How to save money on gas

Environmental Economics and The Energy Collective are among the many voices recently advising consumers they could save gasoline by driving more slowly. I was curious to take a look at the evidence behind such claims.

Source: U.S. Dept. of Energy.

Air resistance increases the faster you travel, which might lead you to think that higher speeds always require more fuel. However, your car’s engine is designed for maximal efficiency in converting fuel into motion when you drive at higher speeds. As a result, the typical car gets much better gas mileage if you drive it at 45 mph instead of 15. However, at speeds above 60 mph, the wind resistance becomes a dominant factor, and miles per gallon for most cars starts to decline significantly if you drive faster than 60.

The basis for the graph above and many of the quantitative claims one finds in newspapers and blogs appears to come from a 1998 study, whose raw data can be found in Table 7.23 of the DOE’s Transportation Energy Data Book. This reports miles per gallon as a function of speed for 9 automobiles on the road in 1997. In principle, this is a straightforward thing to measure– you drive the car around a fixed track trying to keep it at, say 65 mph, and measure the fuel consumption. Then you do it again at 75 mph and calculate the change. Consumer Reports and have run analogous trials with similar results to those reported by DOE. The first column of the table below summarizes those results in terms of how many more gallons of gas it would require to drive 100 miles at a speed of 75 rather than 65 in each specified vehicle, while the second column expresses these same numbers as the percent by which you’d reduce your fuel consumption in that type of car if you reduced your average speed from 75 to 65. For example, with the 20% fuel savings you’d get from slowing down in the Olds 88, it’s as if you could buy your gasoline for $3.20 a gallon rather than $4.00 and still do just as much driving.

Consumer savings from driving at average speed of 65 rather than 75

gallons per
100 miles

% reduction
in fuel cost

equivalent hourly wage


San Diego


1988 Corsica






1993 Legacy






1994 Olds 88






1994 Cutlass






1994 Chevy pickup






1994 Cherokee






1994 Villager






1995 Prizm






1997 Celica






Of course, there’s also an economic cost of driving more slowly, which is you’ll take longer to get where you’re going. Again considering this idealized experiment of driving a distance of 100 miles at exactly 65 mph rather than 75 mph, it would take you

100 miles x (1 hour/ 75 miles) x (60 minutes/hour) = 80 minutes

to get there at 75 mph, but 92 minutes if you only drive at 65. So you’d spend an extra 12 minutes to save 0.94 gallons if you drive a 1994 Chevy pickup for 100 miles.

How much money that saves you depends on how much you pay for gas. My table reports three reference values, first a “national average” based on the current average U.S. price of $4.09/gallon, the second a high-priced community (my home San Diego, where it’s now $4.59), and the third for one of the cheapest spots in the country (Oklahoma City’s $3.76/gallon)– those prices come from You can then convert that to an hourly wage you could consider yourself to be earning for driving more slowly. For example, if you drove that Chevy pickup for approximately 500 miles, you’d do an extra hour’s worth of driving, but save yourself $18.70 if you were paying the current national average retail gasoline price.

And, by the way, that $18.70 per hour is entirely tax-free income from your point of view.

Having done these calculations, I should point that you can’t always drive on San Diego’s freeways at 75 mph. If you try to maintain that speed, you will be braking and accelerating more than someone who tries to keep it at 65. Each time you put your foot on the brake pedal, that’s costing you money as well, in that you’re killing the momentum provided by the fuel you’ve already paid for and burned. DOE separately claims that motorists could reduce their fuel bill by 5-33%, depending on their existing driving conditions and habits, by trying to use the brake pedal less and accelerate more slowly.

Driving more slowly should also reduce your carbon footprint and any other pollutants by the same amount indicated in the second column above, and should lessen the likelihood and severity of collisions. And most urban highways prohibit speeds over 65, so you’re exposing yourself to risk of fines and higher insurance rates if you drive at higher speeds. Even without these added considerations, however, I would think that many people, if they knew that the immediate financial rewards were on the order of the numbers given above, might choose to drive more slowly. In which case, it is perhaps a public service to help call such numbers to people’s attention.

Which, in case you were wondering, is why I wrote this post.

Technorati Tags: ,

52 thoughts on “How to save money on gas

  1. Rich Berger

    I notice a 5% increase in mileage by keeping my speed at 65 instead of 75 in my 2007 Pilot. It increases my trip time, but it has the pleasant side effect that I am being passed by other cars and spending more of my time in the open and not trying to make my way through knots of traffic.

  2. Lord

    The tough thing about that is if you drive in the fast lane you will have a bunch of annoyed drivers behind you, but if you shift to slower lanes you will have more congestion. About the best you can do is select the lane you won’t have any more congestion in.

  3. Dan Weber

    Does anyone know where one could find the above graph, but specific for make/models of car?
    Interestingly, the graph for electric cars has awesome fuel economy at low mileage. The internal combustion engine sucks at efficiency, but sucks less at 55mph. An electric motor gets very good efficiency, so you could potentially push a Volt hundreds of miles on a single charge if you were going slow enough.

  4. Anon

    Sorry but you fail the grade on this particular report.
    The obvious conclusion is that slow speeds in heavy traffic and at stoplights are the big culprit in consumption – not your cruising speed.
    Toyota makes the Dean’s honors list though.

  5. JDH

    Ironman, I was thinking you might make a neat tool out of these kind of data if you had an interest, e.g., input your EPA highway gas mileage, typical current speed, and cost of gas per gallon and it would output your hourly dollar time compensation for slowing down.

  6. JDH

    Anon, I’m not sure how to avoid red lights, but I am sure how you personally can avoid driving 75 mph.

  7. Dave Brown

    I recall a BBC report about a study done in Europe about 8 years ago. The bottom line: for cars with manual transmissions, accelerating at about 3/4 throttle (thats pretty fast) produced the best fuel economy. It makes perfect sense to me; the wider your throttle is open, the better the engine’s volumetric efficiency. And assuming you are going to cruise at the same speed, the trade-off (brisk acceleration = lots of fuel x short time vs. slow acceleration = little fuel x long time)is net zero, until you take volumetric efficiency into account.
    As to cruising speeds, the engine management system in any modern automobile will maximize efficiency at any speed and load. You will have an increase in volumetric efficiency (due to lower pumping losses) at say 75 mph vs. 65 mph, but not enough to overcome aerodynamic drag and rolling drag (tires, drivetrain) both of which increase exponentially with speed.
    The best technique: go with the flow of traffic, leave a generous following distance to avoid excessive braking, and coast whenever possible. And as Max said, avoid fuels blended with ethanol or methanol. Alcohols have less energy (BTUs) per pound (or gallon); in other words, you’ll have to open the throttle wider to maintain a given speed.

  8. odograph

    To throw out a tidbit from a Prius driver who eyes the mpg computer regularly … the absolutely worst thing is a boulevard with a high speed limit and frequent stops: get up to 55, stop, get up to 55, stop …
    It kills everyone’s mileage, but they must build them that way for the perception of speed. If you aren’t really paying attention you think you are getting there faster.

  9. central texas

    Just as a rule of thumb, I was always taught that the most fuel efficient operation was achieved at or near the slowest RPM that could be sustained in the highest gear available. That arrangement should result in the greatest number of revolutions of the wheels with the least number of revolutions of the crankshaft. (and the least drag) This is relatively easy to do in a standard transmission vehicle but much more difficult with an automatic transmission.

  10. Ironman

    It’s a neat idea – I’ll drop it into the development queue for quick turnaround. Look for it either Wednesday or Thursday next week. (I’ll have it coded much sooner, but I’m already going to debut a new price-of-gas related tool on Monday and past experience has shown that back-to-back tool posts are the kiss of death for maintaining site traffic!)

  11. Michael Cain

    Just curious…
    The table shown in the post has no cars newer than a 1997; the Edmunds article used a Volvo and a BMW, neither noted for their fuel economy. Over that decade manufacturers such as Honda have improved aerodynamics, improved engine efficiency, and added more standard gears in the automatic transmission in some models. One might expect such improvements to shift the point at which mileage falls off to higher speeds. Is more recent data available?

  12. Mace

    “if you drive in the fast lane you will have a bunch of annoyed drivers behind you, but if you shift to slower lanes you will have more congestion”
    The rules of the road, to quote Dennis Miller – “Left lane fast, right lane slow!”
    You can also improve your MPG by 10% and reduce gas costs by $.40/gallon by not buying your gas in California.

  13. JDsg

    …you drive the car around a fixed track trying to keep it at, say 65 mph, and measure the fuel consumption.
    Why put the car on a track where other factors (e.g., wind) might muddy the results? One would think they’d have put the cars into wind tunnels.

  14. spencer

    I find it interesting that in your example you find that you save $18.70 per hour by driving slowly and
    this is 104.4 % of the national average hourly earnings of $17.94. This implies that on a pretax basis if you time is worth much more then the average you will actually lose money by driving slowly– at least on an opportunity cost basis.

  15. JDH

    Michael Cain, that’s an excellent question. In the brief time I tried Googling for this, I found lots of very recent statements about this question, but when I tried to track down the data behind them, they all appeared to be derived from the sources I’ve cited. Maybe some of our readers can point to some systematic study using newer models.

  16. Esprit

    Has any institution calculated the cost of gas to the American driver of being legally required to STOP completely at every junction (with a stop sign)?
    I’m not aware of any other country in the world (although I’m only familiar with Europe and the US) that forces drivers to follow this rule.

  17. Bruce Hall

    The Federal Highway Administration published the solution several years ago:
    I have written about this and also been successful, locally, at having the situation address on a major, 6-lane, surface street by simply writing to the DOT with a cc to the governor.
    If enough people became pain in the asses to their DOT and state government, you can bet that there would be action.
    Remember those stickers: 21 mpg city/33 mpg highway

  18. Esprit

    Bruce, the stop signs I was referring to are the typical STOP signs at a T junction or merging lanes area. Sure, the RYG signals certainly need a massive policy overhaul in terms of safety, traffic flow and necessity but the mindset here seems to be of adding yet more general sign-age rather than reviewing existing policies and improving management of traffic flow.
    The estimate of 296 million lost vehicle hours is staggering and even worse is how much larger this figure would become if you added my rant against the ill-placement of standard stop signs.
    I’ve think I’ve found the complete paper connected to your link and I might spend some time seeing if I can calculate the STOP sign cost as it effects driving hours/lost production, safety and gas usage. I will have to make a lot of generalisations and the result could be questionable but it’s a start.
    Would you know if all US states require a driver to stop at a stop sign rather than provide it simply as a ‘be aware’ sign?

  19. flatface

    I recommend tailgating trucks to minimise air resistance. Works wonders for your reactions too!

  20. gin

    I have owned 2 Prius a 2002 that i traded in for a 05 as i needed a larger cargo area , The car if your willing to pay attention will teach you to drive for best fuel economy with the 03 my lifetime economy was 49.9 mpg (miles/fuel purchased)over 75K miles now with the 05 I seldom drive over 60mph and i tend to stay at the speed limit on non interstate roads i am currently getting 55-60mpg in the summer and 50-55mpg in the winter. I not trying to hypermile just driving conservatively

  21. ndallasj

    UPS has modified their local truck routing algorhythm to maximize the number of right turns and claims that they are saving millions! Makes sense–with right turn on red, no extended waiting at red lights and no waiting for “Left turn on Green Arrow Only” intersections, etc.

  22. cerqueira

    No scientific evidence, but I found a post at the Green Car Congress named Fuel Consumption at Higher Speeds about a test by the german Auto Bild magazine. They tested sixteen models and it seems that technology has not revoked the laws of physics: even in recent models, fuel consumption grows steadily from 50 mph on. I would expect the results to be more expressive for SUVs, due to their barn door aerodynamics.

  23. Bruce Hall

    I know of no state where STOP means anything other than come to a complete stop. YIELD means watch the hell out but go ahead through.
    I also like the idea of expressing fuel efficiency in gallons per 100 miles rather than miles per gallon. There are a lot of people concerned about whether they are getting 35 instead of 45 mpg when the big savings are going from 15 to 25 mpg in the course of normal travel per month.

  24. jm

    In the UK there are no stop signs, and the British consider us absolute idiots for having them.
    There’s a good reason why hardly anyone comes to a full stop at a stop sign unless actual traffic requires it.
    It’s stupidly unnecessary, and people know it.
    In the UK the rule on residential streets is simple — keep an eye out and yield the right of way to the vehicle on the right. Where a lower-priority road intersects a higher-priority road, there may be a dashed line on the pavement across the lower-priority, indicating that drivers should yield right-of-way to traffic on the higher-priority road.
    At intersections of higher-priority roads, one finds traffic circles, and yields the right-of-way to vehicles already in the circles.
    Traffic flows more smoothly, with less waste of energy.

  25. Bruce Hall

    Here in the Detroit metro area, several high-volume intersections on secondary roads have been converted to roundabouts. It is an expensive process to retrofit an intersection, but does improve traffic flow at peak volumes.
    There is always a learning period when dealing with no-traffic-control situations. Recently, major storms knocked out traffic signals across the area. Fortunately, it was at night during low volume hours. Nevertheless, the 4-way yield situation nearly doubled travel time.
    Traffic signals are beneficial for high-volume road if they are timed to allow continuous movement of vehicles for long stretches. Yield signs make more sense in lower volume areas. Flashing yellow lights are useless unless limited to school zones to announce lower speed limits. Stop signs should be reserved for intersections with restricted lines of sight… and one street should be given priority [no stop sign] over the other. 4-way stop signs are just congestion producers.
    The point is that while increased traffic volumes are going to reduce vehicle efficiency, much of the inefficiency can be attributed to poorly managed traffic control systems and poorly designed streets.

  26. Willy

    In the non centralised electrical business model, there is an economic/energy saving by locally producing electricity (we haven’t transmission losses), worldwide authority don’t encourage the local production: why?

  27. gubbi

    I drive at 65 mph in california on the 710.I put it on cruise as I hit the freeway, get in the lane next to the fast lane and cruise. the mileage on my car Acura TSX has improved from an average of 29 mpg to 31.5 for the last 3 months.
    The last two times (I fill up every 8 days) I got 31.85 mpg. The total commute is 25 miles one way with 11 miles of city roads. The best way to this is to time the green lights, which I have been doing consistently for the past 1 year. Yippe

  28. 2slugbaits

    When I leave for work each morning at 4:30am, the interstate is largely deserted except for me and a few trucks. I noticed about a year ago that when left to their own preferences, truckers slowed down from the 70mph posted limit to 55-60mph. I lowered my speed to match the trucks. Two reasons really. First, I’m usually not in a hurry to get to work. And second, I let the trucks clear the way of deer and varmits on the road. But in the afternoon the trucks and I all have to increase our speed to 70mph in order to flow with the traffic. This is kind of a reversal of the old thinking when trucks always drove faster than the cars on the road. Now it’s the cars that are pushing the trucks.

  29. Bill S

    Years ago when I studied the impact of the 55mph speed limit, I saw a number of econometric studies of highway safety that looked not only at how the posted speed limit affected the rate of accidents and fatalities, but also at the relationship between accidents and the variance of speeds actually traveled on given roads. Not surprisingly, the greater the variance (with some slowpokes creating road rage among the high-testosterone crowd), the higher the accident and fatality rates. So encouraging some people to slow down with no change in posted limits could well contribute to greater carnage on the highways.

  30. GWG

    Thanks for this post. I previously mentioned in one of Menzie’s posts how most people seem to be driving as if gas was much cheaper. I would have expected to see some sort of public service type announcements on this subject by now. The value of one’s time will always be taken into account but I am not sure that everyone realizes how much their driving habits may be costing them and soceity as a whole. Conservation is certainly part of the solution and the quicker we learn that the better.

  31. DickF

    Why not just pass a law that makes it illegal to sell or own any car or other vehicle that gets less than 50mph? Imagine how nice it would be to drive without worrying about those pesky 18 wheelers.
    Then we could make it illegal to consume any meal that is than 1,000 calories. Imagine the savings in health care cost.

  32. Austin Chu

    Great post. I’ve always wondered how the numbers worked. I work for a company that manages and tracks gift cards and I’ve been blogging about non-technical ways to save money on gas on One thing to consider is to buy discounted gas cards online. As for me? I converted my car to run off waste vegetable oil. I haven’t paid for fuel since January. you can check out my ride on

  33. benamery21

    Note that these results will be skewed upward if you drive with A/C on. The mph sweet spot for the highest mpg will be higher (although the overall mpg will be lower) because the A/C gas consumption is increased by traveling more slowly, since it is required to cool roughly the same heat load for a longer period of drive time.

  34. Paul

    No, we shouldn’t use gallons per 100miles. We should use L/100km like the rest of the world. Then we’d also be able to read other peoples data without having to sit down with a calculator (and make a _lot_ of errors while we’re at it).

  35. Russell L. Carter

    I get about 40 mpg @ 90mph in a Prius loaded with 3 adults and luggage. Around town we get maybe 45 mpg no matter how we drive (wife turtles around, I’m a leadfoot, on both pedals).
    So all the above machinations to eek out just a bit more from obsolete technology seem a little… sad.
    (For the safety nuts, out here in the west it’s possible to drive like 20 miles at a time w/o passing another vehicle. In JDH’s town I drive with the flow)

  36. GNP

    We’ve dropped the highway cruise speed on our ’93 pathfinder SUV from 110-115km/hour to 95km/hour. The fuel savings are noticeable especially when lugging a 16′ Chestnut canoe and gear.

    Moreover, Porcupine Tree sounds much better on the sound system. The ambient vehicle/road noise is radically reduced.

    Otherwise, we walk and bicycle to do local errands as always.

  37. George Barwood

    One thing is to put the car into neutral on downhill sections where your speed can be maintained without power from the engine. This is straightforward, reduces revs from typically 3000rpm to about 800rpm, saving fuel.
    An extension of this idea is “pulse and glide”, which aims to use the engine at high efficiency for a few seconds, then either idle in neutral or switch off** for a longer period. Hybrids automate the process, and by using battery power can glide for longer. But it works perfectly well on conventional cars.
    **Obviously there are safety implications if you switch the engine off – for a start loss of power steering.

  38. aaron

    This is very decieving. The data is over 10 years old. It doesn’t consider passangers per vehicle (BTS has 1.63 as the average occupancy). I took the data from the BTS table. Using those 1997 numbers (there have been great improvements in engines and aerodynamics since then) I calcualed the $ consumed per second and hour at each speed:
    @ 55 MPH $5.89 per person per hour
    @ 75 MPH $7.59 per person per hour
    @ 70 MPH $6.55 per person per hour
    @ 65 MPH $5.58 per person per hour
    or per vehicle per hour:
    But what is most important is the difference between 55 MPH and 75 MPH. At 55 MPH you save only $2.30 per person per hour versus 75 MPH, or $3.75 per vehicle per hour! Average productivity per person per hour in the US is $4.97 (Total 2007 GDP divided by total 2007 population and total hours per year). The amount of savings is more than 50% less than the average productivity per person!

  39. aaron

    Oops! Lysdexia acting up.
    @ 55MPH, $4.26 per person per hour, $6.94 per vehicle per hour.
    @ 75MPH $7.59, $12.37
    @ 70MPH, $6.55, $10.68
    @ 65MPH, $5.59, $9.10
    Thats a savings of $3.33 per person per hour or $5.42 per vehicle per hour. Still 33% less than the average productivity per person per hour of $4.97.

  40. aaron

    There is probably nothing you can do to reduce fuel consumption better than attending your local government meetings and pressuring them to time traffic lights and set speedlimits appropriately (and try to avoid changes in speed limits), and remove unnecessary traffic lights and stop signs.
    Another thing communities can look into is possibly installing variable speed postings to smooth out traffic. Electronic signs could give drivers a target speed that will reduce the number of stops and create openings to allow traffic to clear from queues.

  41. aaron

    Ironman, very nice. But I have one request, please make a clarification on the aggressive driving portion. There is a big flaw in the methodology used in the test. It tests hard acceleration with hard braking, not independently. Hard acceleration is actually more fuel efficient, avoiding high speeds and braking is where you get the fuel savings.
    See this from the New York Times:

    People saved the most gasoline when they pushed down on the accelerator briskly and then shifted quickly, keeping the revolutions per minute low — not by accelerating very gently.
    ”It’s not commonly understood by people who drive,” Dr. Dougherty said. ”They think that the way to get best fuel economy is to accelerate very gently, but that proves not to be the case. The best thing is to accelerate briskly and shift.
    ”Don’t give it everything the car has, but push down when you’re going to shift, using maybe two-thirds of the available power, and change through the gears relatively quickly.”
    Another time when gasoline was saved was during braking. ”The main thing is to anticipate better when you are going to need to stop,” Dr. Dougherty said. Then you should take your foot off the accelerator and use air resistance and friction to help slow the car.

  42. Paul

    What do you know, the peak (using defaults and just changing the speed) is at 110km/h which is the normal maximum freeway speed limit in many countries.
    It’d be a more useful tool if it also did metric.
    Like benamery21, I think we’d be a lot better off if we just used L/100km like the rest of the world.
    If you want a bumper sticker denoting your efficiency in L/100km, you might like to take a look at

  43. supersaurus

    “The best thing is to accelerate briskly and shift.”
    since it takes exactly the same amount of work to increase the same object’s speed a given amount regardless of acceleration, assuming acceleration is always positive (no perverse go/slow/go/slow methods), the above statement, if unqualified, makes no sense, i.e. if you accelerate twice as fast you spend energy at twice the rate while you are accelerating, but you only spend it for half as long.
    back in the good old days I always thought the exact opposite happened, because when you stood on the gas the accelerator pump dumped a bunch of extra gas in there and the mix stayed richer while you had your foot in it, to increase the power. my vague feeling was the mixture was optimized for power, not for efficiency, so I expected the overall fuel consumption to go down when I stood on the gas often, which in fact is exactly what happened.
    nowadays, with that fabulous engine computer and fuel injection maybe the amount of fuel entering the cylinder is always near optimal. if that is so then why does acceleration matter?
    it seems to me the reason it matters is the engine’s efficiency varies with power output, and if you force it to put out too much power to maintain your desired acceleration then the efficiency goes down (and of course the same will happen if you use too little).
    I guess all that boils down to I’ll concede there is an optimal acceleration for best fuel economy, but “brisk” doesn’t give me a clue what that might be.

Comments are closed.