophisticated over the years. So, it’s its own cottage industry, it’s its own world that’s going on behind the scenes. As these highly paid sportsmen are driving the race cars out on track, you’ve got these brilliant mathematicians in the background who are having their own race.

Ben Shields: That leads us very nicely into the thesis that we’re going to explore with you today, James. And let’s get into it. It’s a bold one. I can already, I think, hear some of the drivers lining up outside your office to come talk to you about this particular thesis. And you’re here to say that Formula One data analysts should be paid more than drivers. So, what’s the reasoning behind this thesis?

James Allen: Well, Formula One cars are obviously very complex machines, and one of the questions that you always get asked is, how much is the driver and how much of it is the car? Because if you took the champion driver of the moment, which is Lewis Hamilton, a five-time Formula One champion, and you put him in a car from a back-of-the-grid or middle-grid team like Sauber for example, or Williams, would he still win races? And would he still win the world championship? And the answer is no. He’d probably do slightly better than the current drivers they’ve got, but he would not be winning races and world championships. So clearly you need the car.

But drivers do make a difference, particularly when it rains or in variable conditions. And in certain types of racetracks where there’s a big confidence element, the human side kicks in and the judgment of the skill. But above all that, the definition of whether someone wins or loses is more often than not decided by the decisions that get made around the race. So, like I was referencing earlier on, what tire to start? There’s three different compounds of tire, different levels of softness that you can have. But the softer the tires, the less long-lasting they are. And you know if you have to cover a 200-mile race — 70-odd laps, for example, of a racetrack — you have to figure out what’s the fastest way from the start to the finish. And the drivers are pretty dependent on the mathematicians who are figuring out what’s the best way to do this.

If you start at the front of the grid, as Hamilton often does, it can be quite straightforward, but there’s still quite a lot of jeopardy. You can still throw it away by making a bad decision, and we’ve seen that many times before. Fernando Alonso lost, with Ferrari, the world championship in Abu Dhabi in 2010, because they made a decision. When he had the race under control, they made a decision based on trying to cover one of their competitors in the race, but they completely forgot to realize that another, much slower car had already made a stop and was going to be blocking Alonso when he came out of the pits after his pit stop. And the whole world championship went up in smoke, because Alonso couldn’t get past this car. So literally, world championships, let alone Grand Prix, have been decided on. So these guys should be paid a fortune, because they decide more race outcomes than the drivers do.

Ben Shields: It’s interesting that you make that point, because I want to bring in the business element here, and maybe this is a little bit tongue in cheek, but you do think about the amount of sponsorship attention and dollars that the drivers attract. We don’t see any celebrity mathematicians running around here, do we?

James Allen: No. Well, no question at all that the guys who are crunching the data and doing this kind of strategy, using all these amazing techniques like Monte Carlo analysis and game theory, they provide tremendous value for [the] money, I think, for what they do. And what we find more and more is that companies involved with Formula One want to find out more about this, and they want to see what they can learn about themselves and about their businesses from the way that Formula One teams go about this mathematical analysis, this data analysis. Because essentially this is change management in real time.

And so, a lot of businesses are fascinated by that. They’re fascinated by how they move much more slowly than Formula One teams. And the race outcomes and the evolution of the way the races develop, obviously, is much faster than what happens in a normal business. But this is something that Formula One excels at, and you do find a lot of sponsors and businesses really drilling down into these practices, the way the data analysts, the mathematicians behind the scenes manage these processes. And they take a lot from it.

Ben Shields: That’s an interesting point, and I guess maybe in a few years’ time we’ll start to see a few celebrity mathematicians walking around the paddock. Now, you mentioned the car earlier. I do want to talk a little bit about the engineers and the technical team behind the car, because you are advancing an increasingly persuasive argument about the role of the mathematicians. But wouldn’t you say that the engineers or the technical team deserve the biggest paycheck? Because in the end, it’s about having the fastest car.

James Allen: Possibly, but there’s so many of them. I guess the guy at the top of the pyramid maybe deserves the biggest paycheck, but I mean, a well-funded team fighting at the front of the grid will have over 90 engineers in the R&D and design department, sometimes over 100 of them, all beavering away, trying to find microscopic, incremental gains that will make the car go faster.

Typically, a Formula One car, from the start of the season in March in Australia through to the end of the season in November in Abu Dhabi, will improve by about two seconds a lap, which means that if you went back to Melbourne in November, at the end of November, the same track and roughly the same conditions if you could find them, your car would have gone at least two seconds quicker than it did at the beginning of the season. And the teams that can do the best job are the ones that obviously create a fast car in the first place, but then are able to iterate it and develop it. I mean, the iteration is like a — it’s much faster than iOS. You’re constantly having to update your phone with the latest iOS, but Formula One teams are updating their car literally every Grand Prix. But there are so many of them that it’s a collective, team approach that’s important there, whereas quite often, what I’m talking about with these very brilliant mathematicians, data analysts, and strategists, it’s one person, or at maximum two or three, who are arriving at the conclusions, analyzing in real time what’s going on, and making the big calls.

Ben Shields: We have a very famous metric in the sport of baseball called “wins above replacement.” And to apply that to Formula One, let’s say the best mathematician from a winning Formula One team goes to another team. Would that result in a measurable improvement in that team’s performance, if the best mathematician joined that team?

James Allen: Yes, it would. I think it would. I mean, the very best strategy people make the right calls more often than not. The one thing that is inescapable is that the best teams tend to have the best of everything anyway. And if you’ve got the fastest car, the job of the strategy guy is actually easier than if you’ve got a car that’s running for 15th or 16th on the grid [where] you have to use a lot more ingenuity and cunning and innovative ideas to try and make something happen, shake the tree, because you’re there because that’s the pace of your car. But it is nevertheless the case that the brightest guys and women rise to the top.

And here I would make the observation that there are more women participating in Formula One than people might imagine, and certainly a lot more than there are in a lot of other sports like [the] NBA or NFL. If you look at the composition of the, obviously not the players, but if you look at the support staff, I don’t think there are an awful lot of women participating in those sports. But in Formula One they definitely are, as engineers, as data analysts, and scientists. And indeed, of the 10 Formula One teams, three of the chief strategists, the main decision-makers in those teams, are women. And they do a very, very good job.

Ben Shields: That’s outstanding. I did not realize that, and I appreciate you sharing that with our show and our listeners. And the last question on challenging this thesis is around the importance of the drivers understanding some of the mathematical output. Is it important that they are as proficient in the data as maybe some of the other members of the team? How important is it that the driver is obsessed with the data as well?

James Allen: Yeah. The point is really that the drivers of today have to deal with data far more than the drivers of the past. People love to compare drivers of this generation with drivers of the ’50s and ’60s. And whereas with [the] NBA or NFL or baseball, you can do that, because the sport hasn’t really changed that much, in Formula One it’s very difficult, because the cars have changed so much. They’ve got so much more grip. They are much, much faster.

But, I guess without a doubt the biggest change is in terms of the workload for the driver and that they have to be data analysts themselves to some extent. They sit there and they are given all sorts of sheets and information about how they slowed the car down before the corner, how they turned the wheel in and rotated the car, loaded up the front tires, the rear tires, etc., etc. They can make microscopic changes on all these details to improve their driving, whereas back in the days of Mario Andretti and those sort of guys, they would go out on the racetrack, they’d come back and tell the engineers what the car was doing, and the engineers just had to believe them, because there was no way for them to measure what the car was doing, other than the word of the driver.

So, you could say that Mario and his cohorts had a pretty easy time in comparison, because now the teams know exactly what is going on, and they can therefore measure the performance of the driver literally second by second. So, he has always got a whole set of KPIs hanging over his head, and he can’t get away with anything, because they have full transparency of what he’s doing.

Ben Shields: Yeah, and that historical perspective is really interesting, that the driver used to be the main source of data capture for teams. Things have certainly changed. All right, James, I like to get into some future considerations with all of our guest experts and talk a little bit about where we’re headed from an analytic standpoint in the sport. And my first question is about whether the amount of data that Formula teams have at their disposal is actually a good thing for the sport. Remember Austin this year, and because of the weather it impacted the amount of information that teams had, so there was incomplete information. Teams had to use a little bit more of their experience maybe to make some decisions and even rely on some of the driver instincts. So, do you think that the amount of data that teams have is a good thing for the sport?

James Allen: That’s a very interesting question, and I mean, it depends whose interests we’re serving, I guess, really. I mean, from the team’s point of view it’s a very, very good thing and from the driver’s point of view, because it means they have a very, very high degree of understanding of the car’s performance and how to make it better. But I would argue that Formula One exists not really as a technical exercise but as entertainment. And, you know, there are 300 million fans around the world who would agree with me, and they want it to be — the more entertaining, the better. But what they also want — Formula One fans are pretty savvy, they’re pretty tech-savvy and they like deep insights, and they like to understand data. And I think the journey that Formula One is on at the moment is to try and make more of that data available to the race fans around the world, if they want it.

If you don’t want it, you just want to watch the race, that’s cool. But if you actually want to understand some of the biometrics, [like] the heart rate of the chief mechanic who’s standing there in the pit lane waiting with a wheel gun in his hand as the car approaches him at 50 miles an hour has to hit his marks perfectly — you know, his heart rate is pretty high — the drivers out on the circuit, what the car is doing — there are, I think, millions of fans around the world that want access to that, and I think Formula One is definitely looking at ways in which it can present that data to its audiences.

There’s a thing called the Tata Communications Innovation Prize, which is something that Formula One does together with one of the teams and with this company Tata that does all the connectivity around Formula One, where they’re looking for a crowdsourcing, if you like, competition, that looks for fans to actually come up with ways of servicing this kind of stuff. So, actually, more and more the sport is looking to its fan base to tell it how it would like the data to be served. So, I think we’re going to see, staring into the future, I think we’re going to see more and more of that, and I think a deeper and deeper understanding of the sport as a consequence. So, the answer to your question is, yes, I think having lots of data is a good thing.

Ben Shields: Yeah, and I’m with you from a fan standpoint. It just enhances the viewing experience every single time I watch a race. All right. Last future question. It’s about the role of machine learning in Formula One racing. Will the computers be better decision-makers than the humans? Maybe are they already? And if so, how might that change strategic decision-making in the sport?

James Allen: Wow, that’s a very, very interesting question. I think first of all, going back to the whole point about entertainment, I mean, what is entertaining about sport I think is man or woman mastering machine, when you’re talking about mechanized sport anyway. So it’s an absolute brute, a Formula One car. If you or I tried to drive it, we would break our necks going around a corner, and we would just, you know, we’d be terrified. So that’s the first thing that people want to see.

But they also want to see the human decision-making process. There’s something fascinating about people being under pressure and being forced to make decisions. It’s the same in an NFL game: What do you do on the third down, and you’ve got to make a decision on what the next play is? People want to see that. If they knew that it was all being done by a machine or machine learning, I think there’d be less appeal, and I don’t think it’s any different really in Formula One.

So, in the same way there are rules in the sport that stop driver aids, as they’re called — things like ABS brakes. They’re not allowed in Formula One, because you want — part of slowing a car down for a corner and not spinning it is the skill of the driver — and you want to keep that there. So for me, the decision-making process around this, for all of that machine learning might be helpful for, I would think we have to find some red lines here to make sure that it’s always humans who are doing all the key jobs in the sport.

Ben Shields: All right, Paul. James Allen says that Formula One teams should pay mathematicians more than drivers. Do you buy that?

Paul Michelman: I want to buy it, Ben. I really want to buy it, because I love it. I’m missing a key relative measure, however. You brought up this notion of value over replacement driver in the interview with James, and I thought that was a really astute question. What about value over replacement mathematician? If we’re determining who should be paid more, the question is, who is the least replaceable? Who’s to say the mathematician is any less replaceable than the driver?

James made a really interesting observation about engineers during the interview. He said yes, they’re incredibly important, but they’re also plentiful. So, no one individual engineer is going to demand outsize compensation except maybe one or two extraordinary people. If we’re going to determine whether a data analyst or a driver should be paid more, I need a relative measure that compares VORD to VORM. And as far as I know, that data point doesn’t yet exist.

Ben Shields: Ladies and gentlemen, you heard “VORM” here first. Excited, Paul, to see what some future work could be done around that particular metric. From my standpoint, I would love to see a chief strategist of a Formula One team in a Rolex commercial, for instance. But we’re not there yet. And this specific argument is about whether data analysts or the mathematicians should be paid more than the drivers.

Now, that’s different from whether the data analysts, to your point, are more valuable than the drivers with regard to winning races. So, until a chief strategist is the number one reason fans tune in to watch a race or buy a ticket, or a sponsor signs on because of that strategist, then drivers because of their commercial potential will and should still be paid more. That’s not to say that the mathematicians aren’t valuable. I love them. I appreciate them. But in the end, this sport is commercially driven as well, and the drivers bring in the revenue in large respect.

Paul Michelman: To that point, this is such a nuanced argument, and I love it for that. James brought up this notion that sponsors are looking at the commercial opportunity through a new lens. It’s not just about logo placement on the car; it’s not just about product placement and advertising. It’s also about access to the strategic thinking of the team. And so, if we’re going to be selling as part of a sponsorship, you get to meet the brilliant minds and learn from the brilliant minds behind our competitive strategy, that argues in favor of the mathematician.

Ben Shields: Paul, that’s a very valid point, and I will say that for organizations that are looking to study big data and analytics and decision-making, Formula One teams, based on some of the work that I am doing in this space, do provide a very interesting and applicable example to understand data-driven decision-making in today’s marketplace.

Paul Michelman: Mary, last word is yours.

Mary Dooe: Yeah. You know, I really do like this argument. I like the idea of mathematicians being paid more than athletes in some sector. I think unfortunately, as is true in almost every industry, salaries and pay aren’t always linked to the direct correlation of the success of, whether it be a sports team or a company. And I think unfortunately, to bring it back to data, even if you had the data to prove that someone winning was based specifically on one person versus another, that still isn’t necessarily what goes into salary negotiations. So I’m a little bit of a skeptic here.

Paul Michelman: This has been Counterpoints, the sports analytic podcast from MIT Sloan Management Review.

Ben Shields: You can find us on iTunes, Google Play, Spotify, and wherever fine podcasts are streamed. And please take a minute to post a review. We really want to hear your feedback.

Paul Michelman: Counterpoints is produced by Mary Dooe. Our theme music was composed by Matt Reed. Our coordinating producer is Mackenzie Wise. Jake Menashi is our crack researcher. Our maven of marketing is Desiree Barry.