Potential of Machine Learning in Formula One.

I have been a long time fan of Formula One: the excitement, the glamour, the prestige. What the race car boils down to is technology, mechanics and investment into new technology that progresses the sport and arguably the human driving experience. This piqued my interest. How does data fit into this fascinating world of adrenaline?

On any given weekend you will see the drivers in the garage, snug in the cockpit of the carefully designed F1 car. A lot of people say it’s the design qualities of the car,however, around them are monitors of data analytics to give the driver the extra edge on their next in line competitor. One thing you will notice is each of the 10 teams have their own data analytic company partnership: Redbull has Oracle, Mercedes AMG-Petronas has TIBCO, Renault has Microsoft, Ferarri has Palantir and this is just to name a few. Whereas F1 as a whole is powered by AWS.A giant in the world of data analytics.

“This project with AWS was one of the most revolutionary in the history of Formula 1 aerodynamics,” said Pat Symonds, Chief Technical Officer of Formula 1. “Nobody designs a car to come in second, but for this CFD project we were looking at how cars perform in the wake of another, as opposed to running in clean air. We have been able to use AWS technologies to understand the incredible aerodynamic complexities associated with multi-car simulations.”

With all this data, I was curious what other information are driver being given and what analytics are these big data companies performing throughout a race weekend?

On race weekends you can be certain that teams use information based on wind tunnel testing, computer simulations, tyre choice, ambient temperature and precipitation level to name a few. In all of these aspects Machine Learning is involved and is more then likely to be around for a long time as data analytics are only increasing in use.

“In using Palantir, we have basically taken all of the complexity that lies around managing such a vast amount of data out of our hands. And we really focus our time on extracting value from the analyses that we run, focusing on the product that we develop.” — Marco Adurno, Head of Vehicle Performance, Ferarri.

To start off my exploration in the many seasons of F1, I merged all the tables together to obtain all relevant information. Once the ultimate merge was complete, I began to explore!

First I was curious to see how many F1 Drivers came from their respective country. As you can see below, German, English, and French Drivers were the most popular countries to have a seat in F1. Historically this makes sense, as F1 is heavily UK based, but originally the racing league started in a small town called Pau, France.

After seeing what countries had more drivers, it was a logical step to see what drivers from these countries have dominated over the last decade. Below is a bubble plot to show the magnitude of the wins from a given driver.

Dominating drivers of the last decade

You can clearly see Sir Lewis Hamilton has been a dominating factor in the last decade along with Sebastian Vettel. However, this made me curious as to who was the overall dominating racer through-out the years. Again as you can see below, Lewis has taken charge and has surpassed Micheal Schumacherwho is considered one of the best to get into the cockpit of a F1 race car.

Dominating Drivers in History of F1

However, this year marks the new era of Verstappen. The young RedBull Driver is the rising star that just eclipsed Lewis Hamilton in the 2021 Constructions Championship. To the left is a visualisation of fastest lap times. To me the drivers/teams that achieve these fastest lap times are good predictors of Sunday performance, and ultimately points to be scored for the team.

I further want to test whether these top teams were consistently getting the fastest lap time. To do this, I created a data frame from the csv files available on kaggle. My goal is to predict what team and driver is likely to win in the coming 2022 season using a Neural Network and classical machine learning algorithms.

In conclusion, all of the teams in Formula 1 and even Formula 1 itself, all have data companies helping service their big data needs. From every aspect of the race weekend; such as assuming tyre choice based on weather APIs or predicting win outcome with deep learning models. This all shows how much the world of technical sport has come to rely on the advancement of machine learning.

Related posts:

Cut off from the noise of daily life and find moments of deep relaxation. Inhale and infuse yourself with calm, exhale out the stress. When was the last time you truly relaxed? Took some time off…

I am 20 years old and i need 500 dollars to pay for some fines i got in trouble for. I used to smoke weed i got caught. Im through that like i hate to sound corny and **** but it did help me find god…

As the gangplank of life is walked, I remember that youth lasted an hour, maybe less. Stars and signposts led me through the loves and the lies, so many collected, others lost, broken.