Gas Mileage: Consumer Retorts

Consumer Reports’ real-world mpg figures make the Prius even more appealing 22

Consumer Reports recently claimed that EPA's vehicle ratings routinely overstate how fuel-efficient cars and trucks are in real-world driving. For standard cars and trucks, the magazine says, EPA's ratings overstate real-world fuel economy by 30 percent. But for small hybrids, such as the Toyota Prius, they claim that EPA overstates actual miles-per-gallon by a hefty 42 percent. (Ouch.)

Now, I believe that there's reason to question Consumer Reports' figures. Of course, I have read a number of reports that the Toyota Prius doesn't actually get the EPA-rated 55 mpg in combined city/highway driving (though some people -- particularly those who've optimized their hybrid-driving habits -- get pretty close, and these folks actually squeezed out 110 mpg from their Prius, albeit in highly non-standard driving conditions). But I'd never heard any claim that the typical Prius averages just 32 mpg -- which is what the magazine's figures suggest. See this comment by WorldChanging's Jamais Cascio for a similar take.

But, just for the sake of argument, let's take the CR figures at face value, and assume that small hybrids' mileage really is overstated by 42 percent, vs. just 30 percent for regular cars. Doesn't the higher mpg reduction for hybrids suggest that their fuel-savings advantages vs. regular cars are overstated -- and that they don't save as much money as advertised?

Actually, no. As counterintuitive as it may sound, the Consumer Reports figures, on their face, actually bolster the economic case for buying hybrids.

As we've said before, mpg math is tricky. And though it may be hard to believe at first, Consumer Reports' figures suggest that the Prius is an even better choice in the real world than EPA's fuel economy ratings would suggest.

Consider, for example, a regular, non-hybrid car with an EPA fuel economy rating of 30 mpg. "Officially" it burns 2 gallons of gas every 60 miles. But Consumer Reports estimates that the vehicle actually would get 21 mpg in real world driving -- 30 percent less than advertised. Which means that over the course of 60 miles, the car actually burns 2.9 gallons of gas.

Now consider the Prius, with an overall EPA fuel economy rating of 55 mpg. At its advertised mileage, it burns about 1.1 gallons of gas every 60 miles. But if its mileage is reduced to 32 mpg -- a 42 percent reduction, per Consumer Reports' estimate for small hybrids -- then it uses about 1.9 gallons of gas every 60 miles.

So that gives us...

Gallons consumed in 60 miles of driving
  Ideal Real World
Prius
(rated 55 mpg)
1.1 1.9
Regular car
(rated 30 mpg)
2 2.9

Look at the "ideal" column -- which represents how much gas EPA says the two cars should burn over 60 miles. The Prius has a .9 gallon advantage -- a nice bonus. But look at the "real world" column: as estimated using Consumer Reports' figures, the "real world" Prius has a full one-gallon advantage over the "real world" regular car.

In other words, the Prius is actually an even better deal -- roughly 10 percent better -- in the real world than it is in the abstract. And compared with lower-mileage cars and trucks, the "real world" Prius looks better still.

Now, this obviously isn't evidence for or against Consumer Reports' estimates for hybrids. And it doesn't do much to change my assessment that buying a Prius can be a pretty pricey way to reduce your greenhouse gas emissions.

Still, it's good to remember that, as Barbie famously said, math is hard -- and miles-per-gallon math turns out to be among the more counterintuitive gauges that Americans are expected to understand. So it's important to actually run the numbers. Apparently, at least when it comes to gas mileage, it's just not enough to trust your instincts.

Clark Williams-Derry is research director for the Seattle-based Sightline Institute, a nonprofit sustainability think tank working to promote smart solutions for the Pacific Northwest. He was formerly the webmaster for Grist.

Advertisement
Advertisement
  1. ben Posted 10:32 am
    12 Sep 2005

    diesel is definitely more cost effectiveCars like the VW Golf TDI are definitely more cost effective than many hybrids.

    theWatt.com
  2. Biodiversivist's avatar

    Biodiversivist Posted 4:52 pm
    12 Sep 2005

    The problem with Consumer reportsTheir ratings do not rely on statistically significant samples. They are little better than dice rolls. I have a friend who always turns to consumer reports before any purchase. He was looking for a new wireless phone. I looked at the phone matrix and noticed that it did not include the kind of battery used. I commented that NiMH batteries were far superior to NiCAD and that it was odd that they would leave such a critical feature out of the ratings. My friend, having the utmost confidence in Consumer reports scoffed at my critique and bought the highest rated phone. Two weeks later it caught fire.

    In the end, it all comes down to biodiversity. Help acquire and protect ecological hotspots, give to a conservation organization: http://www.saveourbiodiversity.com
  3. odograph Posted 12:14 am
    13 Sep 2005

    real-world numbersI really think this link should be in every story about "hybrid mileage":
    http://www.greenhybrid.com/compare/mileage/
    It looks to me like most people are beating Consumer Report's numbers, while falling short of the EPA's.
    It also shows clearly which hybrids are high-mpg models.
    FWIW, I've been getting 50 mpg out of my Prius very consistently.  I think that is due to two factors: 1) I try to keep the mpg up and 2) my "trip mix" supports it.  Many short hops around town drag my mpg down, and a few open highway runs bring it up again - back to the 50 mpg average.
  4. odograph Posted 12:33 am
    13 Sep 2005

    survey, don't simulateJust to be clear, I think the best numbers are always going to come from the real world (surveys) rather than from simulations of the real world.
    As soon as a would-be simulator defines his test loop, he eliminates its usefullness to people who don't drive in such conditions.
    I think the focus should be on capturing real-world data from the fleet.
  5. cyrussh Posted 7:20 am
    13 Sep 2005

    problem with your numbers but not your logicLet me start by saying I am actually in the process of trying to buy a Prius so my comments are in no way meant to put down the Prius.  To say that it is actually a better spread of .9 vs 1 is a bit of a misrepresentation since you are comparing apples to oranges.
                   Ideal Real World

    Prius

    (rated 55 mpg) 1.59     1.9

    Regular car

    (rated 30 mpg) 2.9     2.9
    This means that where the Prius had a 1.31 gallon advantage in the ideal world it slips to 1 in the ideal world.
    Everything I have heard about the real world though makes consumer reports irrelevant since I have mostly heard of people getting around 50 mpg.

  6. Clark Williams-Derry's avatar

    Clark Williams-Derry Posted 2:28 am
    15 Sep 2005

    Not sure I get that, cyrussh"Ideal" Prius gets 55 mpg.  Over 60 miles, it uses about 1.1 gallons of gas.
    "Consumer Reports" Prius gets 32 mpg -- or 42% less than the "Ideal".  In 60 miles of driving, it uses about 1.9 gallons of gas.
    "Ideal" car gets 30mpg.  Over 60 miles it uses 2 gallons of gas.
    "Consumer Reports" car gets 21 mpg -- or 30% less than the "ideal".  In 60 miles of driving, it uses about 2.9 gallons of gas.
    So - the Ideal Prius uses .9 gallons less than ideal car over 60 miles.  But the "Consumer Reports" Prius uses 1 gallon less than "Consumer Reports" car over the same distnace.
    So my real point is an entirely geeky one -- that if you take Consumer Reports numbers at face value, a 42% mpg reduction for a Prius, vs. a 30% reduction for the standard car actually makes the Prius look better.  Which is an example of how mpg math is just really screwy.
    And just for the record, I don't take Consumer Reports' figures at face value.  I think they're actually mistaken.  See here, for example, for mileage figures from 900 Priuses, and 9,000 tanks of gas -- they average about 47.9 mpg for the Prius.  And nobody reported the mpg figures that Consumer Reports seems to imply.  Unless this represents a screwy sample, it seems to me that you just can't rely on that CR article.
  7. pblanchard6 Posted 2:35 am
    05 Jul 2007

    prius gas mileageI purchased a 2007 prius touring 2 weeks ago.  We went on a trip from RI to Erie PA. We drove a mix of highway and back roads up and down mtns through out upper state NY. We avg 47 mpg 1st tank 68 mpg 2nd tank and 57 mpg the third tank of fuel.  We drove from Eire PA back to RI on one tank of fuel.  Car has 3000 miles now and lowest mpg we got was 47 mpg.  I always fill the tank, 0 my miles and divide gal by miles. The car handles great has plenty of pep and is fun to drive.
  8. odograph Posted 3:06 am
    05 Jul 2007

    come on pblanchard6We know you Prius drivers do 100 MPH and drugs too.
    (Actually I'm a Prius driver too, my old comments above.  I wonder if the Gore arrest helps or hurts our image?)
  9. Delay And Deny's avatar

    Delay And Deny Posted 3:29 am
    05 Jul 2007

    Green Math

    Why is it that when the IPCC uses crazy convoluted math to try and "force" us to believe in the myth of anthropogenic global heating, Grist laps it up like condensed milk on a cat's tongue.
    Yet, when a simple fact finding establishment like Consumer Reports lays it out for all to see, viz., hybrids are b.s. -- you have to invent all kinds of reason why it's really saying, that you should buy a hybrid?!



    John Bailo


    You Read It Here First
  10. GreyFlcn Posted 3:39 am
    05 Jul 2007

    Pretty muchYou can expect 45mpg highway, 60mpg city
    Basically a hybrid and a diesel are about comprable for highway driving.
    http://greyfalcon.net/electriccars2.png
  11. gmunger Posted 4:34 am
    05 Jul 2007

    keepin' it even realerMy '06 VW Jetta actually gets better mileage in the real world than the EPA ratings (46-52 mpg vs. 42). And the scuttlebutt has it that this is very common for these cars.
    If only they would produce and market these vehicles in serious fashion.....
  12. GreyFlcn Posted 4:38 am
    05 Jul 2007

    Is thatIs that Jetta a Diesel?
  13. Delay And Deny's avatar

    Delay And Deny Posted 4:43 am
    05 Jul 2007

    Get A VoltThis is all why I'm saving my coin for a GM/Chevy Volt.
    150 mpg plugin hybrid.
    Toyota will have a Prius plugin next year.
    Problem solved.
    Gasoline companies, ESOD.

    John Bailo


    You Read It Here First
  14. GreyFlcn Posted 4:49 am
    05 Jul 2007

    UhmLast I checked, the Volt won't be out in any numbers till 2011
    And I haven't heard of an official Toyota Plugin before.
    Hell, last I heard they recanted on switching their batteries from NiMH to Lithium.
  15. gmunger Posted 6:04 am
    05 Jul 2007

    oopsYes, it's a diesel.
  16. odograph Posted 8:30 am
    05 Jul 2007

    I said aboveCheck out the real-world data.  Trust people who own them and drive them.  Average over their experience:
    EPA "Shared" MPG database
    The average for my year (146 vehicles, not including me) is      

    47.6 MPG

    And for what it's worth, Grey, it's not "city vs highway" on a Prius.  It's short trips (less than 10 miles) versus long trips (15+ ?).   You'll get close to 50 if you drive reasonably in city or highway ... but short hops will drag you down.
    more explanation here
    P.S. - the "trip length" thing probably explains CRs bad numbers.  They drove, stopped, compared notes, drove more ... and never got into the groove.
  17. odograph Posted 8:48 am
    05 Jul 2007

    P.S.It used to bug me that reviewers would not point people to the "shared" databases.  But, when you think about it ... why should they?  It undermines their value as the "experts."
    To bad in the web world we consumers can just share data.
  18. spaceshaper's avatar

    spaceshaper Posted 12:04 pm
    05 Jul 2007

    Miracle carSo let me get this straight - the further you drive a Prius - the less gas it uses?

    The true meaning of life is to plant trees, under whose shade you do not expect to sit.
  19. odograph Posted 10:31 pm
    05 Jul 2007

    no ;-)It uses more gas, but reports higher MPG.
    That seems counter-intuitive at first, but it's the truth.  When I started my current job I had an 8 mile commute each way.  Since the Prius spent a higher fraction of that time warming up (and not running on electric power alone), I got about 45 mpg.  The office moved.  Now it's a 20 mile drive to work.  That allows the engine to warm up more.  I get 47-48 mpg.  And of course I burn about twice as much gas.
    I guess that's just the way of the world.  Higher MPG is nice, but a shorter commute (if you can manage it) matters more.
    Back to CR ... I think the bottom line is that their test mix does not match average consumer behavior.  This for the simple reason that large numbers of consumers average about 47 mpg.  If they don't match that ... their test is wrong.
  20. SustainableGreen Posted 12:45 am
    06 Jul 2007

    Gallons per 60 miles?!Hey, all:
    Lest we enter into a new era of using whatever statistic or rate we want whenever we want it, much like Humpty-Dumpty uses words, I think we need to stamp out this silly attempt to substitute an inane measurement for a simple, widely used, understood and accepted one.  
    Otherwise I propose a new system of length measurement: quarks, larger quarks, really large quarks, and 'blow-yer-mind big-ass quarks!  
    The entire discussion here hinges on the validity of mileage ratings, when what matters is real-world mileage.  Either EPA needs to test real cars in controlled tests on public roads, or they need to collect legitimate mileage experience from consumers.  The "rating" practice is fraught with bullshit and distortion and does not serve the public.  
    David

    Sustainability For Life
    Messages done with sustainable energy, with Wind and Sun!  
  21. odograph Posted 4:18 am
    06 Jul 2007

    SustainableI've given the link to the EPA's "shared mileage database" twice now.
    It has the real world numbers you seek.
  22. Nucbuddy Posted 4:39 am
    06 Jul 2007

    The Chevy Volt, and its built-in SegwayGreyFlcn wrote: I checked [...] the Volt won't be out in any numbers till 2011
    Where was it that you checked?
    By the way, the Volt may come with a built-in Segway.

    carmagazine.co.uk/news.php?sid=830&page=1

Add a Comment

You are not logged in. Thus, you cannot post a comment. If you have an account, log in. If you don't have an account, well, by all means go make one! Meet you back here in five.

Hello, Visitor!    Why not register?

Advertisement