Brazillian Grand Prix: Story from the Data

Posted on November 29, 2011


Whilst not a thrilling race, the laptime data from the Brazillian Grand Prix tells a number of stories which are not immediately obvious, nor reported widely. The most interesting of these is that post-race analysis reveals that an unused strategy choice could have gained points for a number of the midfield runners. It just goes to show that these things are much easier in hindsight.

Before the race started tyre wear was expected to be low, and a two stop race was widely predicted, including at intelligentF1. Strictly speaking, the degradation was almost exactly as predicted from the Friday practice in terms of pace loss from the faster soft tyres – but only for about 10 laps. Once past this age, the tyres went into a gentle ‘phase 2’ where the pace did not go off a cliff, but became steady at a slower pace. This is particularly interesting because the harder medium tyres, which were about 1s slower on average, did not have this issue, and so they were the tyre of choice for the longer stints. The fastest strategy, given the loss of pace from the softs and the slow speed of the harder tyres was to do three stops (soft-soft-soft-medium), but the next quickest was two stops (soft-medium-medium). There will be a number of teams (Renault especially) who could have gained significantly doing this.

To the race, then. At the front it was a competition between the Red Bulls – although Button could have got close to Vettel had he gone for medium tyres at his first stop (there will be a piece on this here later this week). Vettel may have had pace in hand with his gearbox problem – it is absolutely certain that Webber did. Fuel corrected he is 0.2s per lap slower in the second stint, and another 0.3s per lap slower in the third stint. His pace on the mediums at the end was good, but there was more to come if needed. The win was even more crushing than it looked. It’s worth showing the race history chart of the Red Bulls just to see how good a driver Vettel is.

Vettel’s is the blue trace without the crosses. Notice that the first nine laps are fast, and then the pace drops significantly. This is most likely to his learning how to drive quickly whilst protecting the gearbox – part of it is tyres, but the shape of his curve is different from the other cars – it is clear he has a problem. It takes him until a few laps after his stop to work out how to go fast – then he picks up pace to being a few tenths down on Webber. Once Webber is past, Vettel’s trace settles down to the beautifully consistent curves that we are used to seeing from the World Champion – he has worked out how to drive around the problem. Impressive.

The battle behind was more fierce. The charts for the Ferraris and McLarens are shown below.

There are a few notes of interest. It is clear that Jenson is slower than Lewis in the second stint – however, just as Hamilton is set to go longer than his team mate and take the place, he has three horrible laps (30-32) before his second stop. On lap 30, the gap was 1s, after the stops it was 2s. He lost 2 seconds in those two laps – a reasonable amount of it lapping one Michael Schumacher. The gearbox problem rendered it immaterial, but Lewis was faster here. The pace of the McLarens on the medium tyres is also worthy of note – Button is 0.6s slower on the mediums (or 0.3s used soft to new medium) whereas the Ferraris are 1.2s slower (0.9s used soft to new medium). That’s a very big difference. Massa is also massively slow in his middle stint – I had originally thought he must have been on medium tyres. It looks like he was being very careful knowing that he had to two stop – but 0.7s from your team mate for lap after lap is a lot.

Working our way down the field we come to the Sutil/Rosberg battle. Interestingly, Di Resta was just as quick as Sutil, but his track position didn’t allow him to do a three stop. He could have tried it – but, as we will see, it wouldn’t have worked out. The chart below shows the Force Indias, Rosberg and Petrov, for reasons which will become clear.

In the first stint Sutil has no problem staying with Rosberg, and is able to undercut him and drive away. At this point it is clear to Mercedes that the Force India is just plain faster, and all the opposition (at this stage Petrov and Di Resta) are also on softs. Force India are unable to get their tyres to last quite as long as Mercedes, but it does not matter as Sutil (just) has enough of a lead over the Toro Rosso/Sauber battle to make a second stop and exit the pits ahead. Rosberg does not – he stays on a two stop in the hope that track position in the final stint may be decisive. It wasn’t. Don’t think that Force India beat Mercedes on strategy here – they did it on pace.

Renault decided to follow the faster (in theory) three stop strategy with Petrov, who was ahead of Di Resta having undercut at the first stops. The Force India was faster, but was stuck behind the Senna/Schumacher squabble so did not pull away. But fast in theory is not fast in practice – Petrov is quick for three laps in his third stint – then he catches the traffic and gets stuck. He did not overtake either Toro Rosso and instead of catching Di Resta to make up the pit stop, he laps slower losing the 10s which separated the cars at the end of the race. If either Di Resta or Petrov had selected the medium tyres at their first stop, they would have been quicker over the second stint (note that Di Resta is slower in the second half of the second stint on the softer tyres), and could even have threatened Rosberg for seventh. Indeed the intelligentF1 model suggests that Di Resta would have caught Rosberg by the end – Petrov may not quite have had the underlying pace. Renault lost out through strategy here, which is a recurring theme of the last few races. They did get the last point, though.

Points also went to Kamui Kobayashi cementing Sauber in seventh place in the constructors’ championship.The Saubers were quicker in the race than the Toro Rossos, and Kobayashi passed Buemi on track in the early laps, something his team mate never managed. The chart (including Petrov again) is shown below:

The pace of Kobayashi is much more impressive in the second stint – this is consistent with the Saubers having saved new tyres in qualifying. I have no confirmation that they did this, but the laptimes of both their cars in the early laps of the second stint are faster (according to the intelligentF1 model) than they should be by just the right amount… Perez lost his chance with a spin just before Petrov got himself stuck behind this battle. Indeed, the Renault stopped early, undercutting Alguersuari to get the final point. The Spaniard for his part was faster than his team mate in the second stint, and he was able to make the tyres last longer. Having just squeaked past Buemi at his stop, he proceeded to drive off into the distance and got pretty close to Petrov. If Toro Rosso had stopped Alguersuari instead of Buemi at the first stops and undercut Perez, he may well have been able to get the last point. Either way, it was a good drive – in fact his performances in the second half of the season have been consistently strong.

Behind this, we have the Williams, the Lotuses, Senna and Schumacher. Williams were just not fast enough, and Lotus were about 0.5s further behind – closer but still not quite close enough.

The pace difference between the Williams and the Lotus cars can be seen comparing Kovalainen’s and Barrichello’s traces. Kovalainen is again in a nomansland – comfortably ahead of Trulli (35s) and comfortably behind Barrichello (60s). The Finn lost time on the medium tyres – Lotus struggles on the harder tyres were worse than Ferrari. After his (harsh in my opinion) drivethrough for the Schumacher incident, Senna found himself in battle with Maldonado on hard tyres. Once the Williams crashed, the Renault began to have gearbox issues – Senna’s pace drops by 1.5s per lap, and he holds up Trulli for a number of laps, and is hardly faster than the Lotus at the end. Schumacher’s recovery drive was reasonably strong – especially on the medium tyres where he was faster than Rosberg. He was, however, not as fast on the softs.

And finally, right at the back, we had some good performances from drivers who can’t be certain of being in F1 next year. Jerome D’Ambrosio brought his Virgin home comfortably ahead of the HRTs and Tonio Liuzzi left Ricciardo behind, which he hasn’t done for  a while.

In the first stint Glock held up the Australian, but the growing gap between Ricciardo and Liuzzi in the second stint was due mainly to pace, and partly to time lost being lapped. Liuzzi would have won the intra-team battle but for his gearbox issues. If this is to be either D’Ambrosio or Liuzzi’s last race, they can be pleased to go out on a high.

Another season in the books. And finally a win for Webber. Over the next few weeks, intelligentF1 will be looking back at races earlier in the season (especially before the summer break when the model was written) and compiling the underlying pace trends over the year to see what this can tell us about the development race. Also, the development of the intelligentF1 model will continue, and there will be posts on the progress made with the model.