The manufacturers do not care. Their goal is to meet the letter of the law and no more. And sometimes not even that if they think they can get away with it (lookin’ at you, VW).
As long as the law continues to allow for non-realworld testing to meet its standards, they will keep playing these games.
Nobody will ever be happy with EPA figures.
For the test to be relevant, it has to be repeatable, so it has to be formal.
You can always find whatever test procedure was used to be “artificial” to compare to “real world driving”.
It will always be a driver who drives in “binary mode” with one of the pedals down the the ground, either gas of brake and another person taking eternity to accelerate.
Both will find EPA figures artificial.
dealing with my lemon Honda Accord (it is in litigation now, BTW), I was given two same model year CR-Vs as loaners - one with older 2.4L normally aspirated engine and one with 1.5 turbo.
I had quite an extensive “test drive”, as I logged almost 1500 miles on these
in the real-life driving, 1.5L turbo was beating 2.4L “classic” one on all accounts: it had better acceleration and better MPG, both in city and on highway and by a good margin
the only thing which annoyed me in 1.5 turbo was the distinct high-frequency mosquito buzz on the turbo, it was faint, but because of that probably even more annoying
Formal is fine, unrealistic is not. The top speed for highway mileage testing is 60mph. The top speed on many highways is 10 higher than that, and in some states it’s 20 higher.
Additionally, the cars are tested inside on a dyno, so wind resistance is not a factor. They’re tested without the AC running, which does not reflect real-world use.
Specifically regarding the turbos, they’re there to make the car accelerate faster, which encourages people to push the go-pedal harder. But that’s not how they’re tested, and the car makers know this, so they get a free mileage bump that isn’t reflected in the real world.
On my last examples, I was able to get EPA numbers on both CR-Vs, highway miles were clocked on 65-75.
On my [lemon] Accord Hybrid, I can easily get highway figure of 48MPG, but by no means can get to that EPA-promised 48 in city, 45-46 is the best I can clock, and yes, once AC is on, I can barely even get to 40 MPG
My two similar turbo 4 cylinder sedans… An '01 Saab 9-5 and '14 Audi A4. Both produce similar HP and torque. The Audi’s engine is 300 cc smaller and rated 10 hp lower but direct injection. Both weigh about 3600 lbs. Both are similar in size and aero drag. Both are similar in 0-60 mph times.
Both cars consistently beat their EPA highway economy ratings by at 10% and the Audi is 10% overall better than the Saab was and is AWD. Doesn’t matter whether I’m driving or my wife (they were both her cars) the results are the same.
It has not been my experience that the EPA tests are overly optimistic with turbocharged cars. As always, the driver is the biggest factor in the mpg story.
In addition, the EPA test isn’t meant to predict what your mileage will be. It is designed to allow comparison between similar vehicles so that the buyer can make an informed choice. Sometimes manufacturers have gamed the system an they were caught eventually. IIRC, Hyundai had overly optimistic gas mileages in some of their vehicles about 8 years ago and made changes eventually to their methods to provide more accurate estimates.
Let’s say the EPA numbers were exactly right, on average. Then there’d be millions of people upset that they were getting less than the average.
Actually, wind resistance is taken into account.
Straightforward calculation, given the coefficient of drag.
" Beginning with 2008 models, [three additional tests] are used to adjust the city and highway estimates to account for higher speeds, air conditioning use, and colder temperatures."
I know some turbo cars appear to benefit from the EPA test, but the Jetta I rented for a month (1.4 L turbo, great performance) got better than the 40 mpg highway EPA rating, typically getting 45 mpg on the highway. That eliminates any need for a diesel option.
I’m not saying you’re wrong, but nothing on the page you linked to indicates that any sort of accounting for wind resistance is being used. It says they have 3 new tests which account for faster speeds, AC, and cold outside temperatures. It also does not say anything that I could find about whether or not those three new tests are somehow factored into the city and highway numbers that you see on the window sticker, or if those numbers still continue to reflect only the standard city/highway tests.
That’s because VW is incapable of designing a diesel engine that can meet emission standards without cheating.
Tester
Um, you have to dig a little deeper if you want details.
“Engineers adjust the amount of energy required to move the rollers to account for wind resistance and the vehicle’s weight.” [/hand-holding]
True, but I’m saying even if their diesels didn’t cheat they wouldn’t be the best option for high mpgs.
Just because you didn’t link properly doesn’t mean you have to be rude about it. Thanks for correcting your bad initial link.
Now, got numbers for how much resistance is added, and whether or not it accounts for the actual aerodynamics of the vehicle being tested or is just a generic fudge factor?
The first link was an overview.
The second link was easy to find from the first, if only you’d make a little effort.
If you’re not fully satisfied with my service, you can have your money back. 
Nothing free about it. But worth the extra cost and complexity.
