The gap between the estimated fuel economy demonstrated on a car's window sticker, and that which is achieved while driving in the real world, has always existed. Further, there has always been a difference between the mile-per-gallon numbers submitted to the Environmental Protection Agency by automakers, and those that the agency has garnered for itself in independent testing. In fact, according to an Automotive News report, the EPA believes that gap has gotten "unacceptably high" over recent years. The difference between EPA and automaker estimates was 1.15 percent in 2007, and jumped to 2.25 percent in 2010.
One of the reasons at the heart of this gap is the ability for skilled test drivers to increase ratings by "smoothing out" the prescribed EPA test procedures. By carefully accelerating and braking, experienced drivers can save fuel, and gain an advantage for the test car in question. EPA spokesperson Christopher Grundler told AN that, "The industry has some very good drivers, and we've noticed."

As the average price of gasoline increases, and the importance to the public of accurate mpg numbers also rises, the EPA is seeing the necessity of making its test standards more stringent and more accurate. The organization is now requiring that automaker submit "drive trace" reports; electronic logs of driving behavior during testing. The EPA is also working with SAE International to further refine the process, and to help eliminate the affects of things like ultra-skilled test drivers.


I'm reporting this comment as:

Reported comments and users are reviewed by Autoblog staff 24 hours a day, seven days a week to determine whether they violate Community Guideline. Accounts are penalized for Community Guidelines violations and serious or repeated violations can lead to account termination.


    • 1 Second Ago
  • 109 Comments
      mustang_sallad
      • 1 Year Ago
      I've spent a fair amount of time on a dyno driving through the EPA cycles, and I've definitely been surprised at how much of a window they allow around the target speed. These guys are smoothing out the accelerations and deccelerations while staying within the allowed tolerance. I think they simply need to tighten this window up. In fact, now that almost everything is Throttle-by-wire, I wonder if it would be much of a stretch to automate the throttle control while on the dyno. Although I suppose there's some braking involved too, that would make it more complicated.
        Seal Rchin
        • 1 Year Ago
        @mustang_sallad
        Any time you have a test you can sort of study to it, they key is to constantly improve it.
          Greg
          • 1 Year Ago
          @Seal Rchin
          CarNutMike - I agree. Tightening the tolerances should not affect year-to-year comparisons.
          CarNutMike
          • 1 Year Ago
          @Seal Rchin
          I don't disagree, but the problem is that as soon as you change it, you lose the ability to compare MPG numbers from year to year.
      imnuts
      • 1 Year Ago
      Solution, make it a computer controlled test that monitors car data via OBD2 and changes throttle and brakes based on the predetermined test route. As long as the equipment isn't failing, the test should provide repeatable results, regardless of what company is doing the testing.
        raktmn
        • 1 Year Ago
        @imnuts
        That might be a good idea for an update on a testing process that was invented way back before ODB2 had even been invented.
      • 1 Year Ago
      [blocked]
        Andrew Ramos
        • 1 Year Ago
        Fuel economy label ratings are made from tests indoors but must be run in environmentally controlled test cells. It's all part of the regulations. In fact, part of the label rating is from testing in hot ambient conditions with solar lights and the vehicle's AC on. Testing must also be done in a cold chamber. Indoors sells the testing a bit short.
        montoym
        • 1 Year Ago
        Indoor testing under strict environmental conditions is the only way to achieve consistent data from test to test. if they tested outdoors, how would you compare a test done in, say Minnesota in Janaury to a test done in Arizona in August? You're never going to find a consistent, reliable environment to test in outdoors, so a controlled, indoor testing facility is the best option.
      EZEE
      • 1 Year Ago
      How's about we educate people on how to drive to get the most out of their mileage? I always beat the EPA when I drive. It don't take no PhD.
        Greg
        • 1 Year Ago
        @EZEE
        Since the problem reported in the article has nothing to do with what Joe Public gets driving in the real world but instead how car companies have figured out a way to take advantage of the tolerances of the test, educating the public on how to drive does nothing about gaming the tests and/or cheating.
        • 1 Year Ago
        @EZEE
        [blocked]
        Andrew
        • 1 Year Ago
        @EZEE
        Word up. "Very good drivers..." People who suck at driving deserve to pay more for gas. Or, put more charitably, people who choose to drive aggressively can pay to play.
          Fonin
          • 1 Year Ago
          @Andrew
          the humor to me is, i see more Prii, Corrollas and econoboxes flat-footing way more than the 'gas guzzlers'
        johnb
        • 1 Year Ago
        @EZEE
        agree, I haven't owned a car where my averages are below epa estimates. I always figured they just drove the cars a bit harder than I regularly do.
        S.
        • 1 Year Ago
        @EZEE
        Same - on the car I always get above EPA without even trying. On the moto, not so much. I'm always under, but I ride it very spirited-ly, which is the entire point.
      Black Dyanmite
      • 1 Year Ago
      Or the EPA can get off their holes and test each car as they make powertrain changes. God forbid they do their own work, and get the roosters out of the henhouse...... BD
        raktmn
        • 1 Year Ago
        @Black Dyanmite
        The decision for the companies themselves to do their own tests and not the EPA was the idea of the companies, their lobbyists, and the Republicans they bought. It was the age old Republican argument that the companies can patrol themselves, and that they know more than any "gubberment employee" at the EPA. But heck, don't blame private industry for their own failures when they make a mess out of their fuel economy tests they do themselves. It must be the EPA's fault, right?
          80Mirada
          • 1 Year Ago
          @raktmn
          Bitter much? The 'republicans' are patrolled by the consumers. Several times customers complained about mpg's and win lawsuits while automakers have PR nightmares.
          raktmn
          • 1 Year Ago
          @raktmn
          80Mirada, nobody is being patrolled by consumers. Congress has a 90% retention rate, while only having a 10% popularity rating. Yes I am bitter about lobbyists having more power over our laws than the people. You aren't?
          80Mirada
          • 1 Year Ago
          @raktmn
          Also, its easy to beat the epa's numbers. It just takes common sense. Unfortunately, most people don't care about mpg's or don't have common sense.
          80Mirada
          • 1 Year Ago
          @raktmn
          Companies. Stupid small keys on my phone.
      • 1 Year Ago
      [blocked]
        The Wasp
        • 1 Year Ago
        I don't have a problem with people driving however they want and not meeting EPA ratings. I have a problem with EPA ratings being unrealistic for normal driving (ie non-aggressive, non-hypermiling). It should be a standardized rating, based on a standardized test, with standardized testing protocol -- carried out by third-party testers.
      Menel
      • 1 Year Ago
      Maybe we should fire EPA, example of poor government if they can't get good drivers. They are probably the reason my 2005 Honda is rated 20c-28h and I routinely pull 32-34 on interstate trips and 27 during the week city commuting. Honda probably could have rated more accurately.
        Spec
        • 1 Year Ago
        @Menel
        The carmakers supply the drivers. Reading comprehension!
          raktmn
          • 1 Year Ago
          @Spec
          I'm sure he read it. He even comprehended it. But it didn't fit in the anti-govt rhetoric he wanted to post, so he just changed the facts he knew were true, until they fit his narrative.
        CarNutMike
        • 1 Year Ago
        @Menel
        if you look at the highway drive cycle, you'll see why. LOTS of speed changes that don't reflect highway driving for most people.
          Greg
          • 1 Year Ago
          @CarNutMike
          I disagree to a point. I don't think the hwy number is supposed to reflect what you get at any instant on the freeway. (After all, most trips with a majority of freeway driving still involve starting the car, getting to the freeway, exiting, etc.) But I also have doubts that the "highway" test is even related to the actual hwy. It may be nothing more than a higher speed, fewer stops cycle, and the EPA needed something to call it. They may have simply picked "city" and "highway" because they couldn't come up with any better names.
      Worx2749
      • 1 Year Ago
      I get 50% better mileage than EPA because I've stopped using my brakes.
      CarNutMike
      • 1 Year Ago
      This positively idiotic and the EPA should be embarrassed. THEY supply the friggin drive cycle. As long as the drivers for the manufacturers are compliant, that's it. Done. "Guys, sorry, but you're too compliant to the test WE defined. Could you do worse?" WTF? Why doesn't the EPA use drivers who know what the hell they're doing instead? And FFS, 2.25% in a 25mpg car is 0.625mpg. ONE mpg on the sticker, at best.
        CarNutMike
        • 1 Year Ago
        @CarNutMike
        Maybe the 1.10% (seriously? 1.1%?) "improvement" since 2007 is just a different EPA driver who sucks a little bit more.
        montoym
        • 1 Year Ago
        @CarNutMike
        The point of this article is that those drivers are not compliant, or rather they are adjusting their style enough as to affect the outcome.
      Alex Ingram
      • 1 Year Ago
      Well, I doubt any of us could lap a Porsche 918 around the Nordschleife in 6:58 either, so it shouldn't be surprising that a professional driver can squeeze a better mpg figure out of a car too. A pinch of salt needs to be taken with manufacturers' claims, whether about performance or economy.
      icemilkcoffee
      • 1 Year Ago
      Finally the EPA wakes up to the rampant cheating that has been plaguing the self reported mpg ratings. It is time to end this joke of letting foxes guard the henhouse. The EPA needs to put its foot down and take over all the testing.
        CarNutMike
        • 1 Year Ago
        @icemilkcoffee
        The difference is 2.25%. That's "rampant cheating"?
          icemilkcoffee
          • 1 Year Ago
          @CarNutMike
          That is the AVERAGE. Ford was just caught cheating by something like 7mpg on the C-Max. This is being horribly unfair to the honest carmakers.
          merlot066
          • 1 Year Ago
          @CarNutMike
          The C-Max wasn't off by 7 MPG because a driver gamed the test, it's because there are ridiculous loopholes that were started in the 70s that multiple automakers take advantage of. The EPA is extremely lazy in regards to vehicle fuel economy. They have absurd exceptions from decades ago and even when they "redesigned" the city/highway testing procedure they just added a new cycle and averaged the numbers with the old cycles.
        David MacGillis
        • 1 Year Ago
        @icemilkcoffee
        Rampant cheating? 2.25%
      sp33dklz
      • 1 Year Ago
      It's a competition. WTF do you expect when you tell them to meet a threshold, do it the most difficult way possible? The EPA needs to get off their high horse and test differently. If an individual decides not to drive a car like Ms. Daisy, it's their own fault. There's a reason my WRX gets 18mpg on some tanks, and 25 on others. Get real people.
    • Load More Comments