How to use Data Analysis

We discuss the pros and cons of data analysis in this blog. Now, technologies improved a lot over the last five, 10 years, and when I was winning all my races in the 2005, 2015 era, we didn't really look a lot into data analysis. It just wasn't something that technology was up to scratch with.

So nowadays when you're looking at, whether it's a Mychron unit, or Starlane, Alfano, Motech for racing cars, the amount of knowledge that that computer system can gain from your driving is incredible.

Whether that's where you're accelerating, where you're braking, how much brake pressure, you can GPS your racing lines, your RPM to make sure when you change the sprocket, whether it's any good or it's bad, engine temperatures. All of these things can be quite confusing and quite daunting as well, if you're new to karting. It's like, "What do I actually need to look at for me to try and establish how to go faster?"

We'll start with the pros on why we think that data analysis is good, and what benefits it has.

First and foremost, it's black and white. It tells you what the driver is doing or what the engine is doing. So if you are running a 78 sprocket and you want to go to a 76 sprocket to try, you can analyze by downloading onto your software how much more straight line speed you've got compared to how much acceleration you've lost. So we know that a smaller sprocket is going to give us more down the straights, but it'll take a little bit longer to accelerate out of the corners. You can see the time difference, we've gained one tenth down the straight, which is great, we've lost two tenths accumulative across the lap with our exit speed. So it's a one-tenth loss. Okay, maybe we don't need to use that sprocket.

The other good thing about having data is, if you've got teammates or other drivers to extract information from, you can basically just analyze where you are losing time compared to them. So just say you're doing a 50-second lap and you've got a friend doing a 49-second lap, the data's going to show you exactly where you are losing that one second. Is it three tenths in the engine? Is it seven tenths in you're driving? Okay, now we break down the corners on where we are losing that seven tenths of a second. Are we braking too early? Are we entering the corner way too fast and then we're losing a lot more on the exit?

Then the other good thing is, when you're analyzing with another teammate or another friend, is that you can GPS the racing lines. So, suddenly you start to highlight, "Okay, maybe I'm turning in a little bit too early, I'm off the racing line, or I'm just turning in too late, or I'm holding the exit too tight so I'm not allowing the kart to flow out of the corners." That's a really good way that you can compare yourself to other drivers, and that just straightaway tells you which corners you want to focus on most. So, if you're losing four tenths on one corner and one tenth on another three, let's just get that one major mistake out of the way and then we can try and halve the distance that we're actually off of another driver.

Where another pro is, is that, when you're changing engines or you're going into a new track, the new track one is going to be hard because you might not have data from another race event, but you might still know the optimal range your engine should perform in. If you know that you want to hit 15,000 on the top RPM and 8,000 on the bottom RPM, for example, you can still see that on your dash, but you can start to analyze how long you're on the right rev range for or how long you're in the lowest rev range for, which corners they are, if there's multiple corners, because on your dash you can only see the highest and also the lowest data. You might want to find that if you download it, you can still see the engine temperature as well. That's a little bit easier to see when you're tracking your lap from corner by corner, that can help quite a lot as well.

Some of the negatives when it comes to data analysis is that if you don't know what you're doing, it's very easy to get confused. Now, there's a lot of channels on our data analysis software. We've got speed, RPM, engine temperature, lambda values, lateral grip, so you can see how much grip level you've got when you go around a corner. If you don't know what you're looking for, it's easy to get confused. If you read the data wrong, that can also confuse you and the driver. It's a way that you then, as an example, if you are entering too fast and you've got too much mid-corner speed and you're going the gas too late, then you'll have no speed coming out of the corners.

Straight away, if you don't explain that to the driver very well, "Oh, I don't know what you're doing, but you've got more mid-corner speed, which is normally really good, but you're not getting out of the corners. So, it could be a sprocket thing."

It's how you translate that information that you're looking at to your driver, to try and maximize the gains. We still see a lot of drivers in this day and age that don't use data that win races. So it's not the be all, end all, that if you don't have data analysis going on, that you can't win races, because it's not quite true. It's how you read it and how you perceive that information to your benefit, that's where it's really going to come into effect.

The other negatives are that it takes away focus sometimes from your actual on-track driving and your preparation. Now, I've got a few drivers that we coach that are really into the data, which is really good, and you can sort of see just as much sometimes from the video. Not necessarily do you always have to be looking at the lines and what it's showing you from the speed and RPM, because the track map sometimes doesn't highlight that you are turning in too late or you're turning in too early or you're not using enough curb, for example. So, we can get so focused on the lines of the speed and RPM traces, but actually just watching a little bit of GoPro footage can actually show you just as much.

All I would say is when it comes to the video side of things, is that it's harder to look at finding one tenth of a second from video. 0.5 to one second, it's easy to see on video, but when you're trying to look for little margins on the track, it's easy to see that on data as well. It just depends on how far off the pace you are and how big are the mistakes you're making as to whether video or looking at the data is better for you.

So that's probably the pros and cons of, "Do I need data analysis?". I don't think you need it, but it's a tool to get you up to speed faster, but you also need to have those good benchmarks to work with in order to actually get the most out of it. Now for our race team, we have some really experienced and fast drivers that are winning the Australian titles, and then we've got some other kids who are just coming into these classes that might be starting in the mid-pack. Now for a 15th place driver to then compare their data off the fastest person on the track, well, that's invaluable because it's the easiest and fastest way for me to coach someone and to learn, is from comparing stuff against the fastest kids.
Straight away, depending on their braking points, their racing lines, their acceleration zones, how much grip their kart has on the track, their engine RPM, their temperature, it's easy for me to break down where that time difference is.

Now, I could also do that on video, but you're probably having to watch a whole session to see where the mistakes might be made, where we can compare on the data pretty quickly once you know what you're doing.

If you are making the same mistakes in every session and you're just comparing yourself to yourself, well, then you're not really going to learn from it. So that's where I find it's sort of the pros and cons.