Can AI Read Your Karting Data? An Honest Answer

Let's be honest, AI is revolutionizing our lives, and it's probably going to do it even more in the future. I mean, I'm one of the power users of Claude, for example. I started using ChatGPT like a couple of years ago, and then swapped to Claude early this year, and it's working really well for me. I am doing lots of tasks on it, and I've been working on PURPL a lot thanks to Claude.

But the thing is that now you can really use AI in racing, in karting specifically, if you have no sort of knowledge in reading data, or even if you do have data analysis knowledge and experience, you'll probably be better off using AI to do the analysis for you, because the reality is that we as human beings will not be able to be at the same level of AI in terms of computation, all the numbers and the math required to compute all those laps, corners, owners, and sectors, as AI is able to do. You will not be able to do that as a human being at anywhere close to the speed of AI.

Not just any AI: it has to be built for karting

But obviously, not just a random AI, just using ChatGPT, Claude, Gemini, you know. You need to have a specific app, an AI app that has been built and developed fully karting-centric, that is gonna know exactly what to do, and has, you know, predefined templates of what results to spit out, and give you the useful insights that you need in order to find lap time. So here at PURPL, I mean, I've been really pushing myself to get the best possible outcomes and get the drivers to find lap time as easily and effortlessly as possible.

Of course, you still have to go out there and find the lap time and do the steps required in order to get faster. I mean, the AI is not going to do that for you. But the reality is that identifying the corners where you're not doing really good right now, and where you're losing time in a specific session, is just super useful, because then you can go into the following session and have a clear idea of what to work on and where to push your limits. It works really well for me. And it's really the tool I build that I wish I had when I drove 20, 15, 10 years ago, and even when I'm driving today. So when I go and drive myself, I am just like using it all the time, of course. But the question that many karting drivers ask themselves is whether AI can actually read karting data and do it in a real way, in a way that's not hallucinated, with fake data, fake coaching, etc.

The rule that decides whether you can trust it

So that's, by the way, the priority here at PURPL. Like I'm just super strict on that. I mean, if there isn't a high confidence in the coaching being provided, I'd rather let the user know that there's not enough confidence, and it's better not to give any feedback instead of giving wrong feedback.

What AI genuinely brings: memory and scale

So, you know, the reality is that we are currently in a position where AI is able to again reach a large number of, of data and parse it all together. But the great thing is that AI has memory. So it's going to remember everything you did in the previous sessions, in that specific track or in other tracks. It's going to know your sort of driver DNA, and have an idea about the areas that you've currently been struggling the most lately, and the areas where you should be working to improve yourself, even though you may not have the data of other, the reference drivers.

That's the thing, you know, with AI you can really find lap time to yourself, and not necessarily find lap time against some reference driver, whether it's your teammate or somebody else. So the thing is that you can find lap time by yourself, and that's what PURPL excels at, because it allows you to see, first of all, among all the laps and all the sessions that you did, which were the best sections, the best theoretical laps derived from the sections. So of course, I, I always prefer to do that from a single session, because comparing different sessions, most of the times in karting, is not a real fair comparison.

It's not really like comparing apples to apples, because every session the track changes, it develops, it either gets a little bit faster or it gets a little bit slower. So I just really believe that it's better to compare yourself within that same session, and see how far you were, for example, from reaching the ultimate perfect lap.

How the lap gets split, and why fake splits ruin everything

And how does the AI do that? Well, it basically splits the lap based on an algorithm that, for example, we created at PURPL, and it basically is able to detect different sectors, where one sector begins and the other ends, but not the same as how MyChron does it. It does it in a different way, so that you can have, at the end of the day, more splits than just the three ones given from MyChron, for example. You end up having six, five, seven, eight, depends on the track. And it's great, because those sectors are real.

I mean, they're not splitting the middle of a chicane where it's going to be unrealistic for a theoretical lap, because in a fast chicane or a slow chicane, if you are just driving for the entry, and would be split in the middle of the chicane, for example, you would gain a lot of time in one lap, but obviously you would lose so much on the exit in that specific lap. But then the theoretical would be unreal. So the problem we faced was like, okay, we got to make sure that this gets fixed, and we needed to create a strong algorithm, which we did. And so now we are able to confidently give you splits that are real, and they are not going to affect the previous corner or the next corner in order to make that split faster. So yeah, they are just kind of like autonomous splits, which is really important.

And once you have those splits, you can then see, okay, you know, in my fastest lap, I gave up, let's say, a tenth in the mini sector one, I gave up a tenth in mini sector two, I gave up two tenths in mini sector three. And maybe at the end of the lap, after seven splits, you will be probably four tenths off, just giving you an example, from your ultimate theoretical best time, you know? And that's showing you that, you know, that was what you were capable of doing with whatever you had in that moment, in your hands, with your engine, with your chassis, with your skills. It just needed to be put together.

How far off your theoretical means "not ready to win"

So I'm a big fan of this comparison against yourself, because the goal is to become the best version of yourself. And what I noticed is that when drivers are not ready yet to win, they typically are very far away from their theoretical. And to me, far away means two, three tenths. That's really too much, because imagine qualifying for the CIK World Championship: within one tenth you find 15, 20 drivers in a junior or senior race, whereas in a WSK qualifying session you will be at least a P15 if you're more than a tenth off P1 in qualifying. So that's really important, you know, it's gonna be very crucial for you to match whatever performance you had in that moment under your ass, so, you know, you can't blame the engine or the chassis or the setup for, for giving you like one tenth lost, if you have three, four tenths in yourself that's displayed in the theoretical.

So I also see myself, whenever I'm driving at the top of my game, I come out of the kart and I'm not even a tenth off the theoretical, you know, the ultimate lap theoretical from PURPL, and I'm probably matching the theoretical from AiM MyChron. So it means that I was able to reach the peak performance of the kart, of the driving, at the same time, and I was able to put everything together. Which is kind of easy to do that in a quali, I would say, because, like, the performance is going to come in the last couple of laps. So if you have a clear lap, it's going to be quite easy to match the theoretical, because by the last lap it's probably, you know, going to be more grip.

And so, unless you make big mistakes on that last lap, you should probably match your fastest lap in every sector. But of course, in a longer stint, that's going to be more difficult, to get closer to your theoretical, because if you do 20 laps, out of which you have like at least 10, 15 laps in the optimal operating window of your tires, engines, etc, then, you know, if you are doing a really good job, you're, you're going to push your limits. At some point you will make mistakes. So it's going to be really hard to get close to your theoretical time. But if you can sustain that level of consistency, which is really important, and closeness to your theoretical time, well, then you're going to be set for many, many successful races.

Why comparing against yourself beats comparing against a rival

And that's the moment when you realize, okay, I'm kind of ready to win now. So that's one thing. So that's the really cool thing about AI, is that you can use yourself as a benchmark, and really extrapolate the maximum performance from yourself. Even though you may not have the data from other drivers, that's fine, because even if you did have the data of other drivers, you will never have the certainty of their kart compared against yours, because you can always blame that, oh, they had a different engine, or they have different tires, or they have a different chassis, or they have a different setup, etc. So you could always have an argument about it. So that's why, even if you had the data of somebody else, I mean, that's useful, of course, if you're in a race, if you have multiple teammates, of course.

Let's be honest, in order to find the best lines and the, the best technique, if you have to drive a little bit harder on the entries, or you know, maybe you have to open up a bit more the entry, etc. But uh, you know, when you're driving by yourself, you have still lots to work on, still lots to learn, and still, uh, you know, the goal of hitting that ultimate perfect lap, which is what drivers like Lewis Hamilton and Max Verstappen and Ayrton Senna refer to, you know, it's, that's the goal, to, to reach your perfect lap. And the perfect lap doesn't really refer to any driver around you, it's just referring to you and yourself, like, you against yourself, you and the kart against yourself. So it's really about that. And I really feel that PURPL allows you to track your performance against yourself, so you can know whether you had a good session or not, and how close you are to your theoretical, and whether you're improving, in, in terms of latest sessions.

Read relative values, not absolute ones

Always remember that in karting there's a huge difference in terms of tires, in terms of grip level on the track, also in terms of air temperature. I mean, of course, you can readjust the jetting on the carburetor in order to have a different carburation, but the reality is that the track is never going to be the same, as well as your kart. I mean, like, if you're not planning to change the tires, because you're trying to save money, and of course not everybody's able to run new tires every single day. The thing is that, uh, you know, if you are driving with the same tires the whole day, I mean, by definition, every session your tires are going to be a little older, a little older. But it's not just that, they also, you know, accumulate heat cycles, and therefore they get harder, ever so slightly they get harder, and with less tread, so with less grip in a way.

So, you know, it wouldn't be fair to compare your last session against your first session. So we should always be quite open-minded, and be like, okay, yeah, you know, um, the last session I wasn't as quick as my first sessions, because, okay, probably my tires were worse, but at least you can have an idea of how consistent you were. So the standard deviation, and also how much your braking points were consistent, how many meters away you were driving from your average line. So you can see, okay, I've been also pretty consistent in my lines, or I'm being too inconsistent. That would be something really cool to analyze, I would say, and that we provide in PURPL. And it's really awesome, every time I go and notice that myself. And the other thing is that you can also know how far you're from the theoretical in the last session. Maybe you have old tires, and you know you're getting two tenths off from your personal best times in the morning, but at least probably you are driving a little bit more consistent, and a bit close to the theoretical, because maybe you have built up a bit more muscle memory during that day, you have kind of removed some rust. Maybe it was a few weeks since you last drove in a kart.

And so, by the end of the day, you'll be able to be a little bit more precise, a little bit more consistent, a little bit closer to your potential, but maybe you're a bit tired. So, of course, if you're able to do that really well, and then next morning you can fit some fresher tires, as well as you are fresher in terms of body, then of course you'll be going in the right direction. You'll be progressing. Of course, yes, in karting, remember that it's not an absolute value that you should look at, you should look at more of the relative value, okay? Because, again, you may always find the track being a little bit worse in a day and a little bit better in another day. You will never have the exact same tracks with the exact same grip level, because you will have different amounts of drivers on the track. You will have different types of rubber on the track. Some drivers are going to run hard tires. That's going to make the track a little bit slower than if drivers are driving with soft tires. So if everybody runs the, cons, soft tires, for example, the track is going to get really fast, even though you may still be on really old hard tires, you know what I mean. As well as the, the opposite applies. So it's really important to just like take every session, every day, as a grain of salt, and you just have to try to improve yourself. That's what I always say. But when looking at your numbers, at your metrics, trying to improve your metrics, and by looking at those and push your limits, you'll get better as a racing driver eventually.

So, can AI read your karting data?

So remember that, yes, AI can definitely read your karting data. I feel that, of course, the role of a driver coach is still very useful, because, I mean, AI is still not going to be able to tell what you've been doing wrong in a race in terms of race craft, also in terms of, like, your body posture, in terms of your psychology and fixing your mindset during the day, as well as, like, reading the body language of how you behave outside the track and inside the track. So all these sorts of things are something that experienced racing drivers and coaches will know how to deal with. And AI obviously is not there, and probably won't be there yet, unless, you know, you can have a camera that literally looks at you, looks at your moves during a race, and tracks your, like, body posture. And yeah, it's going to be really hard for that to happen.

But if that would happen, I mean, for sure, I still believe that the driver coach would have an upper hand with everything that requires judgment. But when it requires numbers and data, I just feel that AI is going to beat a driver coach and a data telemetry guy every single time. And the coaches shouldn't be ashamed of using AI and be like, oh, it's a shortcut, I'm feeling guilty. No, you shouldn't feel guilty. I mean, think about all the professionals in the world, like marketing professionals, you know, finance professionals, sales professionals. They're all using data and AI softwares to rely and improve their workflows.

And that's nothing to be ashamed of. And feeling guilty about it, it's just, you know, the way humans work, they try to improve productivity. So AI is not here to replace humans and driver coaches, they're here to augment them. And I just generally feel that driver coaches are gonna become even better coaches if they implement AI. But of course, there's many different websites trying to solve the same problem, but the goal is to just pick the ones who solve the problems in the best possible way, and they have proven track records, and I'm sure PURPL has it.

So let me know how it works for you. Sign up today, and feel free to hit me up and let me know what you think, whether you have some improvement requests. I'll be here to listen. Take care. Remember to keep sending it. Ciao, ciao.