Retro: 2011 Australian Grand Prix

Posted on March 15, 2012


It’s a little difficult to do a data preview of a race, and I guess the best thing to do is to give an idea of the race analysis which can be expected from this site. So where better to start than the 2011 Albert Park race, as it also gives a direct comparison (once the 2012 event has happened) of where the teams are now in comparison with 12 months ago.

By using the intelligentF1 model logic, the race can be analysed allowing the underlying pace of the cars, the fuel effects and the tyre degradation to be extracted. This gives a better impression of the relative strengths of the cars, based on physical modelling, than looking at individual fastest race laps or race finishing order. The optimal strategy (in hindsight) can also be determined, and we can work out what could have happened had different strategy decisions been made.

So what might we have learned from an intelligentF1 analysis of the Australian Grand Prix 2011? Well, I’m going to put up a few plots to show what can be seen from this analysis of the data, and to get an idea of what we could learn from the 2012 race, and what implications that could have for the rest of the season.

Case study 1: Was Vettel much faster than Hamilton?

The race history chart, with model fits (the model cannot account for losses in pitstops or lapping, or incidents/spins – so it is the gradient of the line which counts for the pace of the car) which are pretty good and consistent from stint to stint.

What we see is that the best fit for Vettel and Hamilton happens to be identical on the soft tyres, with Vettel having a slight edge on the hard tyres, although this may have had something to do with the teatray damage on the McLaren as Lewis was 1.4s slower on the harder tyres when the average was 1-1.2s (as determined by the model fits). With this being the case, the time lost by McLaren was partly due to time lost in the pit cycle, but the big news was that the McLaren was losing time (curve bending away from the fit) in the second stint due (probably) to increased tyre degradation (although it could have been fuel saving). The Red Bull pace was very consistent, which could suggest more in hand, but there are no clear places (not even the opening laps) where there is a clear signal of extra pace from Vettel. This boded well for a close season in terms of race pace between the top two teams (which on the whole we had).

Case study 2: Two stops beats three

Petrov claimed a podium finish last year, beating Alonso and Webber, but how competitive was he in terms of laptimes, given he finished well behind the front two? Was two stops really that much faster? Well the chart below, only the race data is included for clarity, but the fits have been done, and provide useful insight into the relative performance of the cars.

Petrov was not slow. But he did not have the pace of Webber or Alonso. Indeed, of the three, it is the Ferrari which shows the fastest underlying pace, a pace which is about the same as that shown by Vettel and Hamilton. Webber is about 0.2s per lap slower, and both are quicker than Vettel on the hard tyres – which suggests to me that Vettel was taking it easy by the final stint. Petrov? Well his pace is 0.5s back on Vettel’s pace on the softs, and about 0.2s down on the hards. So competitive enough to be a nuisance, but only if he was able to make one fewer stop (as he did). On pure pace, however, it was clear that the Renault was slower – so this suggested that strategy could make a significant difference. Picking where a different strategy could have worked is something which has been seen a number of times in the races analysed using the intelligentF1 model.

Case study 3: One stop wonder

The talking point after the Australian race last year was the tyres – how the degradation was much less than anticipated, with the one-stop race of Sergio Perez coming in for much attention. His race (and the fits) are shown below.

The promise shown by the Mexican was not so evident in the first stint on hard tyres. Indeed, it seemed that the Sauber was struggling in the early laps. It was slow on the hard tyres relative to the softs (1.7s slower by my model), but the car was quick on the softs – as fast as Petrov (about 0.5s slower than Vettel), and the degradation was sufficiently low that he was able to increase pace 20 laps into the stint. It’s also clear from the chart that the tyres held on until the end, completing a very impressive debut for Perez.

Overall pace assessment:

By fitting each of the cars’ data, the underlying pace can be determined, and a league table of the race pace can be provided. Comparison of the 2011 table with the 2012 version (for the same race) should give us some ideas of how the teams have progressed in the last year. The table from 2011 looks like this (on the softer tyre):

Red Bull: Vettel +0.0s, Webber +0.2s

McLaren: Hamilton +0.0s, Button +0.3s

Ferrari: Alonso +0.0s, Massa +1.3s

Mercedes: Rosberg +1.5s, Schumacher +2.5s (damaged)

Renault: Petrov +0.5s, Heidfeld +2.0s (damaged)

Williams: Barrichello +1.5s, Maldonado +1.9s

Force India: Sutil +1.4s, Di Resta +1.5s

Sauber: Kobayashi +1.2s, Perez +0.5s

Toro Rosso: Buemi +1.1s, Alguersuari +1.5s

Lotus: Kovalainen +4.0s, Trulli +4.5s

Virgin: Glock +4.5s, D’Ambrosio +5.5s

The surprise is that the teams were much closer than we might have imagined – the gaps were about 1s larger to the midfield by the end of the season. Let’s see what we get this weekend. There will be a post on the long runs from Practice 2 at some point tomorrow.