The figures are produced independently using a test rig which reproduces the same test “route” each time. It’s designed to give the best figures and therefore an idea of what’s achievable.
Of course manufacturers cottoned on to this and began to tune torque outputs at certain revs and use gear ratios more suited to the test. This had the unfortunate effect of maximising the test figures at the expense of what you could actually do in the real world.
So why do I think they are still relevant?
Simply because now everybody does it, the playing field is now level once again.
Normally there are three figures given:
- Extra Urban
Urban – This used to be simply what the car would do at a constant 30 mph, but nowadays the test “routes” involve differing loads and varying speeds. This figure is supposed to represent the kind of mileage you’d get just driving round town Honestly, I’ve never found a “style” that reflected this figure anywhere near accurately
Extra Urban – This is supposed to represent flowing driving across country on main roads. Actually, until yesterday, I’ve never achieved anywhere near this figure in any car previously. I only managed it on my way back from Norfolk after a holiday where driving home consisted of driving across large, very flat expanses of the country at a constant 45-50 mph, with little variation in speed, for quite boring amounts of time. This says to me that this figure may actually be genuinely achievable in your own car, but very rarely and only in exceptional circumstances, such as driving down a motorway at 50 mph – and who’s going to do that?
Combined – This seems to be the most useful figure I’ve come across. Whenever I purchase a car, I generally drive it as conservatively as possible for a couple of weeks. This isn’t just about fuel economy, it’s also about learning the car’s weaknesses. How does it run? Does it feel like the chassis is OK? How do the brakes feel? It’s good to get to know a car first before you start to push it. In fact, this is why I started tracking my mpg in the first place, back when fuel was reasonably priced. A drop in mpg is one of the first indicators of serious engine wear, such as piston rings on the way out.
Funnily enough, the problem of worn piston rings seems to be a diminishing one in that it barely occurs anymore, mainly due to advanced cylinder lining coatings that are now common place on ordinary cars, and the technology that also goes into your oil these days.
Anyway, back to the main point. Take the combined figure, say 53 mpg for example, and reduce it by 10%.
Working it out is easy enough – 10% of 53 is 5.3, so 53 – 5.3 gives 47.7 realistic mpg.
I’ve found that doing this with the combined figure gives a pretty accurate rule of thumb for how much mileage you are likely to get from any given car. Well, in my experience so far, this has been spot on for my last 5 cars.
Interestingly, while reading this, I scanned autotrader.co.uk for various mpg figures for random cars. OK, not that random, but it seems comparing hybrids like the Toyota Prius and turbodiesels by manufacturers like VW, BMW, Renault and Peugeot, there doesn’t seem to be a fat lot of difference. Perhaps the turbodiesel manufacturers should be the ones turning their cars into hybrids, imagine the mileage you’d get then…