Singapore Grand Prix: Friday Practice and Strategy Predictions

Posted on September 22, 2012


For the first time in a long time, there has been a dry Friday (well, practice 2 anyway) and I’ve had the time to write a post. So what can we say from the Friday activity at Singapore? Well, there are a few things to look out for from last year. Firstly, this race was the biggest performance differentiator of the year – the gaps between the cars were much bigger here than at other races. It will be interesting to see if things are a little more spread out than we’ve become used to in 2012. Also, the tyre degradation was severe last year, and although most teams are managing fewer stops in 2012 than we saw in 2011 (partly due to the harder tyre option being relatively more competitive), we are likely to see most teams do two or three stops depending on how far the harder tyre (the soft compound in Pirelli’s range) will run.

The assessment of the cars pace on the long runs in FP2 is complicated by the fact that the tyre life is going to be critical on the supersoft tyres. Only a very few cars were able to maintain anything like their original pace on the tyres – and given that the length of the first stint is going to be critical in assessing whether a two stop race is feasible, I would expect that many of the cars will be running more carefully in the first five laps of the race. Usually the FP2 stints can be reasonably well matched using a straight line (which is equivalent to the tyre degradation being about equal to the fuel effect – so the lap times are constant) on a ‘race’ history chart. However, in this case, this is only true for a few laps in most cases, with varying levels of performance drop-off from as many as three laps in. Consistency is only really seen from the slower cars, or those who ran shorter stints (which does include Vettel and Alonso).

What I’ve done is to plot the laptimes as if they were a race, but to exclude slow laps from the assessment. This gives us the chart below (Webber, Massa, Vergne, Kobayashi, Senna and Pic did not provide data which I considered useful):

So what we see is that Hamilton is the fastest, and although not able to maintain the pace, is still substantially quicker over his stint than anyone else. However, no-one is able to get past 11 fast laps, and everyone sees some performance drop-off. Next up on pure speed is Raikkonen (about 0.1s down), but the pace drop-off of the Lotus is early (and dramatic). Grosjean clearly tried something different as he was much more consistent, but was 2.3s from Hamilton’s pace, which matched Raikkonen after his initial burst. It does not look very promising for Lotus.

The next cluster of cars are Button (about 0.5s from Hamilton’s initial pace), Vettel (+0.6s), Alonso (+0.6s) and the Force India cars (Di Resta at +0.5s, Hulkenburg at +0.8s). The pace is best maintained by the Ferrari, which is the only car to still be getting quicker at lap 6 – this looks like a good race car, and the most likely to have flexibility on strategy. The degradation curves of the Force Indias match the McLarens pretty well, and they also ran out to 10 (fast) laps, so they look to be a strong bet for some good points. If anything, the Red Bull degradation is slightly better, but it does not stand out. I think that if there is a Red Bull on pole, it is looking like it may well be a rearguard action in the race.

Next up on pace are Perez and Kovalainen (both at 1.3s from Hamilton’s initial pace). Both are consistent as well, but it seems likely that Kovalainen is testing a lower fuel load (he had a slower very short stint – and Petrov is nearly 2s slower). This would put Sauber on the fringe of the points – I don’t expect that they will be coming through the field in this race.

Even with their significant exhaust upgrade, it is not until we get to a full 1.8s from the pace that we find the Mercedes cars, along with the Williams of Maldonado. Indeed Rosberg is even two-tenths slower than that. The Williams is one of the more consistent cars on pace, and runs the longest stint of all, but the Mercedes laptimes just deteriorate. They are about the pace of Grosjean (and Raikkonen after his fast opening laps), but their pace drops-off even from there. It is (whisper it) possible that Mercedes could be the eighth fastest car at Singapore – I expect that they will find something overnight, but it will need to be a lot to get near to the points.

Then, at the back, we have four teams quite close together on the stint times. Ricciardo (+3.1s) and Petrov (+3.2s) are very closely matched, with Glock and de la Rosa (both +2.8s) ahead of them. The pace drops off the Marussia and HRT very quickly (and Karthikeyan is over 1s slower) though. This suggests to me that Caterham could trouble Toro Rosso for the first time in a number of races – especially as their tyre use looks pretty good, and that the Marussia upgrades seem to be doing well. We could see the best performance to date from the newer teams – it could even be a three-way fight for ninth fastest team.

There do seem to be large performance gaps between the teams. Let’s see if it translates to qualifying and the race as it did last year.


The strategy at Singapore will clearly be affected by the safety car. Last year, three stops was the way to go, and the first stops (for those on the supersoft tyres) came at about between laps 8 and 14. From what I can see in the Friday practice data, we are looking at much the same in terms of tyre life for the first stint. This is interesting because it makes a two stop race marginal. Last year the soft tyre (but the harder compound) was good for about 16-20 laps. With the race being 61 laps long, and the pit stop time loss being 30s (the longest of the year), being able to run a two-stop race (with the extra flexibility it gives for the highly likely safety car) would be advantageous. But this requires three stints of 20 laps, or getting 25 laps out of the soft tyres if only 10 can be managed on the supersofts.

Assuming that we have qualified on the supersoft tyres (otherwise start on softs and see how far you get) and are reasonably confident in our tyre performance, I would be looking to get as far past lap 10 as we can (lap 13 is good) and to plan on two long stints on the soft tyre. The alternative (as run by the tyre-troubled Mercedes last year) is to run three stints on the supersofts of gradually increasing length (they managed 10 laps, then 15 laps and before the safety car ruined that strategy). The evidence is that this will not be quite as easy this year, as the tyre life is (if anything) slightly worse.

The pace gap between the tyres was 0.5s last year – which was much smaller than predicted. The qualifying difference was about 2s, and the race difference is usually around half of the qualifying difference. This needs watching in the first stint – my guess is that the pace difference between the tyres will again be of the order of 2s in qualifying (based on the FP2 times of Red Bull/McLaren), and there is no reason not to suppose that the effects will be similar in the race to last year. Therefore, the most likely strategies can be simulated to give the chart below. I have assumed that the tyres last 10 laps and 20 laps respectively, and that the phase 2 degradation is 0.3s per lap – about last year’s average value:

From this, it looks like the two-stopper has about 6s on the three-stopper. The curve ball is the safety car. Because the ‘phase 2’ degradation will be critical to the race time, the key is to make sure that the longest stint is not too long. Therefore a safety car which comes a few laps earlier than you would want to make your longest stint is the worst timing as this will leave you vulnerable at the end of the stint, or needing to make an extra stop. Delaying the stop (as indeed Schumacher/Webber did in Valencia) is probably the way to go here.

This one could be fascinating – and it could be a strategy that dovetails perfectly with a safety car that becomes the story.