Do VW officially state that it isn’t representative of real life situations anywhere? Haven’t seen anything to that effect myself unless you take the use of “up to”.
What I have seen quite prominently on a Mazda TV advert for the Mazda 6 is that they declare that their official combined cycle results may not be representative of real driving results and are to be used for comparison with other car manufacturers only. It would seem then that Mazda are quite honest and by association with their non-real world test figures, the same can be expected of the other car manufacturers also (for it to be a fair comparison). I do think it is completely wrong to just accept 75% of claimed combined figures as “doing ok” when they have sold the MK7 as being 20% better than the previous generation on fuel usage.
I'm not saying that this is the way it should be, just the way it is...
It is not all cars either - being cynical - cars that suffer this problem most are those in the competitive mid-range hatch market. "Honest John" shows that some cars - for example Jag, Porsche achieve the same or better than their official figures and diesel powered cars seem to be worse.
My girlfriend, a few weeks ago, took delivery of a new 120d M-Sport which should officially do 65.7mpg on the combined cycle, the trip computer, last I looked, was reporting 48.5... more than 25% off... Now, she is not the most economical driver
but I would have expected better...
The point here is that the test is not representative of the driving she does... the extra-urban test only briefly reaches a peak speed of 75mph and has an average speed of 40 iirc - she, on the other-hand, has a commute of about an hour, often in congestion but when it is clear she will settle into a cruise higher than the peak speed of the test...
With my likely driving in my GTD when it arrives and is run-in, I would expect to see high-forties to low-fifties as mpg... Anyone expecting to see close to the combined figure for the GTD (67.3) would need to drive as per the tests... and even then I reckon they would only be within 5% because there are so many other factors at play that the test doesn't really address...
The previous tests were not seen as representative (urban/56/75) because they did not represent normal driving but neither do the current tests and having read what is proposed I doubt the upcoming ones will either...