l’ve often wondered how accurate the Driving Data stats for fuel consumption were on my Karoq. This week I have been able to calibrate with brim to brim refuellings. As the image below shows 41 mpg on a 265 mile round trip to Bristol today. Refuelled when I got back 30 litres for 265 miles is 40.1 mpg, so I reckon that’s close enough to rely on. My previous car it was always about 4-5 mpg optimistic, although I believe VCDS could be used to adjust it to be more accurate.

Always found mine accurate to 1-2mpg, but usually high 40s to low 50s.

Edited 23 December, 202523 Dec by Routemaster1461

2 hours ago, Routemaster1461 said:

Always found mine accurate to 1-2mpg, but usually high 40s to low 50s.

This was a round trip MK to Bristol - MK. The 2.0 TSI is a bit thirstier than the 1.5 TSI - I usually manage to get 45 - 48 on this journey but today 44.5 mpg was the best I could manage on the trip, trip to Tescos last night dragged the overall down to 41.

I track my mpg on Road Trip app and generally find its calculations to be similar to the mpg in the dash display. Well near enough for government work anyway. 👍

And, just for kicks, have tracked every penny spent on the car since we got her. That makes scary numbers. 😱

@DSL i have books with refuellings and odometer readings going back to 1977, before that I had a little plastic “thingy” about 3” by 1” with two disks on the top edge and 3 dials on the bottom. The fuel volume was recorded on the top dials and the last 3 digits of the odometer on the lower dials. I have maintained spreadsheets recording maintenance and repair costs since around 1999. As you say the totals can be scary. I will look later to check.

I remember every vehicle i have owned or driven since 1977 and know how much fuel it used on the trips and journeys i do day in day and day out over the different seasons. I pay for the fuel, put it in the tank and drive the miles. I remember well the ones that were high 20 mpgs, the ones in the 30,s and 40,s or 50,s mpg. Petrol or diesels. Same roads unless maybe a new stretch of dual carriageway / by-pass, many of the same filling stations that maybe had different grades of petrol then Unleaded and different dervs. Unleaded, super unleaded, then E10 or E5. I used to write it down back in the day, then it becomes so easy to know which routes are more economic that there is no need. #Memories.

4 hours ago, thamestrader said:

@DSL i have books with refuellings and odometer readings going back to 1977, before that I had a little plastic “thingy” about 3” by 1” with two disks on the top edge and 3 dials on the bottom. The fuel volume was recorded on the top dials and the last 3 digits of the odometer on the lower dials. I have maintained spreadsheets recording maintenance and repair costs since around 1999. As you say the totals can be scary. I will look later to check.

Admittedly this is rather off topic. The boring stats are as follows:

Mitsubishi Space Runner 172K miles (from new) 1995 to 2004 - cost 5p per mile maintenance

Mazda Premacy 160K (ex demo 6 months old) 2004 to 2013 7 p per mile maintenance

VW Touran 108K from new 2013 -2024 9 p per mile maintenance

Skoda Karoq from 1100 miles (used 6 months old) 2024 - current 2p per mile maintenance.

Fiat Punto 30K miles from 9,000 miles (Used 1 yr ) 1998 - 2016 11p per mile maintenance

Skoda Fabia 20K miles from 900 miles (used 1 yr old) 2016 - current 23p per mile maintenance

Interestingly it is the low mileage cars that have the higher cost per mile, the 3 high mileage vehicles cost per mile when allowing for inflation is probably very similar.

The last 12 months I have got 39.9MPG on a Karoq SE L 1.5 manual, according to the Škoda app. However, most of my driving is short journeys round urban areas. When I do the occasional long runs to visit family which are 80-90% motorway, then I am getting over 50mpg. So I am happy overall. I could drive more economically, I suppose but I am too old to change driving habits.

Also, I have the Costco with the cheapest fuel prices near me. (that is cheapest for all the Costco stores) so I am paying about 10p a litre less than the loca Asda/Morrison's etc.

So I can't really complain.

This is a common query on multiple forums.

No human is ever going to 'outso' the calculation obtained from the vehicle management system that is controlling fuelling.

Modern engines engines require such high precision accuracy at the injector, that the stated volume of fluid delivered is the volume delivered, not what is 'brim to brim'.

Besides human inaccuracies, the absolute impossibility of dry tank to dry tank, and allowed variation in pump delivery per cycle, the accuracy of that method isn't within a million miles of the fuelling system.

However, it's quite a testament that so many got so close! 🤣

On 31/12/2025 at 00:16, Frizzbee30 said:

No human is ever going to 'outso' the calculation obtained from the vehicle management system that is controlling fuelling.

Modern engines engines require such high precision accuracy at the injector, that the stated volume of fluid delivered is the volume delivered, not what is 'brim to brim'.

The fuel system is no doubt precisely accurate as you say. But the odometer is not. Cars still rely on wheel sensors to compute distance travelled, just as they do to show speed. And the rolling circumference of a worn tyre with tread depth at the UK legal limit can be around 2% less than the rolling circumference of a new tyre. The car has no way of compensating for that error when calculating mpg. So it's accurate enough, indeed more than enough, but not as precisely accurate as you suggest. At 45 mpg, a 2% error is nearly 1 mpg. (And the same odometer error will also affect the human doing brim-to-brim sums.)

Edited 1 January1 Jan by mumpsim
typo

@mumpsim

That’s a set of very interesting figure your showing there.

Satellite speed apps are very accurate (But ONLY when your travelling a straight flat road).

They can be helpful is checking what your Speedo shows. However as you have indicated 5k miles later that data is incorrect.

Motor Manufacturers are obliviously aware of these issues. As a result they always, always over compensate your speed. Regardless of you having a new or old set of rubber fitted. Your Speedo will not show less and will show a couple of mph more that the actual speed your vehicle is travelling at.

The main reason. That any claim you might make against them re speeding fines. They can disprove in the blink of an eye.

Secondary reason, it makes it seem like the car is more capable than it really is (goes faster, accelerating quicker)l

A win win for them

41 minutes ago, Trickiejohn said:

The main reason. That any claim you might make against them re speeding fines. They can disprove in the blink of an eye.

Absolutely. The Sat Nav and Car companies don't want to be involved in (lots of) court cases of speeding fines. I am 99.99% certain that somewhere in the terms and conditions it says the accuracy is "approximate" and can't be used as evidence in court.
I had this discussion with a Sat Nav company years ago.

I spreadsheet the MPG in our Karoq and since we got the car it's done just under 10K miles and its cumulative average MPG is 38. The cluster display in the car says 42.

In the UK it is actually against Construction and Use (or whatever the current legislation is called) for a car speedo to under-read. That is why they are set up like this. Also, the law specifically requires any speed 'evidence' to be provided from an approved and calibrated device. As far as I am aware, a GPS provides accurate sped, but as it is not properly calibrated there is unlikely to be any circumstances where they can be used as a defence.

3 minutes ago, Routemaster1461 said:

As far as I am aware, a GPS provides accurate sped, but as it is not properly calibrated there is unlikely to be any circumstances where they can be used as a defence.

Accuracy varies depending on the number of satellites and the speed of the car. Also, how fast the navigation system can process and update. Never mind wheel size and tyre changes. So it can probably go from 100% to 95% accurate.

Calibration for legal purposes is not something the Sat Nav people want to get it to, otherwise you would need to re-calibrate every 12 months or every 10K miles AND every tyre/wheel change and.... so it goes on. Calibration would need to be done at a Registered Calibration Station, on a rolling road, take time to do and, like the MOT, only be valid at the time it was done.

In short, it isn't going to happen unless the EU decided it is a Good Idea™ after consultation with the car manufacturers.

Sat Nav reads your movement along the ground, so is affected by hills and bends curves. It is not affected by the size of your wheels/tyres or tread wear.

Put simply it measures the time it’s taken for you to move between 2 and more bearings.

You could be on a ferry or flying. Provided you have the minivan out of satellite reception your good.

12 minutes ago, Trickiejohn said:

Put simply it measures the time it’s taken for you to move between 2 and more bearings.

True, but this depends on how often it does the calculation, If it is a second it will be of little consequence, if a minute it could make a significant difference.

13 minutes ago, Routemaster1461 said:

True, but this depends on how often it does the calculation, If it is a second it will be of little consequence, if a minute it could make a significant difference.

I think it is multiple times per second. However, it depends on how many satellites it is synched with and speed of travel and the changes of direction. At a steady 30MPH with 6 satellites, one system I used came down to a 10m accuracy. In urban areas, where you loose and re-sync satellites all the time, the accuracy can be far less. certainly at higher speeds.

...And don't forget that the constellation is orbiting, so there is a constant rising and setting of individual satellites.

After all that great info.

It sounds like in most circumstances in the UK and Europe the sat nav would be closer to the truth on your average May or major road.

Flights and shipping around Europe also use EGNOS. It produces accuracy to about a 4x6mr cube.

Japanese also only uses geostationary satellites (so my research tells me) every day i sa School day 😂

On 31/12/2025 at 00:16, Frizzbee30 said:

This is a common query on multiple forums.

No human is ever going to 'outso' the calculation obtained from the vehicle management system that is controlling fuelling.

Modern engines engines require such high precision accuracy at the injector, that the stated volume of fluid delivered is the volume delivered, not what is 'brim to brim'.

Besides human inaccuracies, the absolute impossibility of dry tank to dry tank, and allowed variation in pump delivery per cycle, the accuracy of that method isn't within a million miles of the fuelling system.

However, it's quite a testament that so many got so close! 🤣

Hmmm...Googling suggests fuel injected quantity is accurate to "a few % in lab conditions". So it's going to be worse in practise.

Forecourt fuel pumps are accurate to 0.5% - they're checked regularly due to excise duty on fuel.

I've done cumulative MPG over 10K miles so that negates any brim to brim errors. The only unknown error the odometer, but that will apply to the car's own calculation too.

Yep.

Brim to brim with zeroing of trip meter each time can't help giving the best accuracy that's easily obtainable.

If anyone can be bothered to check the accuracy of an odometer, one method would be to see how closely its readings correspond to the distance marker posts (the small posts at 100 m intervals) on the motorways. However, you'd need to check over several km, as individual distance marker posts are not guaranteed to be precisely placed. They are intended to allow people to refer to their location (after breaking down or coming upon an accident, for example) and are not to be taken as survey-grade positioning.

2 hours ago, Breezy_Pete said:

Yep.

Brim to brim with zeroing of trip meter each time can't help giving the best accuracy that's easily obtainable.

I'm not doing brim to brim - well I do, but it varies with type of use, and it's hard to replicate as you're not supposed to literally fill to the brim - I'm cumulative totalling the amount of fuel bought and the mileage. I've been doing that on the Karoq for 10K miles, so it's pretty constant now.

Brim = 1st click on nozzle.