Who Calculates/ How is MPG Calculated?

Status
Not open for further replies.

Nick1994

$100 site donor 2024
Joined
Feb 19, 2013
Messages
16,646
Location
Phoenix, AZ
Who does the MPG calculations for cars that we see on the sticker in the window? Is it by the EPA or the auto manufacturerer? What do they do to estimate this, actually driving it or just calculating the engine size, car weight etc. as to what they think it will be? The reason I ask is it isn't always very accurate. Below is a list of cars I've experienced that aren't true to the EPA/manufacturer's estimates. If car companies went by what their cars actually get then they'd probably sell more, obviously whatever tests they use aren't very accurate. Why can so many cars easily beat the ratings?

2002 Chevy Trailblazer LTZ 4.2L I6 w/3.73 towing gears - Thermostat stuck open, won't warm up past 1/4 way.
Rated 16 mpg city/20 mpg highway
Actual readings from onboard computer (pretty accurate actually) 17 mpg city/ 21 mpg at 80 mph, lots of hills/ aggressive driving, 26+ mpg highway at 65 mph

1996 Jeep Cherokee 4.0L w/a little bit bigger tires
Rated 14 mpg city/18 mpg highway
Actual calculation was 21.8 mpg highway with 800 pounds of people in it plus lots of stuff in it (calculation compensated for tire size)

2004 VW Beetle TDI
Rated 30 mpg city/38 mpg highway
Actual calculation 35 mpg city, 38 mpg mixed driving, 44 mpg highway at 65 mph

1997 Toyota Camry I4 -
Rated 20 mpg city/28 mpg highway
Actual calculation 23 mpg city, 25 mpg mixed, 33 mpg at 80 mph, 36.72 mpg at 65 mph

1996 Lexus ES300 V6 - Burns 1 quart of oil per 400/500 miles
Rated 21 mpg city/ 26 mpg highway
Actual calculation was 28/29 mpg at 100% of tank at 80 mph with some hills
 
On board computer could be off. Only real way to tell is to reset the trip odometer every time you fill up and see exactly how much gas you're putting in. Also if you read the EPA site, there's a formula that they use to calculate mileage, it includes some stops and some varying speed to simulate actual conditions. At a steady speed, you may actually get better mileage which is why the term your mileage may vary is a pretty common saying.

As for who actually tests it, it's the auto manufacturers who test it according to the testing protocol that the EPA has. They randomly do their own testing just to make sure it's accurate. That's why occasionally you see the auto makers get fined when their numbers don't quite line up with what they said it was.

And just to average things out, I've always gotten worse mileage than advertised. I think the computer in my car might be off by .5 to one mpg. I think I got about 12 mpg once when stuck in traffic when I'm supposed to get around 16, it was very slow going through.
 
Originally Posted By: Wolf359
On board computer could be off. Only real way to tell is to reset the trip odometer every time you fill up and see exactly how much gas you're putting in. Also if you read the EPA site, there's a formula that they use to calculate mileage, it includes some stops and some varying speed to simulate actual conditions. At a steady speed, you may actually get better mileage which is why the term your mileage may vary is a pretty common saying.

As for who actually tests it, it's the auto manufacturers who test it according to the testing protocol that the EPA has. They randomly do their own testing just to make sure it's accurate. That's why occasionally you see the auto makers get fined when their numbers don't quite line up with what they said it was.

And just to average things out, I've always gotten worse mileage than advertised. I think the computer in my car might be off by .5 to one mpg. I think I got about 12 mpg once when stuck in traffic when I'm supposed to get around 16, it was very slow going through.
The Trailblazer is the only one with an onboard computer that shows the mpg and I have calculated it by hand and reset the mpg when the tank had been previously filled and it was within 1 mpg. When I drive the Trailblazer around town aggressively it has no problem showing me it's getting 12-13 mpg. Other cars have been tested by hand.
 
They're not always over. I remember hearing/reading about a women who tries to sue Honda. She bought a hybrid that claimed "X" mpg. After she purchased it and drove it for a few months she found out the mileage was far off. Its actually becoming a popular trend to see manufacturers claiming mpg that many people will never see. Be thankful you get more.
 
Officially quoted manufacturer numbers are done on a test cycle set by (I think) the EPA for you guys! We have the NEDC.

Problem is, manufacturers know what these tests consist of and adjust/map/ballast cars accordingly.

The only way to know for definite is to put exactly 1 gallon in your tank and measure the route with gps. Thatll give you a precise figure for that journey. Repeat in other environments to gain understanding of the numbers in those environments
 
If i recall correctly, the car manufacturers calculate the MPG based off the EPA testing protocol. However, I recall the protocol was changed around 1998 so that vehicles would show a more accurate MPG estimate. Prior to that, the MPG listed on the sticker tended to be higher than what the vehicle would actually produce in real world usage.

Considering how E10 seems to have become the norm, I wonder what type of fuel is required for testing now.
 
Hello, To Nick1994: In your Jeep Cherokee example you said, "calculation compensated for tire size".
How'd you do that?

FOI: (For Our Information)
I simply start with a fill up and note the miles on the ODO.
I make sure to fill up at month's end and note the ODO.
Add up all fuel purchases. The total=GALLONS.
Subtract the starting mileage from the ending mileage. The result=MILES.
MILES/GALLONS=mpg.

Here's comes the part where I'll get scolded for flaming.
On board computers are really only good for "on the fly" notation.
It's so simple to calculate accurate mileage. People should just do that.

The methods employed by auto makers and/or a government agency are going to be negotiated with "results" having to fit a marketing profile. Kira
 
I would have to agree that its the automakers who do the testing. Hyundai was off by so much % with my moms Elantra and now every time she goes to the dealer (which is hardly ever.
smile.gif
), they check her mileage and give her a pre paid credit card for compensation. Its like $60 every 9 months or so.. I cant even remember.

The best way IMO to calculate your mileage is to fill the vehicle up at the pump on the slowest setting. Drive and record the mileage.. and fill up at the station with the same pump on the same slow setting. See how much gas you put in there and do the math.
 
Nick, "your milage may vary".

Problem is, the EPA test is a standardized test that simply can't accomodate for everyone in the USA. Some cities are nothing but stoplight drag races; some areas are miserable commutes. And a lot of people plain can't drive, let alone drive for mpg's.

The testing attempted to come up with a standard, which was then mocked for years (decades?). In 2008 they basically went and knocked 10% off in an attempt to "match" real world results. IIRC the test didn't change, they just lopped some percentage off.

I don't think the proceedure allows one to engine brake, for example. Or not aggressively. Standardized shift schedules for stickshifts too. Hence a stickshift owner can trounce the ratings by short shifting into high, and aggressively engine brake when stopping. Automatic owners don't have it so easy, but they too can beat the ratings by good old fashioned conservative driving. Everyone can pump up their tires for a bit more mpg--except for the EPA, which has to go by the doorjamb.

You and I probably have driving routes which look nothing like the EPA highway route--very little stop and go.
 
Originally Posted By: Nick1994
Why can so many cars easily beat the ratings?

I wish that was the case with our Q5. It's EPA rated at 18/26. I'm lucky to get 24 mpg on the hwy, even when I'm very gentle with the throttle.
 
Originally Posted By: Nick1994
Who does the MPG calculations for cars that we see on the sticker in the window? Is it by the EPA or the auto manufacturerer? What do they do to estimate this, actually driving it or just calculating the engine size, car weight etc. as to what they think it will be?


Info from the EPA

http://www.fueleconomy.gov/feg/info.shtml

EPA estimates are based on laboratory tests conducted by manufacturers according to federal regulations. EPA re-tests about 10% of vehicle models to confirm manufacturer's results. For more detailed information, visit our page on How Vehicles Are Tested.

How the test are run:

http://www.fueleconomy.gov/feg/how_tested.shtml

Fuel economy is measured under controlled conditions in a laboratory using a standardized test procedure specified by federal law.
 
Last edited:
"Your mileage may vary, depending on how you drive."
That statement goes back to the '70's and was a disclaimer in car commercials of the day. It's still true.
 
Wouldn't it also depend on fuel quality, brand, winter blend etc?

Not so much for Nick though seeing how he lives in AZ.
smile.gif
 
Last edited:
I've decided mpg doesn't matter at all. My car is rated 28/36 and I get 17 driving in the city in winter, even really calmly. I can get 42 driving highway in the summer, even going 5mph over the speed limit.
 
You have ideal conditions with no rain, warmer air than most of the country and even your city commute is probably not as city as the EPA cycle. Also tires and their PSI can have a strong impact.

Many sporty powerful cars/trucks come with performance tires which don't compromise to get that last 5% fuel economy.

An example my 2015 forester(boring car) came with AWFUL low RR tires to get that last 5% they are below average in winter conditions on a car marketed as being excellent in the winter.. but hey they got that last 1.1mpg to put on the sticker.

Not much more dangerous in the winter than an AWD car on bad tires.. you dont just slip the front or rear tires.. you have all 4 spinning. With winter tires this is minimized as even when they are slipping they still retain good traction including lateral traction.
 
Last edited:
I use Fuelly.

With that I find that my onboard computer is fairly accurate. The GTI gets about what it was rated at; I don't track the Camry.
 
Originally Posted By: sicko
However, I recall the protocol was changed around 1998 so that vehicles would show a more accurate MPG estimate. Prior to that, the MPG listed on the sticker tended to be higher than what the vehicle would actually produce in real world usage.


Yes, the testing protocol was changed, but it was done in 2008, not 1998. Interestingly, the cars I've owned (or at least how or where I drive) produces MPG numbers closer to the "old" estimates than the "new" estimates.

Our 2005 MDX was originally rated 17/23, which is about what we get. The new estimates are 15/21, and we rarely get that low.

Our 2008 CR-V is rated under the new regime at 20/26, and I rarely get that low. The 2007 CR-V, same vehicle and drivetrain, is 22/28, which is about what we get.

Climate and topography makes a huge difference. We used to live in the hills of southwest Virginia, where everything's on a curve or an incline and town roads were curvy and 25 mph. When we moved to North Carolina about 12 years ago, our MPG in the vehicles we owned then improved about 25%, just because of the difference in terrain. Roads here are much flatter and speed limits are much higher (so your average speed is higher). All of that really aids fuel economy.
 
Status
Not open for further replies.
Back
Top