Can testing tell us anything about next season?


Super Hero And All Round Good Guy
Every year at the end of the season we the fans have the pain of a 4 month F1 race vacuum. Which is why the car launches and testing get such intense scrutiny from us the F1 depraved deprived fans.

For me testing has always been a mixed bag, its great seeing the pictures of the new cars on track but the times seem so random its impossible to work out who is quick and who isn't until the first qualifying session of the year.

With this in mind I have under taken a little project in anticipation of the 2013 testing season. I have entered every testing result for the last 3 years and compared them to the first qualifying session of that season. My goal was to come up with a ‘sand bagging factor’ (SBF), that is, a number to adjust the average testing results in order to predict the most likely qualifying result at Melbourne for 2013.

Of course to call it sand bagging is not entirely fair. Many teams are simply concentrating on race pace with others concentrating on qualifying pace. But the result is the same, some teams appear artificially slow and others artificially fast during testing. Sand bagging factor or SBF seems as good a way to describe this as any.

Why 3 years of results? Well I reason that to go back further would introduce errors due to changes in team ownership, drivers, management or simply just philosophy. Yet to base it on less than 3 years would mean factors like crashes, bad weather etc would be too prominent in the calculation. 3 year seems about right to get a reasonable average whilst maximising the potential for accurate prediction in 2013.

I entered into this expecting to see no patterns and to have wasted my time, but I was pleasantly surprised to see that practically every team for the last 3 years has either under performed or over performed during winter testing in a consistent way from year to year when compared to their qualifying result in the first race of the season.

So let’s take a look at the figures.



N.B. a positive SBF means the team would do better in qualifying than in testing. Negative SBF means the team would do worse in qualifying than in testing.

On average the most consistent and extreme of all the teams when it comes to sand bagging during testing. Year on year Redbull have ran their cars on average with enough fuel on board during winter testing to give a false result in their pace by an average 5.7 places lower than their qualifying result at the first race. So if they appear to be running in 5th and 6th in testing chances are they will be on pole for the first race.

With the exception of 2010, Mclaren have actually sand bagged during testing to an even greater extent than Redbull, in 2012 running 8 places lower on average in testing than they pulled out in qually. When you average the last 3 years that works out as a SBF of 5.2 places lower than their qualifying result at the first race.

The least consistent of the top teams with a huge swing in SBF for 2010 (SBF 4.5) to 2012 (SBF -5.33) if we discard the oldest result in 2010 then we see that Ferrari tend to over perform in testing and are disappointing in qually that has certainly been the case in the last 2 years if not 3. Ferrari are the wild card in the pack the rest of the teams are much more consistent. Average SBF -0.83

The last of teams we will look at that run artificially slow during testing. In 2011 they were fairly honest in pace but for 2010 and 2012 they ran around 4 places lower in testing than they managed in qually. Average SBF during testing 2.67 places lower than their qualifying result at the first race.

The team that really goes out to get their car low fuel ready, or if you’re being more sceptical, the team that goes out to grab headlines during testing. Sauber seem to run their cars on very low fuel and in qually trim more often than any other team. With the assistance of the other teams running race simulation Sauber have on average appeared to be 5.5 places faster during testing than they have managed during qualifying.

Toro Rosso
The remaining team worthy of individual mention due to there apparent change in philosophy during testing. In 2010 they were running a massive 7 places faster than they eventually managed in qually. Possibly glory hunting like Sauber. For 2011 and 2012 this seems to of been abandoned with Toro Rosso actually under performing in testing in much the same way as Redbull. Read into that what you will. But I can’t help but think the number 2 team was reminded it was number 2. Average SBF 1.39 places higher in testing than in qually.

Lotus / Force India / Williams / Caterham / Marussia
These remaining teams appear to run fairly honest. On average just 1 or 2 places higher in testing than they manage in qualifying. This demotion in qualifying of 1 or 2 places can be accounted for by the sand bagging top teams moving forward to their correct positions.

So in summary it appears that the top teams only concentrate on race pace and never, not once carry out a low fuel qually simulation. I find that very surprising. At the same time Sauber in particular seem to be concentrating a large amount of their time setting up their car to run on low fuel, possibly glory hunting to help win sponsors pre season.

I feel this has been a worth whilst look into past form for the teams and await 2013 testing with a new enthusiasm.

I will be updating this post with my qually predictions after 2013 testing is completed. It will be interesting to see how accurate this ‘Sand Bagging Factor’ theory turns out to be.
I looked at this for testing before this (2012) season, making a system based on awarding points for fastest laps and most laps completed, the thought being that a team consistently running with low fuel would complete fewer laps. I then separated the teams off in to likely tiers and made the following conclusions, which turned out to be pretty decent in the end!
  • McLaren will be the team to beat, but...
  • Red Bull will have the strongest single lap pace.
  • Ferrari are not as far behind as people think and will be in the mix
  • Lotus are the team most likely to challenge McLaren/Red Bull/Ferrari
  • Sauber will lead the midfield.
  • Caterham, HRT and Marussia won't challenge for points.
The full article is here if anyone wants to have a look:
I was thinking about doing it again this year but there didn't seem to be too much interest at the time.

Have you taken the average of the positions in all the tests (based on fastest times) and compared that with the qualifying result from the first race of the season? The results are certainly interesting. The other factor that could be coming in to play, apart from sandbagging, is development between the last pre-season test and the first race of the season. McLaren famously did that in 2011, for example, successfully abandoning their 'octopus' exhaust in favour of a simpler (Red Bull like) system.
But weren't McLaren dog slow in testing and then cobbled together some makeshift diffuser and exhaust system to take pole and 2nd in Australia? So how does that correlate with their testing times....:thinking: or am I thinking about the year before?
Have you taken the average of the positions in all the tests (based on fastest times) and compared that with the qualifying result from the first race of the season?
Yes exactly. The other point you make about developement after testing has finished is certainly a factor but generally the difference in the cars between testing and first race is limited. Also any changes should be averaged out due to compairing the previous 3 years.
My sand bagging factor adjusted testing results is below giving the predictions for Qually in Melbourne.
This is based on the combined finishing position during the last 4 days testing at Jerez.

Note I have not included Rossiter as it would mean Force India would have 3 cars on the grid at Melbourne and as I have no idea if it will be Rossiter or Bianchi (or neither ) racing this year I have deleted the slowest (he was 14th which is why its missing).
Oh yes and I will replace Pedro with Alonso for the other tests if he can be bothered to turn up at some point.

gethinceri if you read the first post it is explained. Basically I took the 2010-12 testing results average them and compaired them with the first qually result for that year. The difference in qually and testing position averaged over 3 years gives the SBF (sand bagging factor). Simples.
gethinceri I intentionally only posted the SBF results as its a large complex spread sheet that would be a turn off to most.
But a small extract example to help explain below.

Vettel average testing 10.5th (14th jerez & 7th Barcelona)
qually position 1
difference (SBF) 9.5

Webber average testing 10th (18th jerez & 2nd Barcelona)
qually position 6
difference (SBF) 4

Redbull SBF in 2010 ((9.5+4)/2) = 6.75

The SBF for the teams I have used is the average over 2010, 2011 and 2012.

Hopefully you get it now. :please:
I think Force India and Bianchi may have screwed your results over by having Bianchi out there trying to show he should get the race seat. Other than that it all looks very possible. Will be interesting to see how far off it is come first quali session. Remember to gloat big time if your nearly right!
Red Bull have gone relatively unnoticed but they've had almost no issues and have always been there or thereabouts in terms of the time. Combine this with Gary Anderson's comments that they look like they have 10% more downforce than anyone else and it perhaps look like they'll be difficult to beat this year. Force India are the other stand outs for me, people have been slating them for their lack of driver selection but they've had a very good week.

For those of you that are interested in this, I've also doing a 2013 version of my testing championship, there's quite a few similarities but also some differences:
I think that SBF could be a myth, but I could very well be wrong....
I'm convinced it's a myth. Just a few reasons. All the teams need as much information as possible to establish where their attention needs to be concentrated to enable a decent development path. At some stage over the 4 days they all need to push to understand how their car behaves on track when it's nearer its limits. At a purely physiological level there are always mental battles going on between drivers. They can't resist. Hamilton for example would dearly love to win that one over the McLarens, especially Perez. Lastly no one can convince me any one of the top teams wouldn't like to walk away from 4 days of testing with the fastest recorded lap against their name. Based on my first reason, needing to see how a car behaves when it's being pushed, I think the times mean far more than any of the drivers and teams will admit.
Of course the times mean something. The unfortunate thing is that it is only the teams who know what there own times mean.

I would also point out that the key to testing is consistency, ensuring that the line is consistent, it is not a race.

A while ago, I happened to be at Silverstone with a friend, whilst the Jordan team were doing some testing (Early 1998, iirc) The chap that we were with pointed out the key difference between Damon Hill and Ralf Schumacher. Every lap, Damon was hitting every apex cleanly, and taking the same line into each corner, not spectacular, not on the ragged edge of pace, but consistent. Ralf came out, and was pushing it, getting ragged.

The chap with us, who knew, was very complimentary of Damon, stating that it is key from an engineering and data gathering perspective that the clean consistency made for much better understanding of the car.

The times provide us with something to talk about, but I would hazard that the teams are far more concerned about completing their testing programs, and getting some good data, and miles on the clock than the headline times. Regardless of the times, the teams will already have a view of where they stand relating to others, due to sophisticated spying techniques.
Your absolutely right Greenlantern101. We don't know what fuel they are running with, or what Eric Boullier's definition of a low fuel run is. This applies to all the teams. What we do know is Grosjean's times were set on short runs on the soft tyres. We can only draw our own conclusions from that. We do know that there was an estimated difference of 6 tenths between the hard and soft tyres at Jerez according to Pirelli, which make Massa's time, which was a full second quicker than Button's best set on the first day impressive. Yes, the track would have rubbered in by the time Massa set his time, which would have helped make up the remaining 4 tenths, but once again we don't know what fuel Ferrari were running or how hard Massa was pushing. I'm sure he would have been pushing but whether he got the best out of his laps we don't know. We do know he did a second lap in the 1.17's and numerous laps in the 1.18's on a longer run. Barcelona will tell us much more but at this stage I still believe the times are relevant and Ferrari and McLaren look strong. I also still believe sandbagging slows down a teams learning curve.
The Pits......Your point is spot on which is why Alonso has such a good reputation as a development driver. He's also capable of robotic times during a race when he needs to pace himself. Last season during one of the GP's, I don't remember which one, I was following the lap times on the F1 site and he strung a series of 8 laps together that were not only within a hundredth of a second of each other, all three sectors were within a hundredth, lap after lap after lap. It was astonishing to watch. The commentators picked up on it also.
Top Bottom