Self-Evaluation

Published: December 22, 2012

Novice drivers are constantly evaluated by their instructors. The negative evaluations (these would be something like "you missed the apex", "you entered too fast", "you turned too early") are usually not voiced. Instead the instructor would suggest the corrective action: "brake earlier and turn earlier", "brake more", "wait". Without an instructor in the car it is necessary for the driver to self-evaluate their performance in order to continue to improve.

Self-evaluation consists of noticing mistakes, but also opportunities for improvement. The primary benchmark is lap time, however most skills can be evaluated. Evaluation can be performed in real time on race course or after a session while looking at lap times or data from a data acquisition system.

On-Track Evaluation

A driver that is learning a track or driving competitively is constantly evaluating the car's behavior. After every corner the driver considers whether that corner was taken optimally, whether a mistake was made and whether the corner could have been taken better.

Mistakes are perhaps easiest to recognize. At advanced level you should have a plan for each corner before you get to it. If you have a substantial amount of experience at the track being driven, you probably have a plan for each of the corners before you even get in the car. If what you do in a corner does not match what you planned to do, you made a mistake. A mistake can be very obvious, such as putting wheels off pavement, or subtle, like lifting for longer than you know you should.

Opportunities for improvement are similar to mistakes but even more subtle. Entering a corner too slowly (overbraking), taking a line that is too tight or too wide, waiting to get on power too long are all opportunities for improvement. In order to spot some of these the driver must have significant experience with both the track being driven and the car being used.

Optimal driving happens when the driver is not making mistakes and also not leaving much time on the table. There is objective optimality, which is what the best driver in the world could achieve given the car being considered and environment conditions. There is also subjective optimality, which is how good the driver themselves thinks they are.

Traction Sensing

This is what a driver does when they evaluate their car's performance on track. Everything in the preceding section can be reduced to traction sensing. Traction sensing has a simple definition: it is the act, or art, of feeling how much of the available grip the car is using. It is not so simple in execution; see the traction sensing page for more details.

At novice level, the instructor is performing this job. At advanced level, you are in the car by yourself and it falls on you to perform traction sensing, and know how much grip the car has. It is my opinion that driving at 9/10ths or above requires traction sensing.

Absolute Lap Times

Lap times are a useful tool for driver development. They have an obvious utility in competition as well, but here I will look at them only with respect to driver development.

For lap times to be meaningful they should be compared to those of similar cars. Miata drivers have an ever-present benchmark in the form of Spec Miata; top level Spec Miata cars are quite similar or better in performance to most mildly modified Miatas at HPDE events, giving Miata HPDE drivers a reference point that they can use over years and years. And because Spec Miatas run everywhere, both typical and lap record times for them are available for just about every track.

Drivers of other cars need to work harder to find useful benchmarks. A good strategy is to pick a class that their car would fall into, with all planned modifications, and measure yourself against top drivers in that class.

The important value is the difference from the best lap time of benchmark class to your lap time. When you start out, this difference is going to be big. As you become a better driver, the gap will start shrinking. An experienced HPDE driver in a Miata could place in the middle of Spec Miata qualifying times. If your car is actually faster than the reference class' rules, you should expect to eventually turn lower lap times than the entire class. First such goal would be to run lower lap times at a given event than any car in the reference class. The second goal would be to run under the lap record of the reference class.

Lap Times Within A Session

The spread between lap times within a single session tells you whether you are still learning the car, the track or possibly both.

When the driver is in "learning mode", lap times are dropping visibly between consecutive laps. A driver whose best lap is at the end of a session is in learning mode.

Drivers who are not in learning mode take one or two warm-up laps and then proceed to run consistent lap times for the remainder of the session, or until their tires go away, whichever happens first. For drivers in Time Trial or Time Attack competition, that need a single best lap only, their best lap time happens somewhere within the first five laps of a session.

There is nothing wrong with being in learning mode; all drivers go through it at a new track or in a new car. However, it pays to be aware of whether you are in the learning mode or not. While you are in learning mode, you are improving your driving by yourself. When you leave learning mode, you are no longer improving. This could be because you reached the performance potential of your car, or because you lack knowledge or skill that would allow you to go faster. Looking at reference lap times (i.e., what the best drivers in cars similar to yours have achieved) will help you determine which of these possibilities is happening. If you are not learning and you are not at the limit of your car's ability, you will benefit from coaching.

Theoretical Best Lap Time

To get this you need a data acquisition system which allows segmenting the track. With such a system, the best lap times for each segment over a session can be combined to give a "theoretical best" - the lap time you would have obtained had you driven the entire lap at your best.

The theoretical best is an indicator of consistency, and whether you are learning, similar to delta between lap times within a session. Unlike delta between lap times, theoretical best will point out mistakes made in different corners, where simple lap times might indicate a consistent performance.

It is important to note that theoretical best should not be used in lieu of actual lap times. At HPDE level theoretical best remains theoretical. However, with experience you can estimate your best possible lap time given your theoretical best, in a situation where you never have a clean flying lap the entire session due to traffic.

Lap segmenting allows you to perform consistency analysis on individual corners.

Tagged: intermediate, advanced