F1 to re:Invent series through ‘Artificial Intelligence’

Apart from his very comfortable looking loafers, the thing that struck me about Formula 1 technical boss Ross Brawn’s AWS re:Invent conference presentation was the content. Using AWS Sagemaker, F1 is going to offer a series of data-fueled graphics that predict chances of overtake, real-time car difficulties such as tires, reduced power and other elements that would suggest the driver is in “conserving” mode…or as we call it, lift and coast.

I’ve been in the technology industry my entire life and I have never had much time for the development of tech using the stratagem, “because we can”. Just because you can doesn’t always mean you should or that everyone wants you to.

Would a graphic showing each driver’s cornering speed be interesting? Sure. Would another graphic showing current tire degradation and temperature be an interesting thing to look at? Maybe, although I’d prefer other teams didn’t know to what extent the competitor’s tire was in difficulty.

Ross says that F1’s unique position of having all the data on all the cars will allow them to juxtapose two competitors in real time and then go one step further by predicting the outcome of passing attempts, pit stop under and overcuts, who’s under threat of being caught and what chances they have of being overtaken et. al.

Because we can

I’ll be honest, I don’t think I want to know all of that information. I would rather not have Lewis Hamilton’s running behind Max Verstappen boiled down to an 87.4% chance that he’ll pass the Dutchman. Sebastian Vettel is going to pit and there is a 72.9% chance that Lewis will overtake him whilst in the pit lane.

Before this Sagemaker predictive data, Mercedes may know, via lap times and a radio communication at Ferrari, that Kimi Raikkonen was struggling a bit with his tires but now they’ll know Kimi’s exact tire temperatures, wear rate or level of degradation or grip level and that Valtteri Bottas has an 89.2% chance of passing Kimi if he remains at the speed he is. Geniuses at Mercedes will know exactly what it means and what to do if Max’s tires are at a certain temperature.

Not to be a luddite here, I will admit that there is data that could be very fun to have access to. Cornering speeds, G-force loads, any weather-related information track adhesion levels and temps, more detailed pit stop data to help a 1.9s stop become marginally more exciting etc.

Perhaps time gaps between drivers during Virtual Safety Car periods on a real-time track map, throttle, gear, steering and brake input that is more accurate and a better graphic. Current running order with a much better graphic showing tire compounds, sector time deltas to all other cars etc.

Who doesn’t like nice wrapping paper and a bow?

In the end, I think much of this, at this point in F1’s success arch, is gift wrapping of an otherwise much larger issue which is on-track action and excitement. F1 may never admit it but the deep dive into the hybrid power unit era has cost them dearly and the baked-in advantage Mercedes has had since 2014 is still unprecedented with a second a lap advantage at most races this season—or thereabouts.

I said it in 2014, Mercedes would, by their own influence over the sport by virtue of Ross Brawn, create an environment that they would win and dominate in for years to come. They did and they have. They influenced the regulations and have a baked-in advantage.

F1 would do well to stop nibbling at the periphery and start digging in to what is truly ailing the sport. It’s all very well for Liberty Media to purchase Formula 1 and then assign two outsiders to run it. Knowing their lack of granular insight, they hired Ross Brawn who is, by any stretch of the imagination, a brilliant engineer but there’s the rub.

Ross is an engineer and much of me believes this sport has become too technical and too complicated by virtue of the power leveraged by influential F1 engineers and perhaps chief among them, Ross Brawn. Ross’s solutions are going to be very engineering focused and what I would argue is that the series needs to think of its commercial appeal and revisit the notion that they sell entertainment, not mobile R&D rapid prototyping ecosystems for Mercedes, Ferrari and Renault engineers to hone this skills and work on new tech for their EV cars.

It takes more than engineering data streams couched in nice graphics to fix F1. In fact, I’ve argued that the cure may very well taste much worse than the current illness and a serious reset and reconfiguration of F1 needs to happen. Entertainment, driver personalities, brilliant journalism and media outlets bringing behind-the-scene content. Photographers shooting brilliant, high-resolution images that tell stories and capture the season in stunning imagery.

F1 needs to think long term and long form. F1 needs to build generational appeal over time as a slow burn, not a quick hit viral video. Acquiring social capital is exhausting and short lived. These graphics aren’t going to make me like F1 more and I doubt they will appeal to my Gen Z children. What they find much more compelling is the action, entertainment on track, driver personalities, politics in the paddock and the stories—for the love of all things holy, bring back the stories, not the tweets.

Perhaps I can put it in a way Ross Brawn and F1 will understand. The chance of me resonating with the new graphics, finding them outrageously compelling and reinvigorating my love of the sport? Around 37.2%. That’s if I don’t get passed by Gen Y who have, currently, a 43.4% chance of doing so…although several of them are nursing severely overheated bums.

0 0 votes
Article Rating
Notify of

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Newest Most Voted
Inline Feedbacks
View all comments


” F1 may never admit it but the deep dive into the hybrid power unit era has cost them dearly and the baked-in advantage Mercedes has had since 2014 is still unprecedented with a second a lap advantage at most races this season—or thereabouts.” This is exactly the point. He’s hooked on telling you what was, what happened, and only using that to predict what will happen. 1. That means there will be no surprise if they get it right. 2. That means they can design cars closer to spec and thereby ruin the unknown in the sport. For them… Read more »

Tom Firth

I imagine the conversation went something like this…

I read about this machine learning thing everyone is talking about, how can we get involved in that?

Well one of our new partners is interested in it.

Right, lets do that then. It will make us seem relevant.

Tom Firth

The big yellow numbers… Ahh memories


it does have that air of teaming up with ‘Partners’ (who will pay). this like the rise of xG in soccer or football which basically assigns a value to the chances of a shot resulting in a goal. It does so by crunching data from thousands of historical shots and filtering them based on factors such as distance, type of shot, type of pass and the number of defenders between the shooter and the goal. so in F1 it will rate that chances of a pass by a driver at a certain corner judged by the past by employing some… Read more »


“AI As Savior” seems to be the new corporate gobbledygook panacea magic rescue remedy meme. Phooey! There was a very interesting and revealing debrief on Racer or one of the other sites a couple days ago from Marcus Ericsson after his first comprehensive IndyCar test. And it validates exactly why my experience over recent seasons that IndyCar and F2 races have much more drama and texture. Sorry to say, F1 racing isn’t even within shouting distance distance of those series, MotoGP or, frankly and selfishly, any of the other dozen other ways I might choose to spend a couple hours… Read more »


Didn’t F1 say they wanted to make races less predictable like a week ago, and now their plan is to use Skynet to predict the outcome of races? I swear they can’t even keep the bullshit they feed us straight!!!!

Fred Daniell

Dear Todd, Where to begin?? I fully agree with you that this answers a question not yet asked. I also agree that predictive modeling does NOT make the race more interesting to me. In fact, I would argue it makes it LESS interesting. I am with you that not knowing adds to the excitement. While I am not a fan of Christian Horner, I must give him credit when he said F1 should strive for a balance of 1/3 driver, 1/3 chassis and 1/3 engine. I fear Mr. Brawn only knows how to capture the 1/3 chassis and 1/3 engine,… Read more »

Fred Daniell

Sorry, I guess that was Brazil with the engine failure. After the season the races quickly begin to run together!