Posted on November 15, 2011
McLaren
The Strategy Report

This weekend’s Abu Dhabi Grand Prix was very interesting from the point of view of Race Strategy, with six different strategies in the top ten finishers.

Pre race predictions of two stops were the norm, but there was a wide variety of alternatives tried, with Mark Webber doing three stops en route to fourth place and Paul di Resta scoring a points for ninth using a one-stop plan.

We also saw McLaren pull off something very audacious at the first stop; they pitted Jenson Button on the same lap as the leader Lewis Hamilton, with only a 12 second window between them! Button’s total time in the pits was a second slower than Hamilton’s but it was a very brave thing to do and showed the team’s confidence on the day.

All three podium finishers did the predictable strategy, of running two stops with a longish middle stint on soft tyres of around 24 laps before a short final stint on the mediums. Ferrari tried to use the easier tyre use of the car with Alonso to stay out three laps longer than Hamilton and have a go at jumping him at the second stop, because they saw that the McLaren driver was losing time in traffic after his second stop.

It nearly worked. He was only 3 seconds behind when Hamilton pitted on lap 40 and did two very strong laps on worn tyres before pitting. He was about to catch the HRT of Ricciardo and didn’t want to lose time, so pitting then when he had pretty much the right gap (21 seconds) over Hamilton made sense.

But unfortunately for him, HRT called their stop at the same time and he was held up behind Ricciardo on the way into the pits and a slightly tardy change cost him the chance come out ahead. But even if he had managed it, Hamilton would have probably passed him as the McLaren was superior to the Ferrari on medium tyres. This scenario is precisely what happened in Germany earlier this year.

But Ferrari must have felt that they had enough margin at that point to call Alonso in, as he still had good pace from his tyres.


Why did Webber change to three stops?

Mark Webber has done some racy strategies this season, mostly to get himself out of traffic, which is something his team mate Sebastian Vettel hasn’t had to do much as he’s usually been out at the front.

Webber started out the race planning to make two stops like the podium finishers. He lost a position at the start to Alonso. However, with Vettel retiring he was racing Button for a podium finish until he had a very slow pit stop on lap 17, which cost him six seconds. It was an unusually messy stop for the Red Bull team which tops the league table for stops this year, along with Mercedes.

Although Button had KERS problems, the gap was still significant back to Webber after this. Webber was racing Massa for fourth place. The Brazilian wasn’t on great form in Abu Dhabi and it was well known that Ferrari were very wary of the medium tyre as they had struggled to get performance out of it in practice, as they have all year.

When Webber pitted again for soft tyres on lap 35, it was clear he had switched strategy. But it wasn’t realistic to think that it would help him get Button, as to do that would require him to have gained 31 seconds in 20 laps over the McLaren.

There have been suggestions from some in the team that he needed to stop to get off his second set of tyres which wasn’t working for him, but after passing and being repassed by Massa he was sitting behind the Ferrari and the switch allowed him to try something different to get ahead of the Ferrari. There was a big gap in the traffic for him to slot into and use the pace of a fresh set of tyres.

It worked, but even if he had stayed behind Massa and done a conventional two stop, he would have easily passed him anyway once they both went onto the medium tyres because the Ferrari was so slow on them. Massa made life easier for Webber with a spin and by being slow generally.

Massa has been lucky that the gap between the Ferrari and the Mercedes is as big as it is because it means that despite being six tenths of a second slower than Alonso he still finishes in fifth or sixth place.


How to win the midfield battle

Many fans have wondered why Force India, having qualified both cars strongly in the top ten put Paul Di Resta on a one stop strategy. Di Resta ended up at least where he would have been – ninth behind Adrian Sutil – but the strategy didn’t give him a chance to challenge his team mate. The medium was slower than it needed to be on race day to make a one-stop work.

If you analyse the decision making here, you can see why they did it. Force India’s plan was to put one driver on the one-stop plan and the other on a two-stop plan. Locked in a battle with Sauber and Toro Rosso over sixth place in the championship, which is worth many millions of dollars to them, they were more or less forced to cover off the possibility of their rivals putting one driver on a one-stop strategy and hitting the jackpot if there was a safety car, of the kind we saw last year in Abu Dhabi.

The data showed a 50% chance of a safety car and if one of their rivals had managed to get a free pit stop and been able to run most of the race on the faster tyre, they could have cut into Force India’s points lead. This possibility had to be covered off.

The midfield teams race in a different way from the top teams; the leaders are constrained by starting the race on their qualifying set of tyres and secondly they don’t take risks because they have the car performance not to need to.

There is a significant gap between the top three and Mercedes and then between Mercedes and the midfield and this gives them a margin for error.

So they would never gamble on a one stop in case the medium tyres don’t perform, but for the midfield it’s all about gaining track position and gambles often pay off as we’ve seen this year.

So it was a smart move by Force India’s Dom Harlow and his team. Having qualified strongly, Di Resta’s plan cut the midfield pack off from Sutil – by the time the German stopped on lap 15 he had a 15 second lead over the nearest midfield challenger.

And if there had been a safety car Di Resta would have been the first car on the road to benefit from it. This is a great example of the depth of Strategic thinking that goes into planning a race.

RACE HISTORY
The zero line is simply the race winners average lap time (total race time divided by the number of race laps). This is why his curve can go above the line if he’s lapping faster than his average, and below the line if he’s slower than his average or doing a pitstop.

An Explanation of the race strategies from Abu Dhabi Grand Prix
42 Responses

  1.   1. Posted By: vassy
        Date: November 15th, 2011 @ 5:09 pm 

    Nice one JA. Regarding FI strategies, how do they decide which driver will have to do a difficult 1 stop strategy. Is it based on qualifying? and what do you think about the penalties that affected Senna and Maldanado’s strategies? weren’t they hit hard by them?

    [Reply]


  2.   2. Posted By: James Muscat
        Date: November 15th, 2011 @ 5:17 pm 

    James,

    Lewis and Jenson both have solid lines on the chart (following the pattern it looks like Jenson’s line should be dotted).

    I also don’t understand the zero line – if Jenson’s plot never goes above this line but spends all its time below it then surely that’s not his average lap time?

    [Reply]

    Gareth Reply:

    The “average time” is that of the race winner (i.e. Hamilton on this occasion), not each individual driver.

    [Reply]

    James Muscat Reply:

    But Hamilton’s own trace never goes above that line, and therefore it can’t be his average (since most of the time it’s below it)… am I missing something?

    [Reply]

    JF Reply:

    The position on the Y axis is the DIFFERENCE between (1) the average lap time of the WINNER over the WHOLE race and (2)the current (at a certain point in the race) average lap time of a certain driver.
    That’s why the WINNER curve always has to reach the zero line at the end and why everyone else is below at the end, since anyone else’s average lap time is “more” than that of the winner.
    When the curve is going up, the driver is lapping faster than the race average (of the winner), when it’s going down, the driver is lapping slower than the average (of the winner)

    Matthew Cheshire Reply:

    Not quite. The “trace” is the cumulative difference from the average lap, therefore that line for Hamilton goes down hill as much as it goes up hill and always ends at the same level.
    If it wasn’t cumulative it wouldn’t depict race position- the chart would just be the same as the fastest lap chart, and as you say, half the laps would be above and half below the average line.

    PS, where did that fastest lap chart go JA? That might have earned poor Webber a few more driver of the day votes.

    JF Reply:

    Oups, please disregard my previous comment, it’s wrong.

    For each curve(car), the position on the vertical axis is the time gap between that car and a ghost car that would be constantly lapping at the winner’s average lap time (computed over the whole race).

    The winner and the ghost car have to cross the finish line at the same time,that’s why the winner’s curve always end up on the zero line.
    That also means that the winner can be behind this ghost car for the whole race, only catching up on the finish line.

    When the curve is going up, the driver is lapping faster than the race average (gaining time on the ghost car).

    Gareth Reply:

    Lewis won the race in 1:37:11.886, which divided by 55 laps gives an average lap time of 106.03 seconds. At the start of the race, he was lapping slower than this (e.g. 107.13 seconds on lap 3).

    He made a pit stop of 19.439 seconds on lap 16; by this time he was 30.55 seconds behind his average race pace. He made a second pit stop of 19.350 seconds on lap 40, after which he was 28.00 seconds behind his average race pace.

    But, as the fuel burned off, his regular lap times were improving, so by lap 9 Lewis was lapping in 105.99 seconds, faster than his race average. After his first pit stop, all but two of his laps are faster than his average, so he catches up.

    This is seen on the chart as the line sloping upwards towards the zero line, as opposed to somebody consistently slower (like Liuzzi) whose line always slopes downwards. But yes, it messes with your head to think about it too much! (Stats are from formula1.com and f1fanatic.co.uk.)


  3.   3. Posted By: Ananth
        Date: November 15th, 2011 @ 5:37 pm 

    Excellent as always James!!! How did Sutil manage to get ahead of Michael at the end of the first stint. I see that they both had their stops on Lap 16. Did Sutil overtake Michael on the track or did he beat them in the pitstop?

    [Reply]


  4.   4. Posted By: Rob Newman
        Date: November 15th, 2011 @ 5:39 pm 

    As always great analysis James.

    Knowing that the medium tyre doesn’t suit them, I woner why Ferrari didn’t want to try a three stop strategy. After all, there was nothing much to lose.

    I stil don’t get it why Di Resta had to start the race on the medium tyres. It doesn’t go well with a full tank and this is not the first time they did it. Safety car should be taken into consideration but some teams are thinking too much of it. Di Resta could have finished ahead of Sutil had he started with the soft tyres.

    [Reply]

    Kris Reply:

    Isn’t that what James addressed in the report? That it wasn’t necessarily about enabling Di Resta to get further up the field, but more about splitting strategies in order to cover off the possibility of teams behind them taking a gamble. It’s great insight and the timing of these reports is spot on… couple of days to let the dust settle and, at the same time give us readers an insight fix just as we enter a several-day period of little-to-no news or action.

    [Reply]

    James Beck Reply:

    Starting on hards is not a big issue – it doesn’t really matter if you start on hards or softs, the theoretical race time is pretty much the same. What he could have done was to stop on about lap 15 (once Buemi had gone – the window was there) and two stop from there. He would have been about with Sutil at the end by my analysis.

    With the evident pace advantage that they had in the early stages, I’m surprised they didn’t switch. I understand the safety car arguments, and I think he would have had it covered doing it this way.

    [Reply]

    Rob Newman Reply:

    @Beck, I analysed the first 27 laps of Sutil’s and Di Resta’s. I excluded Sutil’s lap 17 because that was the lap after his pit stop and Di Resta pitted on lap 27.

    In these 27 laps, Di Resta lost 47.46 seconds to Sutil with an average 01.83 seconds per lap. That equals to 2 pit stops. Also I noticed between lap 18 to 27, Di Resta was lapping over 2 seconds slower per lap.

    Obviously a full tank load slows down the car. As per my analysis, the cost of being on the medium tyre was 01.83 seconds per lap. Don’t you think a full tank and the medium tyres penalises a driver?

    [Reply]

    James Allen Reply:

    They all start with a full tank.

    James Beck Reply:

    Here’s how I do the numbers. I assume that the Force Indias use fuel at the same rate. I then match the pace of Sutil’s first and second stint on the same tyres to get the fuel load penalty. (Interestingly Force India look to have had a heavier fuel load than most of the cars in this race).

    By fitting a curve to the race history chart I can get the tyre degradation, and the underlying pace. Then I can compare fuel corrected the pace on the hard and the soft tyre. I get that both Sutil and Di Resta are equal in underlying pace on the hard tyre, and I get that Sutil is 0.3s quicker in underlying pace on the soft tyre. I think the difference giving you 1.8s is due to Sutil’s new softs after the stop – I reckon about 0.05 per lap degradation (very roughly) to fit the curves.

    Di Resta is about 1.6s slower on the hards than Sutil on the softs, so on lap 20 (old hards vs. new softs) the difference should be about 2.5s per lap. I think this checks out OK.

    So, on balance, no – I don’t think that there is a disadvantage to starting on the harder tyres. In fact, if you have track position and hold everyone else up, you could argue it’s an advantage.

    Rob Newman Reply:

    I know James but what I am trying to say is, the harder of the two compound always carries a time penalty at the beginnng of a race.

    Andrew Carter Reply:

    The prime tyre will carry a time penalty at the start of the race, obviously so, but if FI had run a two stop strategy on Di Resta’s car that would have been payed back on the final stint by running the softer options whilst everyone else is on primes. Given that by lap 15 Di Resta had given Sutil a sisable cushion over the rest of the midfield I’m surprised they didnt change him to a two stop strategy.

    Rob Newman Reply:

    Exactly. Just like @Beck and @Carter has mentioned, either they should have brought Di Resta in around lap 15 or at least soon after Sutil’s pit stop. The expected safety car didn’t come so that was the right time because soon after Sutil’s pit stop, Di Resta was over 2 seconds slower than Sutil. Between laps 18 and 27, Di Resta lost 31 seconds which is more than a pit stop.


  5.   5. Posted By: Alex_D
        Date: November 15th, 2011 @ 5:45 pm 

    James, I have a different understanding of the second pit stop for Alonso. Ferrari, mistakenly, called Alonso in 3 laps after Hamilton. They should have done it 2 laps because on the 3rd lap Hamilton was already faster than Alonso.

    Now we also know that HRT was on the way and if they would have called him in after 2 laps, he could have been ahead.

    Abu Dhabi is not Germany. Even though Hamilton was faster, it is not a given that on this track he would have overtaken Alonso. He would have been very cautious to do it due to the multiple collisions he had – he would not risk IMHO.

    [Reply]

    BB Reply:

    Ferrari didn’t have the gap required after 2 laps, and when did you last see Hamilton having a coming together with a quick racer like Alonso or Vettel?

    Favourite piece of text this week

    “Massa made life easier for Webber…….by being slow generally.”

    [Reply]


  6.   6. Posted By: gondokmg
        Date: November 15th, 2011 @ 5:47 pm 

    I beg to differ James, from what I heard on the BBC commentary, by the time Alonso pitted his sector times were slower than Lewis’ and Ferrari had to pit him because the strategy was not working. Besides, I thought the pit stop was about 22 seconds in which case Alonso did not have enough to jump Lewis through the pits contrary to what you said.

    [Reply]

    Martin Reply:

    I think you are right, but it would have been interesting for Ferrari to test the cliff on the Medium tyres as Button was well covered. Hamilton was benefiting from the new tyre effect that goes away – you can see a slight decrease in the rate of Hamilton’s lap time improvements just after Alonso stopped. It might have been possible with minimal wear for the softs to be faster than the mediums again, say in five laps time. I don’t know this to be the case, but we saw something like it with the two Mercedes cars in India – Michael was able to overcome Nico’s new tyre advantage by staying out long enough. Mercedes has come a long way with its tyre wear this year. Nico did a second stint of 31 laps, so if Ferrari did the same, it would on lap 47, so four laps would have been marginal on wear vs time to gain, but I would have been tempted.

    The pit lane time (the bit that is limited was 21 seconds for an excellent stop. The cars on track have to transit that distance too, but they are quicker on entry and exit. I get the impression Martin Brundle forget this at some tracks, but for Abu Dhabi, 22 might be about right for time lost. For stops later in the race the penalty will be slightly greater as fuel effect penalises fast corners more than slow ones.

    Cheers,

    Martin

    [Reply]


  7.   7. Posted By: irish con
        Date: November 15th, 2011 @ 7:15 pm 

    alonso and his 10 podiums this year has been fantastic. him and vettel are the best drivers in f1.

    [Reply]

    pargo Reply:

    100% agree. Alonso has taken his Ferrari to another level purely on ability. He and Vettel are pure class.

    [Reply]


  8.   8. Posted By: Jonathan Lodge
        Date: November 15th, 2011 @ 8:04 pm 

    what I don’t understand is the timing of Button’s second stop. With most of his second run missing the extra power from KERS I was expecting him to be able to make his tyres last several more laps. I realise that the later laps must have been terrifying (not knowing what the brake balance would be at each corner) but 20 laps with the balance adjusted for no KERS Button should have been able to go further and have a much shorter hard tyre run – which meant he could have been racing Alonso.

    [Reply]

    James Beck Reply:

    I did the maths on this. I reckon he could have gained about five seconds extrapolating his second stint pace, and he finished 17s behind Alonso. Pretty confident that it wasn’t on.

    [Reply]


  9.   9. Posted By: Anup Kadam
        Date: November 15th, 2011 @ 8:36 pm 

    Thanks for the analysis James…m a big fan of you…thanks once again…

    [Reply]


  10.   10. Posted By: Andrew
        Date: November 15th, 2011 @ 9:53 pm 

    Hi James thanks once again for an interesting and insightful read as always. Was wondering what is your view on Massa’s performance and position within Ferrari? He seems even further off Alonso’s pace and one gets the feeling that Ferrari have simply accepted this. With both Button and now Rosberg unavailable what do you think Ferrari will do with Massa next year if he does not raise his race pace?

    Also can you comment on the Schumacher/Rosberg battle? Last year Jock Clear was working with Rosberg and now he is sitting in Schumacher’s corner. Is this an example of a champion getting his elbow’s out and making sure the right people are working with him?

    [Reply]

    James Allen Reply:

    No, to the latter point. Rosberg wanted Tony Ross his ex engineer from Williams. Schuey wasn’t getting on with Mark Slade so Jock came in to support. He’s a heck of a racer!

    [Reply]

    Dave Aston Reply:

    Jock Clear worked with Villeneuve in 1997, right? I wonder if they have a laugh about Jerez.

    [Reply]


  11.   11. Posted By: Dominic J
        Date: November 15th, 2011 @ 10:46 pm 

    Why did Barrichello have such a short middle stint? Looking at the graph he was impressively quick at that stage (as Di Resta’s pace didn’t pick up even when he was no longer behind RB), but instead Rubens was left with a long, relatively slow final stint on softs (having already used the medium).

    It seems to me that he should have stayed out maybe 5 laps longer and would have stood a chance of beating the Saubers. What a result that would have been for Williams, from last on the grid!

    [Reply]


  12.   12. Posted By: James Beck
        Date: November 15th, 2011 @ 11:02 pm 

    Where did the 50% chance of a safety car come from? I do hope that it wasn’t based on one safety car in two races at Abu Dhabi, and that it was based on a larger sample of about 20-30 races at modern Tilkedromes with large run offs and good recovery facilities.

    The problem with using a sample of two is that you can equally validly say that there is a 0% chance of a safety car outside the first 10 laps, which is obviously not true…

    [Reply]

    AA Reply:

    You can’t calculate the probability of a safety of the Abu Dhabi circuit based on the Korean, Indian, Turkish tracks etc. It’s like comparing apples with oranges. Of course the 50% came from the last two races.

    Perhaps in 50 years time, we will have a more satisfactory sample space.

    [Reply]

    James Beck Reply:

    I disagree. The question is how good do you think the estimate is from two races against the evidence from a larger number of races at other tracks with similar characteristics?

    When Singapore came on the calendar, we all thought that it was a strong candidate for a safety car. Was that based on no evidence, or because of comparisons with tracks with similar characteristics, like walls? Similarly, do you think Valencia is a low safety car probability track as we have had only 1 in 4 races? I don’t.

    I think that there is strong evidence to suggest that the safety car probability at Abu Dhabi was a lot lower than 50%.

    [Reply]


  13.   13. Posted By: Janis1207
        Date: November 16th, 2011 @ 9:21 am 

    Hi James!
    Nice analysis, that, and well written.
    Have you got any insights about Schumi stopping right after the finish line with fuel running out? Rosberg had enough, and Schumi’s pace wasn’t all that good to explain higher fuel consumption. Some kind of technical problems?

    [Reply]

    Mitchel Reply:

    Schumachar said in a BBC interview after the race that he had a puncture.

    [Reply]

    StallionGP F1 Reply:

    wasn’t fuel related he had a slow puncture which was covered in his post-race interview

    [Reply]


  14.   14. Posted By: Roberto Gomez
        Date: November 16th, 2011 @ 10:34 am 

    Dear James,

    After a small thought, I think now I understand the graph, 0 means a ficticious car lapping at constant speed winner’s average time. The graph shows absolute time different between each car and that ficticious one at each lap. Since cars are slower at the beginning of the race they acumulate delays respect the “0 car” at the begginning and recover them later.

    So unless strange situations happen they seldom show up above the “0 car” line.

    I think the “Zero line” explanation is more comfusing than clarifying, anyway, thanks a lot for your great reports.

    Regards,
    Roberto.

    [Reply]

    James Allen Reply:

    Exactly

    [Reply]


  15.   15. Posted By: James Beck
        Date: November 16th, 2011 @ 4:25 pm 

    I’ve been a bit confused as to why Sutil and Schumacher are so much faster in the second stint relative to the other cars. The second stints must be on brand new softs, but the difference to the used softs is very big in comparison with (say) Kovalainen. Normally this isn’t the case. Their difference is about twice as big.

    Did they each do two qualifying runs on their start tyres? Then it makes sense. I think.

    [Reply]


  16.   16. Posted By: Magnus Malmnäs
        Date: November 16th, 2011 @ 8:38 pm 

    We are all curious to why Red Bull are so fast instantly on the first lap in race and quali. When I today read about the possibility that Vettels puncture was caused by the hot exhausts it got me speculating. Could it i be that they blow hot exhausts onto their tires so that they get the tires fast upto ractemperature. Could it be too complicated to contol?

    Swedish regards Magnus

    [Reply]


  17.   17. Posted By: Ashish Sharma
        Date: November 17th, 2011 @ 11:48 am 

    Great to see a midfield teams strategy analysis. It is very rare to see the different strategies and different requirements of the midfield teams and how they optimise their points scoring covered in any depth.
    Kudos to James for covering this in detail across the season…

    [Reply]

Leave a Reply





COUNTDOWN TO NEXT RACE
Strategy Report
Innovation and Technology brought to you by TATA Communications
Senna DVD
Download the Chequered Flag Podcast here
MTS
Darren Heath
Sport Right Now