Mercedes, Renault & Sauber – Q1 Analysis
Part three in the qualifying analysis series of posts takes a look at the performance of Renault, Mercedes and Sauber, three mid-field teams hoping for performance gains that will take them closer to the front.
The method is the same as that used in the earlier posts and compares the fastest Q1 time at each GP against the fastest Q1 lap time for the quicker driver in each team. That might mean that Schumacher has set the time for Mercedes in 4 races and Rosberg in the other 15, essentially I’m trying to minimise the effect of a poor qualifying by one of the drivers in a given team. More details on the method can be found in the initial post.
Formerly Brawn GP, Mercedes should have started the 2010 season in top form fresh from winning the Constructors Championship in 2009, but, with two new drivers at the helm this was not to be the case. Mercedes were on average, 1.005% slower than the fastest team over the whole 2010 season with the gap varying between 0% at Belgium and 5.604% in Malaysia. The Malaysian result was so far out from the mean value (greater than 3 standard deviations from the average) that I’ve excluded it from my analysis and will concentrate on the remaining 18 races that year. The full results for all three teams are as seen in the table below; a second smaller table shows the effect of removing the outlier point from Mercedes and a similar outlier from Sauber (Belgium 2010) from the averaged data.
With the outlier removed, Mercedes were on average 0.75% slower than the fastest Q1 qualifier in 2010 and 0.952% slower in 2011; this represents a year on year loss by the team of 0.202% even though the team set the fastest Q1 time twice in 2011 and only once in 2010. Does the data shed any light on where these losses occurred?
It’s difficult to pinpoint where Mercedes started to lose ground against the leading team with 11 of the 19 races in 2011 coming within 1% of the lap time of the leaders. Of the remaining races, 7 were within 2% and 1 was within 3% of the time set by the lead team. These 8 slower qualifying sessions tended to cluster in the middle third of the season. In contrast, only 5 races in 2010 were between 1-2% slower than the fastest team and 13 were within 1% of the lead time. The regression line (which gives an indication of relative performance over the season) shows that Mercedes lost 0.11% of the fastest Q1 time at each race of the season and a larger margin of 0.04% at each race in 2011.
Renault were another team who were closer to the leaders in 2010 than they were in 2011 with an average lag to the lead team of 0.748 and 0.983% respectively, this equates to a year on year loss of pace of 0.235%. It would seem that consistency is key here; in 2010 they were between 1 and 2% of the lead time on 4 occasions with their slowest time being just 1.322% slower than the lead team. In 2011, those slower times had increased to 7 with the slowest time being 2.655% slower.
The 2011 season started well but there followed a series of races between Monaco and Singapore where the team started to lose ground. It’s appropriate to bear in mind that Senna took over from Heidfeld from Belgium onwards and this may have had an effect on driver motivation and performance. In the final 5 races of the season the mid-season dip was reclaimed. Overall, the linear regression shows that in 2010, Renault lost just 0.007% to the lead team at each race whereas this performance deficit had increased to 0.028% per race in the following season.
Renault and Mercedes 2011 season compared
At the end of the year, 92 points separated the two teams yet the qualifying analysis showed the season average difference between the two teams to be much smaller, just 0.031% in favour of Mercedes, perhaps race set-up and strategy play a larger part in the overall result than the qualifying would predict.
The two teams seem to mirror each other’s performance for much of the season. The only major differences in the data are the performances at the street circuits in Monaco and Singapore where Mercedes had the upper hand and Italy where the opposite was true. I find the Italian result particularly puzzling as in general, Mercedes engined cars had greater straight line speed. Perhaps someone else can shed some light on these observations?
Sauber, the final team analysed, were the only one of the three to have made an average performance gain in 2011 compared to 2010 (however this “gain” is lost when the outlier point is not removed). The year on year gain is tiny at just 0.053% but if you look closer at the data the general impression is of a team who have performed much better this year than last. The relatively poor performance at the Belgium GP skews the data somewhat but since that point is within 3 standard deviations of the mean it has been left in.
As with Mercedes and Renault it was the mid-season dip Canada and Belgium which proved crucial to the final result with Sauber ending the season some 0.323% behind Renault and 0.354% behind Mercedes. In terms of performance gains or losses throughout the year, Sauber kept par with the fastest team in 2010 being on average just 0.008% faster than the lead team at each race of the season. In 2011 the loss of performance at each race was much larger at 0.042% but as we’ve seen, they are faster this year than last so whilst their development rate has been good, the lead team has been better.
That just leaves one final qualification analysis to look at, the big three of Red Bull, Mercedes and Ferrari which I’ll post early in the New Year.
Note: I excluded one point from Mercedes 2010 data as it was more than 3 standard deviations (104.743%) from the average. Similarly, one point was removed from Sauber’s 2010 data set for a similar reason, 3 SD’s of Sauber’s average data was 105.452%, all these values can be found in the first table of results.
Credit: Mercedes GP Petronas