How come every vehicle's MPG computer is off?

Joined
Feb 27, 2019
Messages
631
I always measure mileage like this:

- Fill to max.
- Set Trip B to zero.
- When it's time to fill up again, fill up to max, see how many miles are on Trip B compared to how many gallons I had to fill up to get to max.

I've done this countless times on all kinds of vehicles and the car's computer is always a few MPGs (2-3) better than what it is in actuality. Why is this? If it's just an estimate, I get that, but how come it's always in the favour of showing better mileage than you're actually getting?

I also wonder when people list their MPG figures, if they're actually calculating by hand, or if they just rely on the computer.

PS. I never pump after the first click, so it's not like I'm filling up the hose.
 
ODOMETER GALLONS $
49,240 ---------------------------------------------------Make this fill-up a new zero (as if you bought the car with 49,239 and drove a mile to a gas station)
49,520 10.021 37.50
49,730 7.923 23.25
50,022 11.437 33.68

List every fuel purchase. Tally only after a fill-up. With these figures you can easily calculate mpg, $/g,
Some people like to have a date column.

This is all you need to do. I always assumed the mpg gizmos in a car's dashboard were tainted.
I tally my car's mileage monthly.
The truck gets bimonthly tallying because I drive it so much less.
 
My 2016 Honda CR-V is off by less than 1% for the 2 fills I recently checked. That is very good considering that it only reads down to 1/10 of a MPG

The Honda CR-V has an option I use that resets trip B to 0 when you activate the lever that opens the door to the gas-cap. The thing you have to remember is to write down the trip B miles and MPGs just before you use that lever, so you record that info or it will be lost.
 
Last edited:
I don't buy your premise. Mine are dead on.

E.G. trip computer says 18.0 and trip odometer/gallons calculation showed 17.99. There is no statistical difference between those two.
 
The OBC on every BMW I owned prior to my 2er and X1 was about 5% optimistic when computing average mpg. The OBC on those two newer cars are pessimistic by a similar or greater amount. You can access the service menu and adjust the accuracy of the OBC, but it has never bothered me enough that I wanted to mess with it.
 
My 2019 Honda Pilot from day one has always estimated the MPG in the car 1.0-1.5 mpg more than when I check by hand. I have kept track of every single fill up and the car mpg and the actual hand mpg since day one. Right now just hit 15000 miles and the same 1.0-1.5 difference is still there. I use Fuelly to track everything, along with a dedicated notebook in the car.
 
The one thing VW doesn't cheat on is the accuracy of the mpg readout. All of mine are within 0.5mpg, mom's car is always dead on. The three Toyotas I've driven were always almost 3mpg optimistic. I'm sure manufacturers fudge the dash readout to increase customer satisfaction.
 
Only vehicle I really looked at the on board MPG was my '04 Elantra, and was usually within 0.5 either way of actual.

I use Fuelly.com so I know my mileage. Just enter the mileage, cost and how many gallons and it keeps track of it for me.
All kinds of usless info I can get for each vehicle.

[Linked Image from badges.fuelly.com]

[Linked Image from badges.fuelly.com]

[Linked Image from badges.fuelly.com]

[Linked Image from badges.fuelly.com]
 
2013 Kia Optima. Always less than 1 mpg difference between computer and hand calculations.

The "remaining miles" ALWAYS starts at 523 miles, no matter what. I don't know where they get that from. I always get over 700 miles per tank
 
On my two Scions, the on board mpg readouts have always been very optimistic compared to manual computation. Consistently over countless fill ups.

The xD runs about 8% optimistic, the xB2 about 11% . The tires on both are the correct size. When occasionally using zero ethanol fuel, the error spread is diminished somewhat.
 
Every vehicle that i've owned that had the computer in dash reading was 2-3 mpg higher than my hard reading. The Civic in my signature(daughter's car) is just about spot on each time(within a couple of 10ths).
 
mpg computer is for Facebook users
grin2.gif

I never pay attention to my computer mpg. I do it the old-fashioned way.
 
Originally Posted by OppositeLocK
I always measure mileage like this:

- Fill to max.
- Set Trip B to zero.
- When it's time to fill up again, fill up to max, see how many miles are on Trip B compared to how many gallons I had to fill up to get to max.

I've done this countless times on all kinds of vehicles and the car's computer is always a few MPGs (2-3) better than what it is in actuality. Why is this? If it's just an estimate, I get that, but how come it's always in the favour of showing better mileage than you're actually getting?

I also wonder when people list their MPG figures, if they're actually calculating by hand, or if they just rely on the computer.

PS. I never pump after the first click, so it's not like I'm filling up the hose.


This has been my observation in every one of my vehicles.....the car computer is a bit optimistic.
 
TFL Truck has what I think a solid system of getting accurate mpg numbers.
Fill to the 1st click, wait 30 seconds and then refill to the next click.
And they drive their loop on a known distance vs relying on the trip odo.
 
I suspect they are decent at figuring out MPG when you are moving. However, some may not "pay attention" to how much fuel you are using when you are stopped such as traffic lights or idling in the driveway to warm up, cool down, or finish that song or podcast before you go in the house.

Now my Mazda3 seems close enough. I think the car's computer lists the all time AVG MPG at 29.6 and looking at my data over the past three years I have 29.32 MPG in the AutoMobil app.

Close enough.
 
The calculatuion at fillup is only valid after long averaging.

Average over 7 - 15 fillups and see what you get.

Iv'e read in the past that some DIC MPG readout have been found to be optimistic and these cars have been recalled for reprogramming. I Don't recall the specifics. I might research that item if of interest.

I think the Automakers have learned their lesson that cheating doesnt pay.

Unless you like to pay the Federal Government enough to almost bankrupt you.
 
Last edited:
I have double checked the numbers on the Mazda and they come within a few tenths of what I come up with calculating by hand. I consider the computer to be pretty spot on.
 
My 2.7 was off about a mile and a halfoff . I adjusted it and now I am off about 3 /10 of a mile per gallon on the proud side.
 
Back
Top