A few more thoughts on Barcelona testing…

Posted: February 27, 2012 in Formula 1, Pre-season tests, Teams
Tags: , , , , , , ,

In our previous blog entry, we analyzed the race simulation that Rosberg did. How does it compare to the real thing, though, i.e. the amount of time it took Vettel to complete the race distance in 2011? Let’s examine that:

Vettel finished the race in 01:39:03.301 hours. This, however, included four (4) pistops: Lap 9, Lap 18, Lap 34 and Lap 48. Thanks to F1 Fanatic, we know that Red Bull’s average pit stop time was around 20 seconds, we can therefore safely assume that the clear race time is 01:39:03.301 – (4 x 20 seconds) = 01:37:43.301. This, over a period of 66 laps, would give an average laptime of 01:28.838.

Let’s see how this compares to Mercedes’ race simulation. Nico did 65 laps, at an average lap time of 01:28.127. Over a 66 laps race, this would equate to a total race time of 01:36:56.382, i.e. 47 seconds faster than Red Bull in 2011. Before we get too excited about that, let’s not forget that the temperatures in testing are (traditionally) significantly lower. This has an effect in performance, both in terms of engine and aerodynamics (cooler air = denser air). Mercedes best lap time in Barcelona pre-season testing was 1:21.218, whereas their best quali lap was 1:22.569, i.e. a difference of +1.5%.

By adding 1.5%, finally, we get a total race simulation time of 01:38.23.628. How does that compare to Vettel’s 2011 race? It’s about 41 seconds slower. Admittedly, this is a very risky comparison because (a) we don’t know how hard Nico was pushing, (b) although we know Nico was using hard tyres for his stints, we don’t know their condition (Vettel was used 3 sets of softs and 2 sets of hards in 2011), (c) we don’t know if Merc were running the car at full blast (engine settings, KERS settings, etc) (d) Vettel did 5 stints whereas Nico did 4 stints to cover the same distance. Lots of disclaimers then.

We can, however, be confident enough to say that Mercedes, at this moment, are a little bit behind of where Red Bull were at the start of their 2011 season, with the exhaust blow diffusers. That’s a very good step for the team. Vettel has already admitted that the RB8 is sliding more in fast corners than the RB7. From this, we cannot surmise how much downforce / performance exactly they have lost from the previous season. If it’s around 1.5 seconds (which is approximately the time that they have found between Spain 2011 and Brazil 2011), then Mercedes should be reasonably close to them. Let us also not forget that from now until the first race, the cars will be heavily updated and worked on; we should not expect big gains in terms of seconds, but a 0.3-0.4 seconds gain from the start of Barcelona pre-season testing to Melbourne should be expected for the top teams. All in all, Mercedes seem to have made the necessary leap to bring them into the Red Bull “territory” and I believe that we are going to see differences of less than 0.5-0.7 seconds between Red Bull and Mercedes at the start of the year (last year, I remind you, was around 1.3 seconds). 0.6 seconds over 66 laps translates to 40 seconds difference at the end…

Now, it’s interesting to examine what some of the “smaller” teams were doing on Friday, Williams and Sauber more specifically. They both (Maldonado and Kobayashi respectively) embarked on what looked like race simulations. Kobayashi completed 58 laps (including in and out laps) and Maldonado completed 69 laps; you can see their runs in the diagram below:


Kobayashi’s runs are a little bit short of a race distance. As you can see in the table below, his average lap time was 01:29.802. This translates to a race distance (again, disregarding the pitstops) of 01:38:46.932. By adding the 1.5% percentage that we discussed above, then it translated to 01:40:15.836, which is about 1:52 minutes slower than Mercedes and 2:30 minutes slower than the 2011 Spain GP Red Bull. If that was indeed a race sim, Sauber would have been lapped quite comfortably.

Williams, on the other hand, look much more racy. Maldonado’s average lap time over 69 laps was 01:28.516, which translates to a race time of 01:37:22.056. Again, using the 1.5% addition, we get a final race time of 01:38:49.687. That’s barely 26 seconds slower than Mercedes’ race simulation (0.4 seconds per lap), which (possibly) shows that (a) Williams are in a much better shape than they were in 2011 and (b) the gap in the midfield is going to be very tight in 2012. People should not be surprised with Williams’ testing performance. In the Silverstone GP (2011) they were the team who gained the most from the temporary ban on engine mapping to blow the diffuser, therefore it was expected that they had the most to gain from this change of rules. In Silverstone qualifying, Williams were able to almost match McLaren’s pace

So, you naturally ask, is it safe to draw any conclusions at this moment? The answer is no. We can get a general “idea” of things. Mercedes have upped their game. By how much? Only time will tell. The midfield will be very tight this season… By how much? No one knows – we have yet to see what Lotus can do, for instance, and we have yet to evaluate Force India’s race simulations (none so far). We have also yet to see what Ferrari are capable of. Rumours have it that they have unlocked certain areas of the car – perhaps people were far too quick to rule them out. All in all, provided it’s not another Red Bull whitewash, we should be in for a thrilling 2012… I can’t wait for the next and final Barcelona tests….

  1. Jomy John says:

    Nice article

  2. madmax says:

    Another nice bit of analysis. Even though Brawn has said they don’t have a winning car there seems to be a lot of confidence coming from that team from both drivers so we could yet end up with Vettel fighting toe to toe with his hero Schumacher for the title!

    • abu says:

      Oh… that would be great, wouldn’t it!? I am really looking forward to the next tests, because that’s when we will get a much better picture of where everybody stands. Thanks for your comments mate 🙂

  3. Hi there – I was alerted to your analysis by a comment on my blog (intelligentf1). Nice to see someone else taking a measured approach to the testing data.

    I have constructed a race history chart with fictional in/out laps, and using a model of the 2011 race suggests to me where cars have refuelled – I reckon that only Red Bull, Rosberg (maybe Schumacher) and Maldonado did a ‘real’ race sim. It looks like McLaren, Sauber and Force India were ‘repeating’ stints as the fuel penalty is about right (or equivalently the lap times are the same).

    Comparing notes I think we have similar ideas, but I think that there is a bigger gap between Mercedes and Williams (closer to 0.8s). Fuel corrected (which I do with a model based on a simulation of last year’s Barcelona race), the Williams is consistent through all four stints (same tyres?) but the Mercedes is slower in stints 3 and 4 (harder tyres?). Also, if you fuel correct the stints which make up the Ferrari 41-lap ‘run’ (assuming empty at the end of the run) you get Ferrari in the mix as well. McLaren are a mystery – their runs don’t show too much by my reckoning.

    Do you have more information on which tyres were used in which stint? This would tie up a few loose ends for me. I think that a lot of the ‘hidden’ time is in engine modes as well as in fuel load – I’m tending to think that there is up to a second to come from the big boys which would put Rosberg about 0.5s down on Red Bull (as you suggest). Would you be interested in swapping notes?

    • abu says:

      Thanks for your comment – I just visited your blog and am planning to spend some time later today going through it. It looks really interesting! I am more than willing to share anything, although everything I have, I have posted already. Regarding tyres, I believe Rosberg was using medium for the first stint, and hard for the rest. It’s hard to find tyre info; even journalists on track can’t enlighten you most of the times (how do you tell hard from medium, from apart?). I think Williams was not consistent in all four stints, if you look at the data. It seems they were trying to do a very consistent run in the 1st stint, and then do a balls-out run on the 4th run, and see the difference. Basically, we are seeing this in many of the teams’ stints. They are obviously trying to assess what is the best way to tackle a stint. Maintain steady pace or give it the beans from the start? I’ve seen McLaren doing similar comparisons, Williams, Mercedes, Force India and Sauber. All in all, I think that the only safe way to correlate testing times with race is to factor in the lower temperatures. I’ve tried to do that, and we’ll have to see how it correlates in the end. It will be a useful lesson for 2013!

      By the way, you can e-mail me at abulafiaf1@gmail.com

    • Calin says:

      Regarding Williams:

      Did you take in consideration that Renault engine is using 18 kg of fuel less per race?

      • abu says:

        No, because it’s not necessary to take this into consideration. Assuming that Williams do indeed need 15-18 kg of fuel less, then that’s something that would replicate itself also during pre-season testing, so it’s not a variable that must be taken out of the equation.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s