By on June 20, 2008

corius.jpgTwo Duke University Business School researchers studied the difficulty in perceiving gas saved when comparing vehicles rated by miles per gallon (mpg), versus gallons per 100 miles (gpm). The eggheads argue that the latter more closely reflects real world savings. The article [sub] and accompanying notes illustrate the problem: a nonlinear relationship between mpg and gpm. An example… At 10k miles per year, moving from, say, a Chevrolet Tahoe at 15mpg (6.7gpm) to a Chevy Traverse (that's a Buick Enclave for dummies) getting 20mpg (5.0gpm) nets a savings of 166 gallons. [ED: providing you  could get someone to buy the Tahoe.] Moving from a Corolla at 30mpg (3.3gpm) to a Prius at 45mpg (2.2gpm) nets a less impressive savings of 111 gallons. When given the mpg numbers, tested individuals thought that the latter saved more gas than the former. (Our B&B knew this intuitively all along, of course.) Europe at al are already on liters/kilometers, so why not make the switch?

Get the latest TTAC e-Newsletter!

50 Comments on “Science Magazine Calls Miles Per Gallon An “Illusion”...”


  • avatar
    Paul Niedermeyer

    (Our B&B knew this intuitively all along, of course.)

    Intuitively, or otherwise; but yes, in a number of ways (metric system, health care) the Europeans are prone toward more pragmatic analysis of reality.

  • avatar
    Pch101

    The Science dudes apparently got sucked into the marginal analysis trap and haven’t dug their way out of it.

    In this example, the Prius still uses less fuel than the Corolla, Traverse or Tahoe. If you want to have the lowest total fuel expenditure, you would pay less for fuel by driving the Prius than you would with the other alternatives.

    Marginal analysis is useful, but we spend dollars in absolute amounts, not in percentages. It’s not the amount of the savings that matters, it’s the actual bottom line expense.

  • avatar
    jolo

    Shouldn’t it be gallons per 100miles? 15mpg does not equate to 6.7 gallons per mile.

  • avatar
    seoultrain

    mpg and gpm are just reciprocals. you take the number of miles you drive and either multiply by gpm or divide by mpg to get the number of gallons. No difference, except it’s cool to see yourself getting 100+ mpg’s while coasting. Feels a lot better than 0 gpm.

  • avatar
    Jonny Lieberman

    What if you move from a Tahoe to a Prius?

  • avatar
    carlisimo

    Honda proposed switching to gpm too, since it’d make them look better. And I agree.

  • avatar
    Pch101

    mpg and gpm are just reciprocals. you take the number of miles you drive and either multiply by gpm or divide by mpg to get the number of gallons. No difference, except it’s cool to see yourself getting 100+ mpg’s while coasting.

    Exactly right. Mathematically, there really is no difference, just different denominators.

    If anything, mpg is more intuitive. When using liters per 100 km, it’s a bit like golf — a lower score is better.

  • avatar
    NBK-Boston

    The point made in the cited paper is true but somewhat trivial. The non-trivial aspect is simply the fact that the two researchers went out into the world, surveyed a bunch of people, and found (surprise surprise) that the average American Joe often doesn’t get it. Alternatively, they may have put the questions in such a way as to discourage the actual calculation of the true result (say, asking for one’s “instant impression”), in which case they’re only reporting a sort of psychological perception-bias, which can easily be overcome with a pencil and paper and some grade-school math, and perhaps is when people sit down to crunch the numbers on a potential major purchase.

    My objection actually relates to the way this blog post is phrased. MPG, as such, is not an “illusion,” it’s just that people have trouble instantly seeing how it measures the real world. Similarly, both MPG and GPM differences “reflect” real world savings equally well, mathematically speaking — again, it’s just that people relate to one better measure better than another. “Misleading,” or “unintentionally misleading” seems to be a better way of putting it, because it puts the emphasis on the gap of understanding in the typical reader.

  • avatar
    sean362880

    I think the point made by the researchers is excellent. It’s no surprise that the public doesn’t get it, because they’ve been fed MPG-based information since forever.

    Something will have to change though, when more and more alternative fuels become available. The confusion on E85 vs. E10 vs. regular gas on yesterday’s boondoggle post is evidence enough, never mind EV’s! And these were the so-called B&B talking!

    https://www.thetruthaboutcars.com/e85-boondoggle-of-the-day-blame-big-oil/

    Maybe estimated cost/mile, based on national average energy price?

  • avatar
    CarShark

    @Jonny:

    Then the rainforest self-generates, the hole in the ozone layer refills, the Amazonian spitting tree frog smiles for the first time in a millenium and every hairy-legged Birkenstock-wearer (male AND female) will want to sleep with you.

    OR

    You’ll lose less fuel while giving up some performance and utility.

  • avatar
    shaker

    Well, it’s nice that Science Daily just told dumb people that the Tahoe Hybrid over a standard Tahoe is a fantastic idea.
    GM used the same “math” to justify it to the public — but wait; nobody’s buying it.
    Maybe the unwashed masses are smarter than they look.

  • avatar
    Pch101

    The non-trivial aspect is simply the fact that the two researchers went out into the world, surveyed a bunch of people, and found (surprise surprise) that the average American Joe often doesn’t get it. Alternatively, they may have put the questions in such a way as to discourage the actual calculation of the true result (say, asking for one’s “instant impression”), in which case they’re only reporting a sort of psychological perception-bias, which can easily be overcome with a pencil and paper and some grade-school math, and perhaps is when people sit down to crunch the numbers on a potential major purchase.

    I understand their point, and you’ve provided a nice synopsis of it. But still, it is ultimately flawed to promote marginal analysis to an absolute.

    Let’s get away from cars to illustrate the point. You have a 375 pound woman and a 125 pound woman. Each of them goes on a diet and exercise routine and loses 10% of their body weight.

    Who lost the most weight? The 375 pound woman (37.5 vs. 12.5 pounds.)

    Whose diet was most effective in terms of weight loss? The 375 pound woman — she outperformed the other woman by a resounding 300%.

    Who would you rather date? That’s the answer that really counts. The ultimate benefit is derived from the absolute, not the marginal. In this case, the marginal analysis produces a deceptive result. The thin woman is still preferable to the behemoth, and the efficient car still uses less gas than the guzzler.

  • avatar
    50merc

    I’m with the “eggheads.” It would be good for the government to require disclosing gallons per hundred miles on window stickers. I believe our friends in Canada do that, and they have no trouble with the concept. (Though I think they use “kilometers,” which I hear is 1/10 of a cubic hectare.)

    The problem with mpg is that the average Joe thinks by swapping his 15mpg Tahoe for a 20mpg Traverse he’ll cut gas consumption by one-third, though it’s actually one-fourth. It’s a lot easier to understand the difference between, say, four gallons and five gallons per hundred miles.

  • avatar
    sean362880

    pch101 –

    The ultimate benefit is derived from the absolute, not the marginal. In this case, the marginal analysis produces a deceptive result. The thin woman is still preferable to the behemoth, and the efficient car still uses less gas than the guzzler.

    I disagree. From the public’s perspective, and the government’s in particular, the marginal gain is much more important. The benefits are reduced national dependence on oil, energy prices, and sociopolitical stability.

    Using your analogy, if the goal is to lose 3,000 pounds, it’s easier to start with 100 fat people than with 300 thin people. The goal comes faster with less effort on the part of fewer dieters.

  • avatar
    John R

    Now, I remember why I frequent this site… :)

    [+1 all of the above commenters]

  • avatar
    Pch101

    Using your analogy, if the goal is to lose 3,000 pounds, it’s easier to start with 100 fat people than with 300 thin people. The goal comes faster with less effort on the part of fewer dieters.

    That’s the whole problem with using marginal analysis. It measures change, but ignores totals. Which is another way of saying that it measures the increment, but ignores the result.

    This is exactly the gimmick you find in retail stores everywhere. “You saved 10%!” “Save 20%!” But if I save 10% or 20% on a ridiculously high price, I am still paying too much.

    At the end of the day, your employer pays you in dollars, not percentages. You can’t spend a percentage anywhere.

  • avatar
    KixStart

    sean362880,

    But, to carry the analogy a bit further, it turns out that the fat people didn’t care whether or not they lost weight so, lacking any motivation, they wouldn’t join the program.

    And for those that decided they did care, they realized that joining the “lose weight while still eating fast food” program turned out to be very costly… as much as 10,000 double-cheeseburgers at McD’s, perhaps even more. Just to eat fewer double cheeseburgers overall. It was senseless. The motivated ones just went off to eat carrot sticks and saved a lot of money that way.

  • avatar
    NBK-Boston

    Pch101

    It is fair enough to point to the total cost of ownership, and not just marginal savings, when discussing the costs of motoring generally.

    But, to return to your analogy, while one might rather date the 125 lb. woman (either before or after her diet), one still might pick the 375 lb. woman (probably after some diet and exercise) if one needed a linebacker. This gets us back to the old point of people’s “wants” versus “needs,” and how these things vary based on circumstances and occupation, and how much people can and should spend to satisfy the latter and seek to satisfy of the former, and how much “we” should try to influence our fellow citizens in how they judge these matter.

    The fueleconomy.gov website has a “yearly cost of fuel” figure for each tested car model. It is calculated on the basis of the number of city and highway miles an American motorist drives, divided by the EPA mpg figures for the car, and then multiplied by the average cost of the type of fuel the car uses. The base Tahoe costs $3825 per year, the hybrid Tahoe costs $2913, the automatic Corolla $2111, the Prius $1328. This illustrates both the relative and absolute differences between the various cars, and is essentially equivalent to a GPM measure.

  • avatar

    I have said this before…GM has it right by putting the “hybrid enabled” MPG increase into lower MPG vehicles and not into its smallest vehicles.
    By my calculations (yes, I actually did these calculations, but it was a few years ago), upping the efficiency of the lowest mileage vehicles by 20% will save in the ballpark of about twice the gasoline as doing the same to the most efficient vehicles. (I believe I based this on something around a 15mpg boost to 18 and a 35 mpg boost to 42.)

    Don’t forget: The light-duty trucks/vans that get poor mileage are also the ones that HAVE to be driven a certain mileage. These support out carpenters, plumbers, and others who keep society running. They represent a reliable set of gasoline use, as opposed to personal vehicles, which are likely to represent discretionary mileage.

  • avatar
    Pch101

    But, to return to your analogy, while one might rather date the 125 lb. woman (either before or after her diet), one still might pick the 375 lb. woman (probably after some diet and exercise) if one needed a linebacker.

    While I do believe in equality, I will default to chauvinism in both dating and when staffing a football team…

    I understand the point, of course. I don’t have a problem with using marginal analysis, per se, it has its benefits. But business school people tend to rely excessively on it, and it often plays itself out in negative ways.

    The conflict arises when developing and measuring goals. If you heavily weight margins, you will tend to underperform, because you are focusing on increments instead of results.

    A study like this has business school PhD written all over it, and sure enough, you can see that’s who prepared it. Kinda figures that a business school study could find a way to argue that 20 mpg is better than 40 mpg.

  • avatar
    Luther

    “What if you move from a Tahoe to a Prius?”

    The Jungles will all die from lack of CO2 and water…The earth becomes one big desert…The spitting tree frog gets dry-heaves and every hairy-legged Birkenstock-wearer (male AND female) will want to sleep with you.

  • avatar

    Yes, so a 5mpg difference becomes a proportional less big of a deal when you go up in overall mpg efficiency. However, this was already shown when the EPA revised its testing program and massive SUVs and trucks showed little change, while way more efficient cars were hit a lot (like the Prius, even if most of the demographic that owns them wouldn’t drive like the EPA cycle) when using the same testing cycle.

    However, efficiency improvements are just that: improvements. It’s a gain, simple as that. I would prefer the EU methodology simply because it makes more sense in general, even if the testing cycles themselves are totally different between the nations.

  • avatar
    hwyhobo

    Frankly, I would rather just switch to the metric system, get it over with, and avoid the whole discussion.

    US Military is already on the metric system. Yes, I know, costs and all, but we will do it one day anyway, let’s shorten the pain.

  • avatar
    paoleary

    The study is basically one of heuristics and biases, which goes back to Tversky & Kahneman’s work which started in the early 70’s (the authors cite Kahneman and Frederick, but it’s the same content.) What I don’t get is why they only appeared to study strategy choice for the MPG case. The GP(H)M study is only done comparatively, so they only check outcomes.

    However, if the GP(H)M subjects are using the same faulty linear reasoning as the MPG subjects, then you get the problem opposite the one they cite for MPG: small improvements in inefficient cars will be overvalued. You can test this yourself by applying the linear reasoning to a 1/2 vs. 1/3 GPM car, which is the same marginal improvement as going from a 1/4 to a 1/6 GPM car. While both sets represent a 50% improvement in fuel economy, the first change is valued at double the second.

    The only way to make the linear thinking accurate in terms of marginal changes is to express fuel mileage on a logarithmic scale (I propose “decibelairs“). However, by doing this you lose the ability to quickly calculate the financial or environmental impact of a fuel efficiency change.

    What we’re left with is, as the authors indicate, a policy question, since the psychology (presumably, since they didn’t actually test!) goes both ways. The authors effectively conclude that it is better to overestimate changes in inefficient cars than to overestimate changes in efficient cars.

    Conversely, I conclude that people need to learn to perform division (3/2 = 6/4 => same marginal gain).

    Personally, I feel that scaling tricks like “per 100 mi” and “per 100 km” are more confusing than not, since the majority of travel is over much shorter distances. If people already show an unwillingness to perform division, no one is going to think about how much that 2 mi trip to the grocery store costs (about $1.20, I think, using the ~6 GPHM car), especially when, due to the use of air conditioning and the influence of cold oil, among other things, they will not be meeting peak efficiency along their short route. This is just as likely to cause people to underestimate their fuel usage with their current vehicles as the current MPG-based system.

    In other words, this little notational issue is possibly less significant in causing poor estimation of fuel use than the nature of the EPA test itself and its relative conformance to one’s particular driving habits.

  • avatar
    Jonathon

    Whatever your feelings on mpg versus gpm, there are always worse ways to measure gas mileage or fuel usage, like dollars to fill up your tank or miles traveled on a tank. I’ve really never understood why so many people go by those measures.

  • avatar
    dkulmacz

    PCH101 . . .
    I’m not quite sure what argument you are actually trying to make. I don’t think this study is aimed at the effects of a single decision, but at the aggregate results of many decisions.

    It’s obvious that if Driver A trades a Corolla for a Prius, and Driver B trades a Tahoe for a Traverse, then Driver A is absolutely better off because he has the lowest absolute fuel bill. That’s not the point.

    Suppose Driver C owns both a Corolla and a Tahoe, and he only has enough money to trade one vehicle. In which case is he better off? Using the egghead analysis, we can see that he’s better off trading the Tahoe for a Traverse and keeping the Corolla, since that is the path to the absolute lowest fuel consumption. This, I believe, is the point of the study. Most ‘men on the street’ would tend to put their money towards the sexy, uber-mileage Prius thinking they were getting the best answer . . . but would be wrong.

    (And please no one pipe up with any ‘trade em both for a Prius’ suggestions, since that simply evades the question . . . some people need Tahoes/Traverses, and some need Corollas/Priuses.)

    Extend this logic to the US vehicle fleet in aggregate. If we have limited resources to apply towards improving FE, then this study shows that we can get the most bang for the buck by spending it on improving the worst offenders. Thus, the marginal analysis at the individual level — along with the reality of limited resources — produces the best absolute answer in aggregate.

    I believe this concept is also known as the ‘law of diminishing returns’.

    Now, combine this with the fact that it is usually easier to find ways to improve the gross offenders than it is to squeeze more performance out of your already best-in-class. Together these make a very compelling case for the GM strategy of focusing FE efforts on the gashogs.

    However, this is not the politically correct answer (and not coming from the political darling), so it is easy to slag on it. This reminds me very much of the battle to reduce non-CO2 tailpipe emissions (boy, the good old days) . . . California found it easy to mandate ever lower and lower limits for new cars that were harder and more expensive to meet, all while neglecting the fact that removing a few thousand blue-smoke-spewing hoopties would have a greater overall impact on crud flung into the air. However, blue-smoke-spewers tend to be owned by the poor and underpriviledged, and it’s not politically correct to force them to give up their transportation. Hence, the PC but suboptimal solution.

    Another note . . . I think it’s a bit funny to see the back-patting here, saying that the commenters at TTAC have known this all along. I don’t think so. I’ve seen plenty of discussions about CAFE here, and the CAFE calculation deals with basically the same issue (the ‘average’ in the name is a geometric mean, not the normally understood arithmetic mean). The majority of the commentors do not understand a calculation as simple as figuring the ‘CAFE’ of two individual vehicles. I don’t think this was any more obvious to the readers here than the general public . . .

  • avatar
    improvement_needed

    several comments:
    1) this got published in science????!!!!

    2) car makers already put an ‘estimated fuel cost / year’ on their window stickers… – this scales the same as g/100miles…

    3) if you really want people to consider things, you should calculate your TRUE cost to own… (ala – edmunds)… – imagine if that was put on window stickers???
    from edmunds (base models)
    tahoe: 78 cents/mile
    oddesy: 56 cents/mile
    enclave: 75 cents/mile
    sienna: 59 cents/mile
    caravan: 56 cents/mile

    all automatics
    civic (EX): 46 cents/mile
    civic hybrid: 46 cents/mile
    corolla (XLE):52 cents/mile
    prius (base): 54 cents/mile

    unfortunately, when doing this, i calculated that it costs me ~ 50 cents/mile to drive my current vehicle – thus – when making a trip, the question needs to be asked: do i need to drive 60 miles to go somewhere? – that trip will cost me 60$… (round trip)…
    of course, by owning a vehicle, your sunk costs are typically much greater than your variable costs…

    edit: who would buy a chevy traverse????
    much better off in a minivan – ridiculously cheaper to own / operate

  • avatar
    Lumbergh21

    All it tells me is something that I already knew, always run the numbers, don’t guess. We need to change the rating system to gallons per 100 miles just to protect dummies or is it to be more like the wonderful Europeans? I know why don’t we switch to carbon ratings!

  • avatar
    Bunter1

    And once again a comparison makes the incorrect assumption that the Prius is a compact.
    If the Travacadoutclave is a Tahoe equivalent while being smaller, lighter and pulling less based on interior volume then lets compare the prius to a midsize.

    Haven’t time to run the numbers but it might just save more fuel then the ‘Hoe vs. Traverse.

    Not saying that most ‘Hoe drivers won’t be better served by the Traverse, they will.
    Shoot most will do even better with the “horrors” Minivan (improvement_needed, dead on).

    Worth a thought.

    Chuckles.

    Bunter

  • avatar
    Kevin

    Depends what your question is. If you are wondering which switch gives you more miles per gallon, than intuition is absolutely correct that the Prius upgrade gives more additional miles than the Traverse upgrade.

    But the supposed point of the article is pretty pointless. When does anyone in the real world need to analyze whether switching from a Tahoe to Traverse will save more gallons than switching from a Corolla to a Prius?

    Never, that’s when. You’re either looking to trade in your Tahoe or you’re looking to trade in your Corolla. No surprise people haven’t developed intuition about a tricky arithmetic problem they NEVER face in real life.

    If you’re some economist or analyst doing a research study, then you are probably analytical enough to think through the problem. No one else would need to bother.

    If you’re looking to improve fuel economy and you live in a MPG regime like we do, you pick the car with the highest MPG. Period. End of story. That’s simple. Nothin’ wrong with that.

  • avatar
    Bunter1

    Haven’t seen a Lambda come close to 20 mpg in a road test. Make the comparison to the ‘Hoe 16mpg vs. 13mpg. Prius vs. midsize 4, 45mpg vs. 25mpg.
    Probably closer to the reality.

    Bunter

  • avatar
    yankinwaoz

    Both are wrong.

    There are two problems we are now facing, and will be dealing with more and more in the future.
    (1) Multiple energy source vehicles.
    (2) Fair road use taxes.

    Multiple energy source vehicles:
    Automobiles are going away from a single energy source (gasoline) model. There are now, and will be even more, multiple energy sources.

    Using distance/gallon or gallons/distance doesn’t matter anymore because it is only measuring one energy source. If you have a plug-in hybrid, how do you compute the mileage?

    A new inclusive system needs to be used. Cars need to be measured as distance traveled for a given unit of energy, lets say a kilowatt. That measure needs to be an average of the energy fed from OUTSIDE the vehicle. Thus, energy recapture systems won’t skew the average.

    That kilowatt can be derived from gasoline, diesel, fast food grease, ethanol, natural gas, electricity, compressed air, or other sources of energy.

    The mix of fuel sources varies. But for the most part, a car will have at most 2, perhaps 3, sources of energy. The mix ratio can be measured based on averages from city or highway driving.

    So the buyer can have the kilowatts by energy source for two driving conditions. But at the end of the day, what people want is a rough guess of how much a vehicle will cost to operate. So how do you covert these numbers to something that is not impacted by the price of an energy source?

    You can’t. In the old days it was easy. There was only one variable because there was only one fuel source. However, it doesn’t have to be that bad. With the Internet almost everywhere, and small computers everywhere (cell phones for example), a buyer could enter the base kilowatt numbers into a website or application and ask it to derive a cost based on fuel prices that day.

    An online operational expense calculator can be run by a government website. In fact, a good web site can have all the ratios and fuel types on file. Just enter a make/model and let it look up the ugly details and do the math. Then you can compare car A to car B and now how much it will costs you to operate them.

    Fair Road Use Tax:
    The idea of MPG also directly relates to how road uses taxes are assessed. The current system, collect at the fuel pump, is becoming more and more unfair. What if your car gets 50 percent of it’s operational Kw from your home electrical outlet? That electricity is not taxed with road use taxes.

    It is also unfair that one car of equal road wear-n-tear uses half the taxed gasoline of another car.

    The bottom line is that cars will need to be taxed annually (or so) based on a sum of actual energy used plus miles driven. The energy consumed by the motor can be tracked, no matter what source of energy the motor draws on. The miles are tracked via odometer, as normal.

    By collecting these two variables, a fair tax can be computed. This way, two cars of equal road wear-n-tear will pay the same. Cars that carry more weight, or drive more, will pay their fair share of road use taxes.

    By thinking about cars in terms in kilowatts instead of gallons, we can solve a lot of complicated problems that we will be facing in the future.

  • avatar
    ret

    @dkulmacz

    Hooray!!! Someone got it right!!!

    The Law of Diminishing Marginal Returns dictates that you will choose that which improves your utility the most first, then on to the next, the the next until you reach a point where there is either no improvement or an actual worsening.

    In this instance it means the Tahoe -> Traverse trade is the most economically rational choice of the given options.

  • avatar
    ttacgreg

    just a few random comments . . . .

    The level of discussion (with some exceptions) on this site is far and away the most rational and enlightened of any commentary site I visit , BY FAR! What a pleasure.

    Intersting to think othat my recent careful driving of my corolla results in its burning 2.24 gallons per 100 miles.

    Here is another good write up of this concept.
    http://www.caranddriver.com/features/columns/c_d_staff/patrick_bedard/mileage_no_it_s_your_gallonage_that_really_counts_column

  • avatar
    SkiD666

    Bunter – Compare the ’08 Prius and ’09 Corolla and you will find very comparable dimensions and volumes. Prius has a little more interior volume because of the hatchback design, but they are very comparable. If the comparison was between a Prius and Yaris, that would be apples to oranges.

  • avatar
    kph

    Good comments, I have another idea to throw in… how about gallons per 1000 miles? Yes, I know it’s just moving the decimal point over, but the truth is keeping track of where the decimal point goes confuses the hell out of a lot of people when they multiply it with price of gas per gallon.

    By doing this, you can round to two significant digits, similar to that of typical mpg figures. 1000 miles is also close to the average number of miles a car is driven in a month, which makes monthly budgeting a bit easier.

  • avatar
    Pch101

    It’s obvious that if Driver A trades a Corolla for a Prius, and Driver B trades a Tahoe for a Traverse, then Driver A is absolutely better off because he has the lowest absolute fuel bill. That’s not the point.

    Actually, that is the point.

    What should happen in this scenario is that Driver B should assess the big picture, develop a holistic goal and figure out how to meet that. That may mean punting on the Tahoe and the Traverse entirely, or finding a way to fundamentally change his manner or level of consumption so that he uses less of either.

    When abused, marginal analysis tends to be destructive because it actively encourages small picture, short-term thinking just like this. It’s that very mentality that allows someone to say that 40 is less than 20 with a straight face.

    Mathematically, I understand the point being made and the authors are accurate in the details, but in the big picture, the analysis falls flat because it fails to look at the aggregate.

    It’s a bit like the allegory of the blind man and the elephant — these guys have stumbled upon the trunk, and they’re now telling you how to extract more water out of the hose that they’ve just found. By missing the big picture, the wrong little picture gets drawn.

  • avatar
    Blastman

    Here in Canada we’ve been using liters/100 km as a fuel economy rating instead of mpg for over 20 years. I still find myself calculating out the mpg and using those numbers to compare cars.

    MPG just seems like an easier way of comparing the efficiency of vehicles. Even though, either way will work. Using a figure like 100 km always seem just like some arbitrary number to me … oh wait …. gallon is an arbitrary number too …. Fooey,… I still prefer mpg.

    Frankly, I would rather just switch to the metric system, get it over with, and avoid the whole discussion.

    You could switch your units of gas over from gallons to metric quite easily. A liter is 1.0566 US quarts, so you could call it the metric-quart. And 4 metric-quarts could = 1 metric-gallon.

    A metric-gallon (MG) would be 5.6% bigger than your current gallon … MPMG…!!!

  • avatar
    bleach

    The need to switch is not nearly as great as the researchers think. It’s particularly useful for comparing two incremental changes but that scenario is just not that common for a consumer. In the above example of Driver A going from Tahoe to Traverse and Driver B going from Corolla to Prius, if I’m Driver A, I don’t give a hoot how my change compares to Driver B. My decision is going to be Tahoe to Transverse or Prius or something else. In that case MPG is sufficient because the Tahoe is the basis for all comparisons.

    So yeah if you’re trying to figure which of your cars to sell to get higher efficiency, keep gpm in mind, but I bet which one is older or needs repairs, life style changes or equity in the vehicle all take priority.

  • avatar
    willbodine

    There are 2 main ways to measure fuel consumption. All freeway. And urban/suburban. When comparing one vehicle to another, the freeway figure is the fairest way of measuring this. Mixed urban/suburban is just too variable to be meaningful. However, I find it somewhat amazing that the cars I regularly drive (and the ones before these) all have just about the same ratio between freeway and mixed. The mixed urban figure is almost always 2/3s of the freeway useage. My Tacoma V6 gets 21 mpg freeway, 14 around town. My Lexus IS gets 30 mpg freeway, 20 around town. Both my 94 Cad Brougham and 94 Buick Roadie Estate got about the same as the Taco; 21 fwy, 13 town. Weird, huh?

  • avatar
    NBK-Boston

    dkulmacz

    If you assume that Tahoe drivers will just continue to drive Tahoe-sized vehicles, and Corolla drivers will just continue to drive Corolla-sized vehicles, then yes, starting a hybridization program with the biggest gas hogs is the most efficient way to go — also assuming that the per-car cost of hybridization does not vary much based on the size of the vehicle. So yes, under those circumstances, GM made the logical choice.

    But there are two assumptions there, and it seems as if both turn out to be untrue in the real world. As for the first, plenty of Tahoe drivers don’t seem to value their Tahoes all that much, as people are shifting in droves to smaller cars due to additional fuel costs of around ~$2000 per year (or less) compared to two years ago. So the question becomes, why bother hybridizing Tahoes, when an even easier solution is to scare three quarters of your Tahoe drivers into buying smaller cars? This is what Pch101 seems to be getting at in his posts, and what Europe has been doing for years with high gas taxes.

    The second assumption, that the hybrid premium for large vehicles is not much different than the hybrid premium for small vehicles, also seems to be untrue. This is just based on a casual comparison of MSRP differences between cars like hybrid/regular Civics, Prius vs. optioned-out Corollas and Camrys, regular vs. hybrid Tahoes, etc., accounting for various option packages.

    Given all that, it’s not hard to understand what’s been going on. The gas-hog drivers who felt they had a choice are now fleeing the gas-hogs, mostly for ordinary, non-hybrid compact or midsize cars. The seriously green crowd has been piling onto the super-efficiency bandwagon, as a fashion statement as much as anything else. The “sensible shoes” crowd that actually cares to live modestly or frugally still buys used Corollas and drives them into the dust, 150,000 more miles down the line, and has barely missed a beat. Political candidates who need to drive green and drive domestic at the same time will buy Escape hybrids. Eventually, the few Tahoe-drivers who actually still value the utility of such a vehicle will inquire into hybrid options, and buy that option only if it saves more than it costs.

  • avatar
    Wheatridger

    I propose the clearest measure possible, one that directly expresses fuel expenses — Cost per Mile. Simply divide the current cost of your recommended fuel type & grade, and divide that by your average MPG. (For my TDI, with D2 at 4.70/gal., that would be $4.70/42mpg= $0.11 CPM. For comparison, my Subaru Forester’s figure is $0.15 CPM.

    It’s a clear way to compare any set of vehicles at a given moment. This CPM would vary from week to week as fuel prices change, of course, but this would affect all cars equally.

    It’s much more difficult to predict payoff points for fuel-saving improvements. As fuel prices rise, so does the value of every gallon saved.

  • avatar
    Pch101

    So the question becomes, why bother hybridizing Tahoes, when an even easier solution is to scare three quarters of your Tahoe drivers into buying smaller cars? This is what Pch101 seems to be getting at in his posts

    You’re close, thanks for going through that.

    The subtext to all this is that I see a study like this as being fodder for gross misinterpretation and abuse, particularly by the supporters of gas guzzlers who will surely post this story all over the internet as “proof” that getting 20 mpg is better than getting 40 mpg.

    Based upon other news accounts, the point of their study was that gas guzzlers from the road are the low hanging fruit in any program to lowering gas consumption. Their work, at least as far as I can tell, was not intended to encourage crossover sales.

    Absolutes still count for more than marginals, because they get to the bottom line and force a view of the big picture. It’s that same abuse of marginal analysis that makes a Civic a better car than a Corolla. Honda benchmarked the absolute (reviewed the entire market, took the best of the market, and benchmarked them in order to beat them), while GM benchmarked itself (looked at its own last effort, then added a few percentage points of improvement.) When the focus is on the marginal, the tendency is to not do much.

  • avatar
    NBK-Boston

    To be fair to dkulmacz, his “driver C” scenario really seems to be what inspired the study. That is, the family of one of the researchers had an old minivan/SUV and an old midsize sedan, had only enough money to trade in one of them, and “needed” at least one three-row vehicle in the fleet.

    Do you trade the old Accord for a Prius, or do you trade in the Tahoe for a CUV? Answer: just eyeballing the MPG figures can be misleading, and you’re better off doing the math — calculating the gallons used per 10,000 miles driven in each choice, to see if trading in the sedan or people mover would save the most gas.

  • avatar
    Gardiner Westbound

    Canada imposed metric measure on a kicking, screaming public 40-years ago. The government believed people familiar with miles per gallon would be comfortable with kilometers per liter, and adopted that standard.

    The auto manufacturers went ballistic. They had a vested in an obscure fuel consumption convention; one that muddied information, maximized confusion and inspired disinterest. They successfully lobbied for the convention to be changed to liters per 100 kilometers.

    It exceeded their expectations! So few understand metric people stopped thinking and talking about mileage. I doubt one in 100,000 knows his car’s fuel consumption.

  • avatar
    Pch101

    Do you trade the old Accord for a Prius, or do you trade in the Tahoe for a CUV? Answer: just eyeballing the MPG figures can be misleading, and you’re better off doing the math — calculating the gallons used per 10,000 miles driven in each choice, to see if trading in the sedan or people mover would save the most gas.

    Fair point, although I still think that it is prone to abuse.

    For those who prioritize fuel economy, the formula is straightforward — drive the most fuel efficient vehicle possible. If the goal is to save money, rather than fuel per se, then do a cost-benefit analysis to see whether paying for a more efficient vehicle is worth whatever increase in expense may come from the change over.

    To the extent that a larger vehicle is needed, use the larger vehicle as little and as efficiently as possible, and to defer to using the more efficient vehicle or another alternative as much as you can.

    (In the previous post, I meant to refer to the Cobalt, not the Corolla, so sorry for any misunderstanding.)

  • avatar
    gsp

    the L/100km we use here in Canada is stupid. KM/L would have been better but i still think in MPG and i was born in the 70’s.

  • avatar
    RogerB34

    The purpose is to get the Peasants to think in terms of weight of tail pipe carbon per mile as the EU’s do.
    With realization of the evil they are inflicting on the environment, then carbon tax gasoline at the pump. UK has such a proposal pending. A government card would automatically record weight of gasoline to be taxed.

  • avatar
    John Horner

    The researchers got it right, but the US will never change. I still remember that we got 20% of the way along to metric conversion and then said … nah, forget it.

    In many ways the US is a very backward place. Certainly our population on average shows poor math and language skills.

Read all comments

Back to TopLeave a Reply

You must be logged in to post a comment.

Recent Comments

  • fazalmajid: The car makers need to provide more transparency because they’ve been lobbying the Federal government to...
  • dal20402: Yep, in the ’21+ the outer seat portions are exactly the same.
  • JonBoy470: It is of course illegal to operate non-compliant vehicles on public roads, but the law, as written, also...
  • JonBoy470: These speed shops are operating without even the merest whiff of plausible deniability that parts vendors...
  • Carlson Fan: “Start saving, spring 2024 will probably be your first real opportunity.” Yep I’d say...

New Car Research

Get a Free Dealer Quote

Who We Are

  • Adam Tonge
  • Bozi Tatarevic
  • Corey Lewis
  • Mark Baruth
  • Ronnie Schreiber