The Mysteries of Testing

Posted on March 8, 2013


Firstly, thanks to those who have asked for my view of things. With life being pretty busy, I wasn’t sure I was going to be able to keep doing this – especially with the lack of nice analyst-friendly race simulation data from testing. But you’ve encouraged me to have a go. In comparison, last year was easy. Very easy. We knew that Red Bull and McLaren were quick, Ferrari were in trouble, Mercedes and Lotus were thereabouts but not quite there.

This year is different. The big issue is that there are just not enough examples of consecutive stints without refuelling, such that we can be confident of the stint starting fuel loads. Knowing when the cars are full is about the whole game – then the fuel levels can be inferred for other stints and we can build a picture. Trouble is, this year, it is not so easy to identify the ‘full tank’ stints as (inconveniently) not everyone has clearly done them. Added to this, the teams are playing with the tyres – what do you do at the start of a stint to keep them alive, or does pushing hard make little difference. Interestingly, I suspect the latter, as there are many examples of two fast laps followed by a very consistent stint which happens to be about 2s slower.

Although the qualifying pace will be about those opening laps, the race will be about the consistent pace, so that is what I have tried to measure. It is an intriguing thought, though, that the teams who can get up to five faster laps at the start of the stint (yes, stints like that exist too) could well get opportunities to gain track position early on. An aside to this is that the undercut will be particularly powerful this year – perhaps four-six seconds faster than the opposition on old tyres (2s for fuel load, 2s for normal tyre degradation and 2s for these ‘golden laps’). There are going to be some difficult strategy decisions – and probably unnecessarily early stops – as we had in 2011 when the Pirellis were new to F1. How much of this is due to the (relative) cold of Barcelona in winter is impossible to say.

So what can we do? Firstly, I have only looked at the front five teams (according to popular thought) so far. I have tried to categorise the runs roughly by stint, so similar (but not necessarily identical) fuel loads. I have done this using the assumption that things are pretty competitive and that everyone is within about 1s of each other. Fortunately I get a picture that makes some sense, but unfortunately there is no way to verify it. I have also looked at the tyre performance – do the cars hold on to the pace from the tyres, or do they lose pace rapidly. Some of this may well be from experimentation, but I have taken a general sweeping view to suggest that the team who see more issues are having more problems – either because they need to try stuff or because they can’t get the tyres to work.

Oh, and the pace is not as fast as we might have expected. Rosberg’s fastest lap is 2s faster than the pole from last year (even without free DRS use), but the pace (after golden laps) is about 1s down on last year for those runs which look like a first stint pace.

Anyway, the stints that I have show the following (relative to Maldonado first stint race pace from 2012):

Stint 1:

Red Bull: 3 stints; +1.5s, +1.0s, +1.0s; all good tyre behaviour

Ferrari: 1 stint; +1.0s; good tyre behaviour

McLaren: 3 stints; +1.5s, +1.4s, +1.6s; poor-OK tyre behaviour

Lotus: 2 stints; +1.6s, +1.0s; one good and one poor tyre behaviour

Mercedes: 1 stint; +1.9s (damp?); good tyre behaviour

Stint 2: (~1s faster)

Red Bull: no stints

Ferrari: 1 stint; +0.0s; good tyre behaviour

McLaren: 3 stints; +0.3s, -0.2s, -0.1s; very variable tyre behaviour, one very poor

Lotus: 2 stints; +0.2s, +0.3s; OK-good tyre behaviour

Mercedes: 2 stints; +0.5s (damp?), +0.2s; OK tyre behaviour

Stint 3: (~2.5s faster)

Red Bull: 2 stints; -0.7s, -2.2s; one good and one poor tyre behaviour

Ferrari: 2 stints; -0.9s (damp?), -1.5s; OK-good tyre behaviour

McLaren: 1 stint; -1.5s; good tyre behaviour

Lotus: -1.0s, -1.7s; one good and one poor tyre behaviour

Mercedes: 7 stints; -1.0s, -1.7s, -1.5s, -1.6s, -1.5s, -1.6s, -1.0s; one poor, 2 OK, the rest good tyre behaviour

Stint 4: (~4s faster)

Red Bull: 2 stints; -3.5s, -3.6s; OK-good tyre behaviour

Ferrari: 1 stint; -3.3s; good tyre behaviour

McLaren: 1 stint; -2.5s (damp?); OK tyre behaviour

Lotus: no stints

Mercedes: 4 stints; -3.2s, -3.5s, -3.3s, -3.0s; one OK, 3 good tyre behaviour

And what can we say? Nothing conclusively, but there are a few themes. McLaren have the least consistent tyre performance, and Ferrari (although the stint number and lengths are small) the best. For long stints, Lotus and Mercedes look to have decent tyre life, McLaren don’t and Red Bull/Ferrari haven’t really shown their hand. For the stints analysed here, McLaren and Lotus have concentrated on higher fuel levels and Mercedes on lower fuel levels. Can’t really say why.

Competitive order? Well, Red Bull appear to be at least as good as the others in each of my stint categories, so they would probably have to be put first. For the others, it’s more difficult. The heavier fuelled stints (one and two) suggest that Ferrari is reasonably competitive, and that McLaren have pace if they can keep the tyres together. Lotus and Mercedes are at most a couple of tenths away, which is only two laps of fuel! The stints with the lighter loads look better for Mercedes – for stint three we cannot really separate the teams, and for the lightest load Mercedes look to be at least as fast as Ferrari. The noise in fuel level and performance must be at least 0.5s, so we have to be very careful with any conclusions.

The best guess, however, is that it is close. I would guess that Red Bull have 0.2s (maybe closer to 0.5s if they are really sandbagging), and that the others are really close together. I would expect Ferrari, Lotus and Mercedes to race better than McLaren on the evidence from the tests just because they seem to keep the tyres going better. In fact, I’d probably put Ferrari second (with the very big caveat that they have done very few long runs), and not separate Lotus and Mercedes at all. Much will depend on how each car gets the tyres to behave in the first few laps – and how they qualify.

I’ll try to have a look at the next four teams in the next week or so. In the meantime, only the teams know exactly what they have run, how they expect their car to behave on the tyres and what sort of race pace they expect. Given that the teams don’t really know how the land lies, it’s not a surprise that I can’t be conclusive with the pecking order, but I’ve had a go… I am almost as intrigued by the new tyres causing qualifying to be much faster, but the race much slower. Can this really be true? I guess we only have a few weeks to wait to find out. Friday practice could tell us a lot. Maybe.