Cliff Notes

Ever since I started track driving, some 6 years ago, I’ve watched videos of people driving Thunderhill. It’s the closest track to me, and also my favorite (the West side is actually my favorite, but the East side is near the top of the list). In the beginning, I was just trying to learn the track. Later, I wanted to see how my lap times stacked up against other drivers. Today, I mainly watch to analyze driving technique.

When watching videos at Thunderhill, I like to focus on Turns 1-3. Each turn exposes specific driving errors and the entire sequence from the tower to the apex of T4 is under 40 seconds.

The video I have for you today features Cliff, a coach with Audi Club. His YouTube channel features a video from 6 years ago with HoD A and S stickers, so it looks like he got started track driving around the same time as me, or possibly earlier. Cliff is driving a 2015 Golf R with a shitload of upgrades. The Golf R has 292 hp stock, and the Stage 1 tune upgrades this to over 350 hp. The car also features Ohlins suspension, StopTech brakes, and 200TW rubber. The car is properly built for track duty. The description of the video says it’s his fastest lap of the weekend. The video doesn’t feature a lap timer on screen, but from the video timestamps I estimate 2:13.9. Seems like he should be able to go faster. So let’s dive in and see if we can figure out why.

Watch the video and then follow along below.

Some of the things I like about this video are the picture quality and overlays. It’s too bad the camera isn’t mounted inside the car, because it would be great to get an idea of what the driver is doing. Given that most overlays don’t show steering data, it’s useful to watch the driver’s hands to see if he’s fighting understeer or oversteer. I also like watching shifting technique. Oh well, not today.

One of the most useful tools for analyzing drivers is a speed trace. Ideally, your data acquisition system updates at 10 Hz or better. Phone-based apps like Harry’s Lap Timer or Track Addict typically record at only 1 Hz unless they are provided with an external antenna. Since I don’t have data for the lap in the video, I made my own by recording the speed of the car in 1 second intervals using the video timestamps. This provides a low-resolution speed trace (blue) very similar to what you would see from a phone app. I’ve also drawn a theoretical speed trace based on my own imagination, which I’ll discuss below.

Turn 1

In the first few seconds, you can see a major problem. The speed trace has a very rounded top. The car is coasting into the brake zone. I don’t actually subscribe to the phrase “you should always be on throttle or brakes” because people who hear that think it means 100% throttle or 100% brakes. And there are also instances when coasting is actually appropriate. But 2 whole seconds of coasting on the main straight is not one of those times. The amount of time lost is only a couple tenths, so it’s not that big a deal in terms of lap times. But it is a big deal in terms of technique. One should drive the car all the way to the brake zone.

The next thing to note is the deceleration. It’s not very steep. A car with 200 TW tires can decelerate at 1.0g. From observing the G-meter, the car never gets close. It spends most of the time at less than 0.5g. Technique-wise, I also brake more gently in high speed corners. My mindset is that I’m trying to set the corner speed at a specific value rather mash the pedal. But the braking here is just too timid. Looking ahead at other corners, he appears to always brake gently. The car and tires are capable of much more.

The final thing I want to point out is the location of the apex. This is the black arrow. The slowest part of a corner should be before the apex, not after. He’s coasting through the corner trying to hold speed. In the overused phrase “in slow, out fast”, the in slow happens before the apex and the out fast starts occurring a little before the apex. Here, the slow is still after the apex.

Turn 2

Like T1, there isn’t enough commitment to the brake pedal in timing or pressure. But the overall shape is pretty good. I see a lot of drivers mash their brake pedal and over-slow the car. Not so here. He’s using the brakes to set his speed, and then he drives through at the speed he set. Good.

Unfortunately, the mid-corner speed of T2 on 200TW tires is not 61-63 mph. Looking back at some of my data, I drive a couple mph faster in the pouring rain or when joking around on 185/60/14 Douglas Xtra-Trac II tires ($38 Walmart tires with a 420 treadwear rating). On a dry track with 200 TW tires, I’m around 72 mph. Why is he driving so far under the limit? Probably because he doesn’t like the feel or sound of sliding tires. Tires are supposed to slide a little on track. That’s where the optimum grip is. Driving a sliding car can be uncomfortable if you’re not used to it. The way to get used to it is to do it.

While T2 is a carousel with a constant radius, it is almost never the case that one should drive a constant radius. On a long corner like T2, you should use the first half for braking and the second half for accelerating. You may be slightly slower on the way out of the corner, but you gain a lot more by using the first half as part of your brake zone. Since there’s such a short straight from T2 to T3, it’s better to take this as a double apex rather than single.

Turn 3 and Turn 4

T3 is tricky because it’s off camber. There are lots lines through the corner, especially when racing wheel to wheel. Although Cliff’s overall grip level isn’t where it should be, the shape of the speed graph is just fine.

T4 is a typical 90, so the minimum speed should be before the apex. Here, like in T1, the minimum speed is actually after the apex. If you’ve got a car with 350 hp, you should use a driving line that optimizes the power of the vehicle. That means getting the braking and turning done early so you can throttle on a straight line. This is doubly true for FWD cars.

Rant on

It’s not really Cliff’s fault that he under-drives his tires. The E in the HPDE system is totally broken. If you haven’t read “Optimum Drive”, by Paul Gerrard, I highly recommend you do. He talks about how backwards the HPDE system is. I won’t repeat that here. Go get his book. Paul also says that if we want to solve a problem, we need to get to its root. The problem isn’t that Cliff coasts into brake zones or drives at 0.8g. The problem is that he’s not comfortable driving a sliding car. Fix that problem, and all the symptoms go away.

What’s the first lesson we usually teach new students? The racing line. As if that fucking matters. The line is a result of optimizing grip. Teach drivers to feel grip and the racing line will follow. The reverse isn’t true. Fuck the fucking racing line. I’d much rather have students drive in the middle of the track. There’s less chance they go off track and roll or hit something.

When drivers get comfortable just under the limit they reach a performance plateau that’s hard to break through. And the better they get, the harder it will be to unlearn later. Stability control, sticky tires, and 500 hp monsters all conspire against acquiring actual skill. But the students show up in Hellcats and Vipers, and I’m not getting in the right seat of one of those things without nannies.

If you really want to get better at driving, you have to have the right environment. Thunderhill in a 500 hp monster is not the right environment. The consequences of crashing at 130 mph are just too great. There’s a reason that the Kenny Roberts school is on dirt and why the Skip Barber school uses all season tires. Learning car control is safest when tires are slippery and speeds are low. Simulators are cheaper and safer still.

Rant off

On the other hand, not everyone needs to be a driving ace. Lots of people enjoy listening to music. Fewer people play music. Even fewer compose. If someone is having a great time driving around a track at 6 tenths, do they really need to turn it up to 8 or 10? As a coach, my #1 priority is safety. The #2 priority is to make sure the student is having a great time. For novices that probably means teaching them the racing line and “advanced techniques” like heel-toe shifting. As students graduate to intermediate and advanced, they need level-appropriate instruction. And just like with music or anything else, the lessons become less entertaining and more work. Drivers who didn’t start with a foundation of car control will take longer to reach whatever level they are trying to attain because they will have to unlearn a bunch of bad habits along the way. Who cares? It’s just time, and last I checked, time on track is a lot of fun.

Conflicted

Personally, I’m really conflicted about driving education. I firmly believe that car control is the only thing that matters, and if I ran a driving school, it would be mostly drills on a skid pad or simulator. However, I also believe that as long as drivers are safe, they should do whatever optimizes their fun. If I ran an HPDE organization, we’d do burnouts, drifts, jumps, and of course, the racing line.