Well, well, well. Precise predictions from testing may be notoriously difficult to make, but as far as an approximate understanding of the pace from the testing data is concerned, I think that the work done using the long run data on this site was about as good as I could have hoped for. As a reminder I had:
Group 1: McLaren, Red Bull and Mercedes. It looks like there are only a few tenths between these teams (Red Bull had KERS issues with Webber and Vettel made an error, so they are really only a few tenths down).
Group 2: Lotus, Ferrari and possibly Sauber at perhaps 1 second down. This is harder to assess – Ferrari look to be the second away, perhaps a little more and Lotus look to be genuinely faster, possibly even on the edge of the first group. Sauber had some problems in qualifying and underperformed, but it seems that they do not have the pace suggested by the analysis of the test running and are clearly in group 3.
Group 3: Force India, Toro Rosso, possibly Sauber and Williams at about a further 0.5s down. It seems that the gaps are a little smaller to the front (which is clearly a good thing), but that Williams are better than I had predicted from testing (another good thing for the purists).
Group 4: Caterham. I had Caterham perhaps 0.3s behind the midfield pack, but still just about adrift. With the gap being closer to a full second in qualifying, the hope is that they can show a bit more in the race.
So, on balance, I think that the pace predictions from the analysis of the long runs in testing was not bad at all. Sauber seem noticably slower (although we will see what their race pace is like), as do Caterham, and Williams (and you can argue for Lotus as well) are noticeably quicker. Everyone else is well within 0.5s of where I would have put them. Not perfect, but it makes me feel that the time doing it was well spent.
Now the title of the post says strategy predictions – so I guess I’d better do something on that or face accusations of false representation. There were only a few long runs to go on in free practice 3, which don’t really give too much away for the race in terms of the relative race pace of the cars, but they do show that the degradation is not high. The runs only stretch to about 12 laps, but the times are getting faster within the stints, suggesting that the degradation is not too high. With the medium tyre being about 0.5s slower than the soft, this makes things interesting. The penalty for using tyres which had previously been used in qualifying was consistent at about 0.3s per lap in 2011, but there are suggestions that the penalty may be greater this season – it will be interesting to see if a reasonably accurate value can be extracted in the intelligentF1 post-race analysis. Whether the penalty is 0.3s or 0.5s, it will make the medium tyre competitive as a race tyre, which indeed opens up the race options as intended by Pirelli. All good.
I have done my usual analysis of the strategy for the race, and it so happens that my model does not give the same results as those presented by Gary Anderson in the BBC programme. He has three stops faster than two stops – which I only get if the ‘phase 2’ degradation kicks in after less than 15 laps. As there is no sign of this in the free practice running, I am inclined to think that the tyres will last. In fact, if the mediums last better than the softs in the first stint (assuming someone starts on mediums so we can tell), a soft-medium-medium race may be the way to go. I get that this would be about 3s slower than soft-soft-medium if there is a 0.5s gap in pace between the tyres. So I expect a two stop race for most, and I think that I would be thinking in terms of using mediums for the majority of the race. The graph of the simulation is shown below, with the codes for the tyre choices in the key being ssm=soft-soft-medium and so on.
If the first set of softs get past about lap 14-15, then we should be in for a two-stopper. And it will be interesting to see if there are many takers for the mediums at the first stop – or indeed the start.
Let’s see. Given that this is the first race of the season, and I get something different from the experts, I could easily be wrong.