Pat Symonds on the secrets of F1’s speed data
THE DATA SIGNS ARE GOOD FOR A CLOSE BATTLE
After the last test of the winter, or as was the case this year the only test of the winter, fans are always asking themselves what the pecking order is and how close is the competition going to be in the coming season. Even the teams, which actually have very little to gain materially from such knowledge, spend hours analysing every run their competitors make.
The understanding they gain does not impact on the work they do – one always works at the limit to add performance to a Formula 1 car, whether it is the class of the field or sitting sadly on the last row of the grid. However, F1 personnel are hugely competitive people for whom success is measured only by a tally of points at the end of the season, and they really can’t help themselves from gauging their likely success as early as possible.
I always advise that one should not form opinions on such things based on a sample of one, and therefore never read too much into winter testing times. Now, however, with a few races under our belt, what can we see and how might it affect how the season progresses from here?
It certainly doesn’t need close inspection of the data to see that we have a much more competitive front of field this year, although it is hard to determine if the cars are overtaking more – the number of overtakes both with and without DRS were similar in Bahrain to 2020, and while the statistics show a lot more overtakes in the next two races, we have to take into account the weather at Imola and the extra DRS zone in Portugal.
We have been tracking overtakes in some detail for four years now, with more precision than using lap charts, by utilising the 200m timing loops at each circuit. The propensity to overtake is very much a function, at the moment, of exactly how the race plays out with Safety Cars and other such disruptions than it is of the closeness of competition. Let us hope the better wake characteristics of next year’s cars changes this.
What we can begin to see is the strengths and weaknesses of each team. To do this we again use the 200m timing loops arranged around the circuit, and we classify each one of them where the car is not on full throttle as being part of a low-speed, medium-speed or fast corner. If the car is on full throttle and there is significant cornering, then we classify the section as power limited. If there is no lateral acceleration, then it is classified as a straight.
The choice of low, medium and high speed descriptions is a little arbitrary but we choose to classify slow-speed corners as those with an apex speed of less than 125km/h, medium-speed between 125 and 175km/h and high-speed anything above that.
Caution is still required in interpreting the results. For example, Bahrain is thought of as a circuit with good fast corners but in reality, such is the performance of the cars now, only Turn 6 is actually classified as a high-speed corner. Other great corners, such as Turn 12, are now completely flat out, at least in qualifying, and therefore classified as power limited.
To say in Bahrain that Red Bull was clearly faster than Mercedes in the fast corners is true, but the overall effect on laptime is less than it would be at a circuit like Silverstone where there are more corners taken at over 175km/h that are not full throttle. To try and normalise this a little we express the difference in performance as a percentage time loss compared to the fastest in any particular type of corner.
Of course, depending on who you choose as reference the time loss could be negative. In other words, for example, while we might choose Mercedes as a reference Red Bull might be faster on a particular circuit and in a specific type of corner, therefore showing a ‘negative loss’.
With those caveats in place, we can start to examine the data. At the time of writing this is just for the first three races of the season, but there are some patterns emerging. It doesn’t need data analysis to
say that Red Bull and Mercedes are closer than last year, nor to say that Ferrari has recovered somewhat from the disasters of 2020, but some other trends are emerging.
Going to our slow, medium, fast-corner analysis we can see that Mercedes still holds sway overall, but this year it is only by around half a percent over the top four. In fact, in the fastest corners Red Bull holds an advantage while in the slowest corners, somewhat surprisingly, it is Ferrari which holds the smallest of advantages. This may bode well on some of the slower circuits later in the season.
While it is extremely difficult to separate downforce from sheer tyre grip, or drag from engine power, it would appear the latest Honda power unit is on par with Mercedes in qualifying trim, and that the Red Bull might be fractionally higher on downforce than their close rivals. Speed trap figures
“IN OUR SLOW, MEDIUM, FAST-CORNER ANALYSIS WE CAN SEE THAT MERCEDES STILL HOLDS SWAY OVERALL, BUT THIS YEAR IT IS ONLY BY AROUND HALF A PERCENT OVER THE TOP FOUR”
from the early races have shown more of a drag discrepancy than normal, the Williams for example being consistently faster on the straight than the works Mercedes with the same engine.
Data is interesting. Sometimes it enlightens and sometimes it merely confuses. When it confuses it is because it is either sparse, inaccurate or inconsistent. I think at the moment analysis suffers a bit from all three aspects, but nothing can disguise the fact that we are in the closest, most competitive season we have seen for a long time. Even the midfield, which has already been close for many years, is edging ever closer to the front. Everything is set for the best F1 season in ages.