Why 'gallons per mile' is better than 'miles per gallon'

http://datagenetics.com/blog/april12014/index.html

by squeakynick • 12 years ago

115 121 12 years ago

datagenetics.com

anigbrowl• 12 years ago

I'll settle for reciprocal distortion if more people would switch to the metric system.

dredmorbius7• 12 years ago

Most metric countries report liters/100km, solving both problems at the same time.

ProblemFactory• 12 years ago

An useless but fun observation: you can do the litres/100km division and arrive at units of area. A family car might have a fuel consumption of 0.05 mm^2.

This is the cross-section of the imaginary pipe of fuel along the highway that a car with no fuel tank would have to pick up as it travels.

dnautics• 12 years ago

Fuel consumption is a better measure of a vehicles performance because it is a linear relationship with the fuel used, as opposed to fuel economy which has an inherent reciprocal distortion.

Fuel economy is a better measure of travel potential because it is a linear relationship with the distance you can go, as opposed to fuel consumption, which has an inherent reciprocal distortion.

RaptorJ• 12 years ago

An interesting switch, but who cares? There are gas stations everywhere.

username2232• 12 years ago

Everywhere? Try driving across Nevada or southern Utah.

analog31• 12 years ago

I think there are other examples of this, such as price-to-earnings ratio, and focal length of lenses. In those cases, since a denominator can go through zero, the ratio goes through a big discontinuity. And I've seen p/e ratios reported as "n/a" when they should really be represented by a big negative number.

sillysaurus3• 12 years ago

Also time-per-frame instead of frames-per-second.

deletes• 12 years ago

Frames per seconds is better in almost every way. When you use it to watch movies it is better since a normal human understands seconds but not what a frame is. In gaming, frame times usually vary quite a lot, and the average, which is fps, is the important part.

ygra• 12 years ago

When doing real-time graphics and looking at a profiler, it does help to know that you want to have less than 16 ms of calculations per frame. But agreed, in most other respects fps is nicer.

werdnapk• 12 years ago

Canada pretty much uses X litres/100km nowadays recently replacing mpg. The lower the litres, the more efficient the vehicle. It actually was fairly easy to become accustomed to these new metrics.

burke• 12 years ago

It's an extremely convenient metric.

L/100km is roughly equivalent to L/h at highway speeds, and you can figure Litres to destination when you see a road sign by multiplying L/100km by the hundreds place.

With gas stations being upwards of a hundred kilometres apart in the prairies and northern ontario, especially at night, it's nice that this is so easy.

blahedo• 12 years ago

Ah, good point: so that means we should be using Gallons per 60 miles! :)

dredmorbius7• 12 years ago

MPG is useful for range estimates: if you've got n gallons, you can travel m miles.

GPM is useful for budget estimates: if you have a trip of m miles, you'll need n gallons of fuel, with some given cost.

Or you can use a tool such as GNU units which handles reciprocal conversions with ease and aplomb.

gamegoblin• 12 years ago

Never used GNU units, but Wolfram Alpha will also do this trivially [0].

[0] http://www.wolframalpha.com/input/?i=16+mpg+in+barrels%2Ffur...

return03• 12 years ago

and google

cbr• 12 years ago

As will google: https://www.google.com/search?q=16+mpg+in+barrels%2Ffurlong

dredmorbius7• 12 years ago

GNU units works offline ;-)

astrodust• 12 years ago

I don't know how a/b is somehow magically easier than b/a. They're both involving division.

Most cars outside of the US report L/100km.

For example, if you have to go 250km and you're running at 8L/100km then you need 2.5x8 = 20L. If your tank holds 50L, then you'll need nearly a half tank.

If you want to know how far you can get on a half tank, say 25L, then divide that by 8 to get roughly 6, or 600km.

bigbugbag4• 12 years ago

AFAIK, MPG is not used anywhere in the world but in one country where people don't even grasp the beginning of it.

MPG is actually quite useless to do range estimates. If you got n gallons you could possibly maybe travel more or less m miles depending on a variety of factors unrelated to the vehicle fuel efficiency such as speed, terrain, open or closed windows, driving with A/C on, and much more. There are a few mythbusters episodes dedicated to the myths of fuel efficiency/consumption.

GPM doesn't seem that useful for budget estimates either as this depends on the fuel price, well maybe it is in the US where oil price is maintained artificially low but in other parts of the world you have a hard time guessing the price of oil 6 months from now or at the next gas station. I've seen price fluctuation of 0.8$ per gallon from a week to the next and 1.5$ per gallon from a gas station to the next in the same city. GPM is only useful to compare vehicles between them, which is something people in position of power in the USA do not want.

dredmorbius7• 12 years ago

If you got n gallons you could possibly maybe travel more or less m miles

I track my mileage in a hardcopy log: miles driven, fuel consumed, and mileage.

While I see some variation (about 35% difference between worst urban and best highway driving), I've got a very solid idea of how much range I can get on a tank under various driving conditions. Sufficient to make "stop or continue" decisions when passing fuel stops on long road trips (though sometimes cutting that closer than I'd prefer).

kalleboo2• 12 years ago

And range estimates are usually only needed when you're sitting in your car driving ("do I need to stop at this gas station or can I go on to the next one"), and every car I've ridden in made in the past decade has a remaining range display.

Budget estimates and car fuel consumption comparisons OTOH are needed when you're either buying a car, or at home planning a trip ("is it cheaper to drive or fly?")

dredmorbius7• 12 years ago

every car I've ridden in made in the past decade

You're in more up-market cars than I, apparently.

taeric10• 12 years ago

Citations, or you are just promoting your opinion.

I'm also curious about the numbers showing that 55mph is the most efficient. I do not particularly dispute that this is the case for the majority of vehicles. I am curious as to whether or not this is as it has to be, or because that is the way vehicles are built in the US. That is, could you do better with different high speed gearing?

squeakynick• 12 years ago

There's a wealth of information here http://www.fueleconomy.gov/

exue8• 12 years ago

It's not always the most efficient for every car, but there are lots of charts showing this tradeoff. Some cars are more efficient at 65 or 70 than 55 (97 Celica), but in modern gas cars, 40-50mph is where you hit a peak, and you start getting worse from there. 40mph on the highway is painfully slow though, so people generally pick an optimal point higher than that.

(One inconsistency I've found is how the Motor Trend test below sees a very high peak at about 40MPH and a decline after, while other sources show a flatter peak). And 55MPH seems more like a magic number the author picked, but not a bad one if we don't know the model.

Generally cars are geared so the engine begins entering its optimal / most efficient RPM range where it can produce more torque in top gear around 40-50mph, and rolling resistance and wind drag (which is a cube function of speed) contribute to make mileage worse after that. Usually the speed at which you first shift into top gear and cruise comfortably is close to optimal. Efficiency doesn't start dropping for a while because of the cubic nature of drag, and the engine sometimes is more efficient at higher RPMs for a bit (see nols' post on manufacturer optimization).

Fun fact: At top speed (254mph), a Bugatti Veyron will use its 26 gallon tank in 12 minutes.

An older chart:

http://upload.wikimedia.org/wikipedia/commons/a/a9/Fuel_econ...

A newer chart: http://blog.automatic.com/wp-content/uploads/2013/12/mpg-vs-...

Motor Trend chart: http://image.motortrend.com/f/roadtests/sedans/1208_40_mpg_c...

http://www.motortrend.com/roadtests/sedans/1208_40_mpg_compa...

(One thing that bothered me is how the Civics in the last two charts were so different)

taeric10• 12 years ago

As I stated in a sibling, this still boils down to "this is how current cars are built." I would imagine if you looked at the average vehicle built in, say, the 1940s, you would see a vastly different "optimal" number.

Which is all to say, do we expect this to stagnate forever at 55? Is this a hard physical limit? I understand drag gets higher there. Are there no tricks left to us?

And again, I fully concede that this is likely not the most pressing fact around. Just one I am curious on.

userbinator• 12 years ago

When taking into account wind resistance, the optimal speed for efficiency varies greatly with the current weather conditions too: with a tailwind, the optimal speed goes up, while the opposite is true with a headwind.

Usually the speed at which you first shift into top gear and cruise comfortably is close to optimal

Although automatic transmissions and lack of tachometers are probably responsible for this, I've noticed that a lot of drivers remaining at the upper RPM range for a certain gear while cruising, when they could've sped up 2-3MPH and upshifted. I don't know if there's a term for this, but it's certainly recognisable as a passenger: the engine is much louder than it should be, often with accompanied higher levels of vibration and discomfort.

nols2• 12 years ago

Generally the most efficient is highest gear-lowest RPM, and with car makers that's usually around 55-60. You could change the fuel efficiency to different speeds(I'm sure they're adapted to local markets with metric vs Imperial), but it would be worse off for most people to do that. It's not simply a gearing issue to change fuel efficiency, it's also the ECU controlling the fuel/air going into the engine. They aim for optimal fuel efficiency for normal driving, and that seems to be the sweet spot.

Contrary to what I've heard before, changing fuel efficiency to higher speeds wouldn't make things better. It's also not that simple because drag increases exponentially with speed.

taeric10• 12 years ago

This is still not any more satisfying than "because this is the way they are built."

Also, I should say that I realize this is a side show. One that I think should get less attention. No matter how much more efficient you make a car, I doubt I would ever be able to do better than local transit. (I, of course, welcome evidence otherwise.)

sliverstorm3• 12 years ago

Supposedly peak engine efficiency is at peak torque, but of course no car is designed to cruise at peak torque.

sliverstorm3• 12 years ago

55mph is a sweet spot where the engine is operating close to peak efficiency and air resistance hasn't yet exploded. There very much are cars that are most efficient at 65, but that is because they sacrifice efficiency elsewhere.

Peak efficiency is the intersection of many curves- gearing, rpm, aerodynamics, speed... Aerodynamics is possibly the biggest factor though, hence the common point of 55

trhway• 12 years ago

55mph is a sweet spot where the engine is operating close to peak efficiency and air resistance hasn't yet exploded.

i think 55 is just the speed of EPA "highway" efficiency tests

wodenokoto• 12 years ago

Citations, or you are just promoting your opinion.

This is an inherently flawed statement. If you follow the citations in a scientific paper at some point, the citations are going to end. A scientific article that consist only of citated information doesn't bring anything new to the table, it only summarises.

taeric10• 12 years ago

But this offered nothing other than some slight preference with a compelling argument that the differences in spending are more obvious. Yet, even in stating this, the conceit is held that "higher mpg is always better."

So, my citations desire would best be settled by any sort of empirical study showing that they actually affected buyers in the desired way.

baddox• 12 years ago

I think most people, when requesting a citation in this context, would also be fine with a description of an experiment (preferably a repeatable one) and a record of the empirical result of that experiment. That's the "base case" you're talking about in the otherwise infinite regression of citations.

pbreit• 12 years ago

Either I did not understand the post or it is totally uncompelling. MPG makes much more sense to me. Gallons per 100 miles (or whatever) sounds very clumsy.

sopooneo• 12 years ago

A typical person drives a fixed number of miles per month, rather than using a fixed number of gallons of gas. So one advantage of "Gallons / 100 miles" is that it requires one fewer steps of arithmetic to figure out how much gas you need every month.

lotharbot4• 12 years ago

On the other hand, a typical person driving between Denver and New Orleans knows they just put 14 gallons in the tank and would probably like to figure out how far until their next required fuel stop.

Different use cases mean different natural measurements.

MisterBastahrd• 12 years ago

A typical person knows how many gallons their gas tank can hold, and thus how far they can go without needing to refuel. For day to day driving, it doesn't really matter. For long road trips, it can be essential if you're on an extended barren interstate highway.

lttlrck2• 12 years ago

How many times do you need to make that calculation? I know how far I can go on one tank and how much it costs to fill - everything that follows is trivial arithmetic.

amluto• 12 years ago

I was hoping that an article about careful use of measurement units wouldn't contain a blatant mathematical error.

The article calculates, presumably correctly, that driving 70mph costs $3.66/hr more than driving 55mph in some reference car. It concludes that driving 70mph is worthwhile if getting to your destination an hour early is worth more than $3.66.

This is completely wrong. Driving 70mph for an hour gets you 15 miles farther than driving 55mph for an hour. Doing that costs you less than $3.66, since you're driving for less time. It also doesn't save you anywhere near an hour.

vlasev2• 12 years ago

I think that's as good of an example as any provided by the author as why the US should do away with the mpg measurement. If even the careful author can make a mistake like that, then think about the average person.

pash2• 12 years ago

This topic came up on HN a couple of years ago [0], and I posted a comment that was well received. So here it is again, edited for the present context and updated with my recent thoughts:

People do seem to misinterpret MPG ratings. A study published in Science in 2008 [1] found that participants consistently overvalued vehicles with high MPG ratings. They assigned values linear in MPG rather than linear in its inverse.

The study's authors told participants to "assume you drive 10,000 miles per year for work, and this total amount cannot be changed." The participants were then asked to come up with values for vehicles of varying fuel-efficiencies. That is just the sort of optimization problem people face when choosing which car to buy, and apparently a fuel-efficiency metric that puts the amount of fuel in the numerator makes the problem easier to solve because expenditure is proportional to the amount of fuel burned, at least when distance driven is taken as given.

But in reading many of the words expended on this topic over the last couple of years, my lasting impression is that this a lot of hullabaloo about the wrong problem. It's the lack of attention paid to the "miles" part of the equation that most needs fixing. If the goal is to reduce carbon emissions (or any of the other negative externalities of driving), then taking distance driven as fixed frames the problem in a way that obscures the real solution: we should be encouraging people to drive less.

Yes, reordering your daily life to drive fewer miles is more disruptive than simply buying a car that goes farther on a gallon of gas. And, granted, once you've chosen your style of life, minimizing the amount of gas you burn as you go about your daily routine is the thing to do (even if your optimization problem is a purely financial one). All the same, it's ludicrous to ignore the basic inefficiency of the suburban lifestyle that predominates in America while we wait for automotive engineers to come up with clever solutions to pricey gas and to carbon emissions that are twice as high per capita as in similarly wealthy countries.

Elon Musk isn't going to save us all by himself. Surely living closer to where you work, using mass transit, cycling, and walking more must be part of the solution as well. ... So maybe houses and apartments should come with a "miles per day" rating suggesting how far you'd travel getting to and from shops, restaurants, entertainment venues, and your place of work every day you live there. ...

0. https://news.ycombinator.com/item?id=4020885

1. http://www.sciencemag.org/content/320/5883/1593.summary (paywalled); http://nsmn1.uh.edu/dgraur/niv/theMPGIllusion.pdf [PDF]

jseliger• 12 years ago

Great comment. I'll add an observation:

we should be encouraging people to drive fewer miles

To do this we basically need denser neighborhoods, as Edward Glaeser points out in The Triumph of the City (http://www.amazon.com/Triumph-City-Greatest-Invention-Health... he points out, as does Matt Yglesias in The Rent is Too Damn High, that the big problems are with local zoning requirements, which by and large forbid density increases.

There are lots of local battles going on regarding density, and I agree that these are good things: "living closer to where you work, using mass transit, biking, and walking more," but they can all be encourage or discouraged by zoning. In most of America they're discouraged.

In the meantime better mileage is at least an improvement.

jimmaswell3• 12 years ago

A lot of people simply don't want to live in a crowded city in an apartment crowded on every side with other apartments. I'd prefer a location in a low-density suburban area (with a lot of forest around me) where I could get to the city in a few minutes.

exue8• 12 years ago

The thing is we have a lot a lot of leverage on the technological side. Choosing a behavioral and societal change vs. technological is always a trade off - but for this case some things make improving the technology much easier. The median age of a car in the US is about 11 years vs. 36 years for homes. Cars are individually replaced whereas reconfiguring homes that are on fixed land is a much more difficult task. If you're asking about an individual level I agree it's a choice, and you've covered those pain points say within a metro area. However if your family is in many South/West cities - say the Houston metro area or Las Vegas, distances are going to be pretty far. Amazingly, Los Angeles actually has pretty low average miles per year - one of the lowest for metro areas.

While the technological advancement is pretty much a universal win, the behavioral change isn't, or is at least debatable/not everyone's cup of tea - a lot of Americans prefer to live in large houses with lower density neighborhoods (of course others prefer large dense cities). Many also live in rural or semi-rural areas, and the US has a lot of cheap land compared to say Western Europe. Zoning in a city is dictated by those residents. However with better technology and cheaper energy, the expense of this lifestyle will be lowered and allow more choices (similar to how remote work allows people to remove geographical limitations). If we think about developing countries, any efficiency gains will also reduce their energy impact as consumption increases

api_or_ipa• 12 years ago

Your suggestion to rate houses with a "Miles per day" rating is brilliant. With some moderately rigourous standardization, I could see the concept becoming more commonplace in urban centres where private vehicle travel is simply unfeasible.

twic3• 12 years ago

It's an intriguing idea.

Just as we have mile-per-gallon ratings for city driving, highway, etc, we could have mile-per-day ratings for different lifestyles, based on how much you eat out, whether you have children, whether you cycle, etc.

Commuting distance might be tricky to incorporate, as there are a wide range of places you might work. If you live in Guildford, then if you work in Guildford, your commuting distance could be 3 miles, but if you work in central London, it could be 30 miles. Perhaps that falls into the classification of lifestyles as well. Perhaps we just exclude commuting as too variable.

If we knew how long it took to get to places, we could make this measure in the time domain rather than the distance domain - minutes per day. Say you work for Google in London. If you live opposite East Croydon station, your commute is 11.3 miles by road, and takes 16 minutes on the train. If you live opposite Wallington station, your commute is 11.1 miles by road, but takes 38 minutes on the train.

danbruc• 12 years ago

Next: Why liters per 100 kilometers is better than gallons per mile.

pash2• 12 years ago

See https://news.ycombinator.com/item?id=4020885 :P

notduncansmith• 12 years ago

Well played, sir.

Vik1ng2• 12 years ago

It's the only one that doesn't make my brain hurt xd

Every time I'm confronted with MPG I first have to think what is better and when I calculate I always confuse M/G and G/M.

While always having that 100km distance in mind makes it a lot easier. I'm not suddenly going to think about 100l.

thomasahle• 12 years ago

With most cars today driving more than 20km/L, it would perhaps make sense to change to L/1000km already.

ntaso• 12 years ago

On paper maybe, but in reality, most cars still consume between 5-10 liters per 100km and more, if you have the horse power.

leccine• 12 years ago

I guess this is why l / 100km is used in the EU instead of km / l.

Ecio78• 12 years ago

Actually in Italy I've always seen km per liter for years and years, let's say all the 80s and 90s ("my car runs 14 km per liter, and yours?" "mine is crap, I can't do more than 10km with a liter of gasoline" and same for auto magazines). I think only recently (new century?) they started using more the "liters per 100km" approach

shmerl• 12 years ago

How about switching to the metric system altogether?

SixSigma• 12 years ago

The United Nations creating a uniform system of regulations, WP.29 was established on June 1952, for vehicle design to facilitate international trade.

http://en.wikipedia.org/wiki/World_Forum_for_Harmonization_o...

As of 2012, the participants to the 1958 Agreement include 52 countries including all of Europe, China, Russia, Japan, Australia.

Part of the regulations is a common specification for fuel use - litres per 100km.

There is one significant non-signatory to the agreement.

jaggederest• 12 years ago

Could even use gallons per 250 miles as a fair approximation in US-style units, and I think it's a more practical measure than gallons per mile since it indicates perhaps a week of moderately heavy driving.

twic3• 12 years ago

Gallons per 280 miles is even closer. And since there are seven days in a week, you can weakly justify that as 40 miles a day every day of the week.

bigbugbag4• 12 years ago

Could it be the very country that is among the three that refuses to use the metric system with myanmar and liberia ?

http://deandecrease.wordpress.com/2012/11/22/we-dont-need-no...

return03• 12 years ago

Well, "liters per kilometer" would be even better.

bigbugbag4• 12 years ago

I disagree. First, the human mind is not that good at dealing with decimal and numbers between 0 and 1. My car is rated 4.5l / 100 km, which is easier to deal with than 0.045l / km.

For a simple calculation, say 5km, what the easiest 0.045 * 5 or (4.5/10)/2 ?

Secondly, a car is designed to be driven over distance, so making measurement on only one kilometer wouldn't be representative of real world usage.

return03• 12 years ago

agree, anything but gallons or miles. L/Km is the standard in europe anyway.

bigbugbag4• 12 years ago

From a European point of view it seem obvious that the gallons per mile trick is a small but effective cog in the larger US automobile conspiracy going on for decades. From the destruction of public transportation[1] to the many billions of dollars bailouts and current tesla battle this conspiracy is quite an obvious and documented one.

[1] https://en.wikipedia.org/wiki/General_Motors_Streetcar_Consp...

Why 'gallons per mile' is better than 'miles per gallon'

Related Stories