Formula One is a sport driven more by data than any other. Despite austere regulations to restrict performance and development, there is no doubt data steers the design and engineering of the cars and plays a dominant role in managing the on track action.
While data has always been a part of F1, its acquisition and analysis are today as turbocharged as the engines. Witnessing firsthand the way teams been able to make use of data over the last two decades is 13 times Grand Prix winner David Coulthard. Throughout his 15 year Formula One racing career that started in 1994, he saw teams embrace and leverage all sorts of technology, from 3D printing to lightweight materials. However, none have had a bigger impact on the engineering of the cars than data harvesting and advanced analytics.
“Since I started in Formula One it has gone from conversations around the coffee machine and intuition, to all kinds of information being harvested in real time and sent back to the pits,” said David Coulthard, talking at a recent PTC event in Stuttgart. “It is instantly looked at by a team of data engineers at the back of the garage, but it is also sent via satellite to team control centres all over the world where even more engineers and analysts will look at it.”
Now a BBC F1 commentator, Coulthard is able to draw on his 246 race starts to give viewers an insight past the glamour, into the heart of today’s competitions and in to the mind of drivers.
Drivers and cars are no doubt the focus of coverage, but they are the culmination of a much larger engineering effort. Taking Coulthard’s career as an example, he spent around 450 hours driving the car competitively, but more than 4,000 hours looking at data with engineers.
“One of the big issues with a lot of data is being able to take the good bits and not get lost in the noise,” he explained. “A lot of the data coming back is fairly benign, it says that the engine temperature is in the right window or tyre pressure is ok. So, being able to translate that data in to information that can be used to add performance is a large part of racing today.”
The use of ‘Big Data’ in F1 is in reality more discrete compared to other applications, generally being confined to two cars, over a weekend, and season. Despite this, the amount of data gathered is immense. Managing and analysing it requires teams of 50 or more, per team, per race. It allows teams to adjust parameters on the car for a safety car or yellow flag, adjust for aging tyres, and even make subtle tweaks to brake balance and differential settings on a corner by corner basis.
“It doesn’t matter if it is Grand Prix racing or advertising, whatever product or service you are proving to a customer, it is all about being first to market and best in class, which will see you having continued business,” said Coulthard. “In F1, it is about finding anything that will affect your pace. How you react and manage that makes the difference between success and failure.”
And this is also a key difference when looking at the way F1 and other industries look at and analyse data. During a Grand Prix analysis needs to be almost instant as live data is streamed in real time from the car.
“It is very much about the here and now,” said Coulthard. “Teams can’t reflect after the event and come back to rerun a GP the next day. They need to take performance out of the data during an event. How will the changing wind and temperature affect the cars performance as they go around the track on a given set of tyres and fuel?
“Teams are much better at planning race strategy and optimising when they make pit stops, which are mandatory.”
All this refinement doesn’t come without an additional burden. The workload inside the cars is now arguably at its most intense ever, with drivers having to make precise on the fly adjustments as they are racing wheel to wheel at 200mph.
“If you look at a modern day steering wheel, there are 40 odd buttons so the driver can self engineer the car on track,” said Coulthard, something that has changed considerably since he first lined up on the grid. “When I first started in Formula One, we had two or three buttons for things like radio and the drinks button. There were very few parameters you could alter once you had taken to the track, and very little that could be analysed when you were off it.”
The use of data to bolster car performance is no doubt something that is only going to continue. While driver performance is important, getting a good car is a much larger piece of the competitive puzzle. While it is difficult to put an exact figure on it, the generally agreed rule of thumb is that the car makes up 80% of performance, and drivers the remaining 20%.
“To win today, it is all about optimising the technologies available,” he said. “This is where data really comes in to its own. Data doesn’t lie. It tells us the facts about performance.”
CV – David Coulthard
David Couthard began racing in Formula One in 1994 following the untimely death of Ayrton Senna, who he succeeded at Williams. He won his first Grand Prix in 1995 and then moved to McLaren in 1996. After winning two races in the 1997 season, he then finished 3rd in the World Drivers' Championship in the 1998 season. He moved to Red Bull in 2005 and secured their first podium a year later. Coulthard retired from Formula One racing at the end of 2008.
He continued working with Red Bull as a consultant and joined the BBC as a commentator and pundit for their coverage of Formula One.