Blog: How reliability reduces F1’s records
F1’s all-time records, and those vis-à-vis Ayrton Senna, have been quite the hot topic lately. One fell in Singapore, though perhaps not the record some anticipated, in that a certain Sebastian Vettel nosed ahead of Senna’s win total, and did so in fewer starts. And then the race after in Japan Lewis Hamilton matched the Brazilian’s number of Grand Prix triumphs albeit a race too late to give us the extraordinary coincidence of the same number of wins in the same number of starts as his late hero. Yet in neither case was there that much whooping in response that their respective legacies had matched that of Senna. Instead they were accompanied by a conspicuous round of ‘Yeah, but…’. Statistical comparisons of different drivers in different eras is a mug’s game when judging pilots in the round, they said.
And in F1 they likely have a point. Any sport will evolve over time which with it means imperfect comparisons of different ages, but it’s hard to think of a sport that evolves as much as F1. At the other end of the scale a 100m sprint has remained almost entirely unchanged, as have most athletics events, meaning there is little dispute over their marks. Even football laws have changed only in the detail. F1? You could make the case that at various points of its history it has quickly become unrecognisable compared with what was before.
There are a few reasons for this: tracks, technology, regulations (technical and sporting), danger levels, challenge levels, numbers of races, lengths of careers. None have stayed still for long. And unlike with the afore-mentioned 100m it is not the unchanging stopwatch that F1 judges its records by –at least the ones we tend to talk about – but rather those relative to contemporary competitors. Wins, championships, pole positions… And add to this that F1 is a pursuit of driver and machine – and that machine changes things. The extent that the driver has a technical advantage; the extent that their rivals do. All are forever moving targets. It is trying to factor all this into our interpretation of where drivers sit in relation to each other where things get really messy.
But there is another big factor that ensures that the ‘you can’t compare records from different eras’ point holds true. It is hardly mentioned but should be. That of reliability. And this is somewhere that the sport has really changed between Senna’s day and now. Now the F1 car seems close to unburstable; back in the Brazilian’s pomp it could be reasonably expected that around half the field, perhaps more, wouldn’t make the end. And the implications of this are obvious, particularly for win totals. By definition the less likely you are to make the finish the less likely you are to win. As well as the less likely too that the nominal pace setter (of which Senna, Vettel and Hamilton often are) will be the one to win out.
And the statistics back this up. It can be seen from the chart that I lovingly created below, including compiling the stats myself (I hope you’re grateful), of the percentage of starters that were still running at the chequered flag in each F1 season, stretched back to Senna’s debut campaign in 1984. It is in some part a crude measure, such as that it won’t necessarily be the front-runners, let alone the three in question, breaking down. Also some of the non-finishes will be down to spins and accidents, but it seems reasonable to assume the number of those will not have changed much over time.
But still the decline of unreliability is clear. In 1984 when Senna debuted around 60% of cars were not running at the end of an F1 race, and it remained at roughly that level for the rest of the turbo era. Come the 1990s and Senna’s untimely passing it still was at around the 50% level, though drifted down a little subsequently. It wasn’t until the mid-2000s though that the levels of low unreliability that we know today kicked in. An ironic feature of the introduction of post-qualifying parc ferme rules in 2003 is that while mass unreliability was predicted it actually coincided with non-finishes nosediving. Perhaps mechanics just fiddled too much given the opportunity. Part of it too may have been a response to the astonishing Schumacher-Ferrari dominance of that time, part of which was down to near-bulletproof reliability, and with this others had to sharpen up their acts in response.
In the 2005 Italian Grand Prix not a single car dropped out (thus matching the 1961 Dutch Grand Prix in that regard – it applies technically to the 2005 US race too but that hardly counts…) something since matched in Valencia in 2011 as well as was oh-so nearly the case in the Japanese race just passed.
By the time Hamilton debuted at the start of 2007 and Vettel later that season reliability had been far reduced compared with Senna’s day. Taken in total, in Senna’s F1 career there was a 54% chance a car wouldn’t still be circulating at the end. In Hamilton’s career this probability was near enough half at 23%. In Vettel’s case it was 20%.
Further I drilled into the F1 careers of our trio to see how many wins we could add to their totals had technical unreliability not intervened. Obviously there is necessarily a little bit of interpretation here, given it’s hard to know what would have happened.
Spins/accidents or running out of fuel (which we can at least in part lay at the drivers’ door) are not included. Neither are gremlins at the start (so Australia last year is not included in Hamilton’s total and neither is Vettel’s mysterious puncture at the first corner in Abu Dhabi in 2011) as they don’t give enough to project forward from. Essentially it’s when something broke when winning, a bit into the race, and which cost the win in all probability, that I’ve included.
Hamilton has surprising few additional wins on this basis. I made it only three – Singapore and Abu Dhabi in 2012 plus I’ve generously included Canada in 2014 as he was actually ahead of Nico Rosberg (briefly) at the point that his brakes finally went. We could of course argue all day about that one.
Vettel has more though, six. Bahrain, Australia and Korea in 2010, Brazil in 2011 (though he was able to finish second, despite his gearbox problem), Valencia in 2012 and Britain in 2013.
Senna however can count an imperfect ten. Monaco, Germany and Australia in 1985, USA, Canada, Britain and Italy in 1989, San Marino and Australia in 1990, and Canada in 1992.
It can be argued credibly also that there is a flipside to unreliability, in that when unreliability is higher drivers also stand to inherit wins that they wouldn’t otherwise have got. This hasn’t been factored into the analysis I’ve done, frankly because it’d get really knotted, as well as that in any case for the reasons given drivers such as these three will not have been in a position to inherit too many. The pole master Senna certainly wasn’t.
But it’s worth reflecting that for Vettel at least even accounting for this difference of mechanical gremlins his win record would still be capable of near-direct comparison with Senna’s. He’d have three fewer victories than Senna in eight fewer starts. It’s something that overarches such debates – and that perhaps we sometimes forget – that these records being debated still all are great however we choose to interpret them, and in this spirit this article is not at all intended to discredit any of them. It’s just to provide an element of context.
Yet there’s another reason why F1 records only mean so much, related in part to the things we’ve already outlined. Any statistics must be appreciated in their context. And furthermore qualitative judgments should be made as much as quantitative ones. Maybe more.
To take one case – an outstanding one in every sense – had Stirling Moss won the 1961 Monaco Grand Prix in a vastly superior car rather than triumph in the outdated and underpowered Lotus that he did have, the record books would have treated it exactly the same. Even though in every other sense it scarcely could have been more different. Rather sums it up, I’d say.
Author: Graham Keilloh
Blog: talkingaboutf1.com Want to be a guest writer on VitalF1.com?
It couldn't be simpler, just submit an article
via this link OR drop us an email on email@example.com
and we'll publish your views.
It's easy to join up and join in
on VitalF1.com, simply click the link and enjoy getting involved!If you are already a member of Vital Football, your logins will automatically work on VitalF1.com!