Basic Knowledge - Video and Data Analysis

Hi folks,


I'm currently on a plane over the Atlantic Ocean on my way to the IMSA GT3 Cup Challenge at Road Atlanta, where I'm supporting a driver in the so-called "Gold" class, just like last year. Vehicles in the GT3 Cup "Gold" class are the first generation 991 GT3 Cup: 3.8 ltr. engine, 450 hp, no ABS, no ESP and no traction control - so a real driver's car.

Usually I drive the car on Thursdays during the so-called promoter test to get the setup and reference laps sorted. Friday to Sunday the actual driver drives and I supervise him in the area of video and data analysis, also I am responsible for him on the radio as a "spotter" and give him important information during the different sessions via radio. The difficult part is then the follow-up of the different sessions, where I evaluate the performance of my driver with the help of data and videos and give him tips and clues to improve.


The data is recorded via the Cosworth system included in the car, and in the end almost everything is recorded: Steering wheel movement, accelerator pedal position, brake pedal position, longitudinal and lateral acceleration, vehicle speed, individual wheel speeds. In addition, the whole vehicle vital values such as water temperature, oil temperature, voltage, transmission temperature etc. are recorded. I could go on for hours about all the infos that are recorded :-)

As a Driver Coach you are of course limited to the values that come directly from the driver, but with the data that I have mentioned above I can overlay my driver's laps with mine to filter out differences:

- brake pressure
- Time difference - how the distance builds up over the rounds
- gear selection
- Accelerator pedal position
- Speed
- Vehicle speed
- Steering angle
- Wheelspin

With these simple values you can do quite a lot and work out many basic things, e.g. how different the brake points are (brake pressure), how the driver builds up and reduces the brake pressure etc. With "Compare Time" you can see for example if the time difference is building up constantly over the lap or if there are certain points where the time difference is created. Additionally you can dive a little bit deeper into the data, e.g. for the accelerator pedal there are the following evaluation possibilities:

- Throttle on distance = how many meters per lap the driver presses the gas pedal
- Wheelspin = how big the maximum wheelspin was in the round
- Coast Time = How many seconds per lap the rider does not use the accelerator or brake
- Coast Distance = How many meters per lap the rider does not use the accelerator or brake
- Full Throttle = How many seconds per lap the rider is WOT (Wide Open Throttle)
- Coast Time Peak = the maximum time the rider is in the coasting in one piece

 There are also many evaluation options for the brake:


- Brake On Distance = how many meters per lap the driver has applied the brake
- Brake on Time = how many seconds per lap the driver has applied the brake
- Brake on Percent = how many % of the lap the driver has applied the brake
- Brake Press Front = the maximum brake pressure of the front brake circuit
- Brake Press Rear = the maximum brake pressure of the rear brake circuit
- Brake Bias Gated = the actual brake balance during braking
- Speed FL / FR Locked = how many seconds the corresponding front wheel has been stationary in the respective lap


just to name a few:

For example, you can also display a histogram for the accelerator pedal, which shows the percentage of the accelerator pedal position over the lap:

 Then there is e.g. the so-called G-G diagram, which represents a combination of lateral and longitudinal centrifugal forces:

Basically this diagram should look like a heart, depending on the track it can sometimes vary a bit. The data shown here is a test from Road Atlanta and there are hardly any left turns with high centrifugal forces, so the deflection for left turns is not as high as for right turns. With the G-G plot you can see if the driver takes everything out of the tires and the car or if you are having handling issues you can also see in which areas it effects the performance the most.

In the end, there is really a lot of data that can be evaluated, but does more mean better? That is indeed one of those things.


Basically it is not bad to have a lot of information available but there is, as with many things, a "sweet spot" - if you go beyond this point the analysis work becomes very complex and the added value is relatively small. In addition one must also consider that the topic time plays naturally also a role. For example, if there are only two hours between two sessions, then it will be very tight to read out all data, evaluate it and give the driver a feedback that he can use to do something. In addition, video recordings will also be viewed and discussed.

This is the next point I want to talk about: With a simple GoPro camera you can really do a lot. Line selection, the way a driver drives the vehicle, comparison of brake points etc. - all this can be done, of course in a much slimmed down version, even with pure video recordings a lot can be recognized and in my opinion a full analysis is only possible with both things: video and data. Sometimes there are situations on the track, such as traffic, which are difficult to recognize on the data at the first moment. It helps if you have the corresponding video at hand, and vice versa, of course.


You got questions or you are interested in a coaching session? Then feel free to contact me! ;-)


So long,