.... the problem with the stats is that they are put up by owners, some of whom like to let others know that their 4.6L V8 does 45 MPG on the motorway.

Many owners don't even do several brim to brim calculations using the correct formula, but rely on what their car computer tells them (which they probably re-set whilst going downhill at 60MPH in 6th gear) and then post these as "true" figures.

Dudes, as your posts reveal you to be an arch-sceptic, I would just ask whether you have posted your own verified figures? I have posted mine for a 207 SW 1.6 diesel giving 61 mpg. That is about what the car computer tells me, but my figure comes from driving 30K miles using 2205 litres - you can do the sum if you wish. I haven't calibrated my odometer against the M-way markers for some time, but when I did it seemed pretty close. I've not managed to check whether the pumps have delivered true litres; I have to take that on trust and rely on Trading Standards.

I agree that there must be some dubious data up there, but not all of it is. Brim-to-brim is OK, but pretty rough really.

Edited by Andrew-T on 08/07/2014 at 21:38

As I'm sure we all know, it's very common for trip computers to flatter the fuel economy. If people are submitting those, then they are worthless.

The new Roomster seems unusually accurate - just as well, as it is only claiming 41mpg, not the 49.6 official combined cycle.

Only 800 miles though, might loosen up. I did a few miles in S mode last night to increase the rpm a bit - it barely gets past 2000rpm in D.

Outlander combined figure is 43. The trip meter gets near that, but unfortunately exaggerates by at least 10%.

Interesting that the fuel computer in the lorry i drive is accurate, our depot fuel pump automatically calculates the consumption every time you fill up, and whenever i've zeroed the OBC (usually on one of our regular assessment days) its invariably within 0.2mpg accuracy.

What is interesting is how much difference terrain makes, on one hard run i do the vehicle can be down to around 5.5 mpg, but on another easy motorway run around 9.4 with a mean best of just over 10, these figures might look awful but we run at just under 44t and at discharge the engine has to be run for about 1 hour and up to 2 hours @ 800rpm to power the hydraulics and blowing eqpt...thats probably where the 0.2mpg difference sneaks in, the lorry won't have calculated PTO fuel used but the depot pump will have.

I'm not convinced about car odometers being any more accurate than the average speedo, who's checking them and how, you'd really have to cover a set distance (not a couple of mile markers) that had been checked by a calibrated odo.

My car consumptions are horrendous compared with some of those above, 96 MB E320 around 21mpg general running about, 02 Outback H6 the same, they are both on LPG though, so maybe petrol would give about 2 or maybe 3mpg more, i don't drive my cars for economy nor does SWMBO, thats why they are on LPG so they can be enjoyed.

The 207 with the 1.6 HDi figures above are good, on a par with the C2 VTS 1.6 HDi we had for a while, kept below 70 it would return 60 ish, but anything over would see a massive drop, that car could easily return 45 or less.

Edited by gordonbennet on 09/07/2014 at 23:28

Have checked the odometers in both our current cars against both our Garmin and AA Autoroute over a route we cover quite frequently, about 200 mile round trip.

The Garmin and AA Autoroute show exactly the same distance with a 1/10 of a mile. The odometer in the Kia Ceed records a distance 3.5% lower and the odometer in the Seat Leon records a distance 1.1% higher.

Without the exact figures in fornt of me I think the distances are 195.3 for the Kia, 202 for Garmin and AA and 204.2 for the Leon.

That can make a significant difference to the calculated mpg figures.

But I suspect that many people who post up the real MPG figures simply quote those shown on the dash computer. In the case of our 2 cars the readout in the Ceed is pretty accurate but that in the Leon reads over 10% more than the calculated figures. Anyone with a Leon posting the computer figures are making a mockery of the site.

But even those people are making less of a difference to those who insist on posting up they did "over 40 mpg on a slow motorway run in a petrol Jaguar".

There should be stricter rules and a way of ensuring that the "Real MPG" site has really worthwhile information within it.

Reading skidpan's post about odo accuracy it just occurred to me I could easily do the same using my hiking GPS, and add a correction factor to my mileage spreadsheets.

Then I remembered tyre wear.

A quick calculation shows that for a 205/45R16 tyre on the Roomster, a 4mm reduction in tread depth will reduce the circumference by c. 1.35%, resulting in an increase in the reported distance travelled of c. 1.37%

So if I want it to be accurate, I'll have to adjust the factor from time to time. A more practical way might be to adjust the factor to what it would be for a half-worn tyre, and just use that.

Then madf can point out that unless I have measured the mpg over the life of a complete set of tyres, I am fooling myself.