(Motorsport-Total.com) – The list of the fastest Formula 1 drivers recently published by Formula 1 in collaboration with Amazon Web Services (AWS) has generated some criticism. Not only that drivers before 1983 were not taken into account, but that surprising names like Heikki Kovalainen or Jarno Trulli were in the top 10, but not respected qualifying experts like Mika Häkkinen or Juan Pablo Montoya, raised eyebrows.
Sports director Ross Brawn and technology and data expert Rob Smedley are now trying to shed light and explain how the list of pilots came about. “There were a surprise or two, but if you dive deeper it makes perfect sense,” says Brawn at formula1.com. “There has been a lot of discussion, but if you understand the methodology, people will understand.”
The basic plan is that teammates are compared with each other in qualifying: “Same day, same situation, same opportunity,” said Brawn. The data is collected all the time, and when a driver changes teams, you get new data. If driver A beats driver B in one team and then beats driver B driver C in another, then A is logically faster than C.
And with enough information, you can then create virtual pairings – such as Jean Alesi and Pascal Wehrlein – and calculate who would likely win.
Why Trulli and Kovalainen are in the top 10
“You get this complex network of drivers, and with the virtual connections we can build a confidence level that tells us how accurate the connections are in the network,” says Smedley. “It’s not just about the distance between teammates over a season, but also the accuracy is achieved by using the network.”
A big point of discussion were places eight and nine for Heikki Kovalainen and Jarno Trulli, who together only have five pole positions. Trulli was considered a brilliant qualifier, but according to the former Renault engineer and sports director Steve Nielsen, he was never able to deliver more than five laps in a row. “That’s why he never became world champion,” says Brawn.
But he doesn’t want to criticize the Italian, he just wants to explain why he is so far up on this list. “It’s about the fastest driver, and the fact that some of them weren’t world champions is because their other skill set wasn’t strong.” This also had an impact on Kovalainen: Because Trulli was so strong in its heyday, but was beaten by Kovalainen at Lotus, the Finn is so high up.
Thanks to data analysis: Hülkenberg almost at Mercedes
Brawn even admits that he worked with data analysis as a team manager at Mercedes and that as a result, Nico Hülkenberg (twelfth place) almost landed in the Silver Arrows. “Lewis was the logical decision after Michael, but we made numerous analyzes of other drivers and ended up with Nico Hülkenberg,” he says.
Chief strategist James Vowles analyzed Hülkenberg’s entire career – but in the end the driver pairing was Hamilton and Nico Rosberg.
In the future, the two would like to use data from before 1983 to bring Juan Manuel Fangio or Jackie Stewart into the system. The problem: These are handwritten and have to be entered first. “That can be done, but it’s a question of capacity,” said Brawn.
Finally, Smedley stresses that this list should in no way be considered definitive and non-challengeable. “It’s about coming up with something that we think is pretty objective,” he says. “It’s about math, physics, data and machine learning – and then discussing and arguing about it.”