# Auto question re: mileage **URL:** \ **Category:** In My Humble Opinion\ **Created:** [November 28, 2015, 5:21am UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807 "2015-11-28T05:21:43Z")\ **Posts on this page:** 18\ **Page:** 2
**Author:** ![jz78817](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/jz78817/32/1028_2.png) [@jz78817](https://boards.straightdope.com/u/jz78817)\ **Post date:** [November 28, 2015, 10:56pm UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/21 "2015-11-28T22:56:16Z")
> [@Roderick\_Femm](#): > > I have said twice now that I drive pretty conservatively, meaning that I don’t rabbit from stops and race to the next stoplight, or that sort of thing. Given that, maybe you (and **Rick** ) have something concrete to offer in the way of driving advice. People in general are terrible at evaluating their own driving. Just because your tires aren’t squealing every time you take off from a stop, it doesn’t mean you’re driving “conservatively.” I’ve been along on rides with a few people who drive Ecoboost Fords, and the ones who complain about their gas mileage invariably accelerate briskly and wait until the last possible moment to brake. but they inevitably claim they drive “gently.” > [@zombywoof](#): > > I know how many miles I’ve actually driven, and how many gallons I’ve actually pumped - tracking over several weeks and dividing the sum of the former by the sum of the latter gives you your real-world mpg. yes, but you’re pumping the gas _after_ you’ve driven the miles. you’re counting on the # of gallons you’re pumping in to be the same as the # of gallons you’ve burned since the last fill. ---
**Author:** ![Dangerosa](https://avatars.discourse-cdn.com/v4/letter/d/22d042/32.png) [@Dangerosa](https://boards.straightdope.com/u/Dangerosa)\ **Post date:** [November 28, 2015, 11:08pm UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/22 "2015-11-28T23:08:26Z")
Colder weather will drop your mileage about 10%. People who drive very conservatively in terms of fuel mileage spend a lot of time with their foot not on the gas pedal, or maintaining a very consistant speed on the freeway. This is beyond quick stops and jackrabbit starts and into zen driving where you purposely maximize your mileage and do things like time lights so you never need to stop. Hills hurt - gas mileage is great going down, but up is hard. Idling your car hurts your mileage, you are burning gas going zero miles an hour - so if you spend time idling your car (like waiting for kids) that can hurt mileage. Tire pressure should be optimal - a little less air in your tires and the car is harder to move. ---
**Author:** ![Pork\_Rind](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/pork_rind/32/111_2.png) [@Pork\_Rind](https://boards.straightdope.com/u/Pork_Rind)\ **Post date:** [November 29, 2015, 12:12am UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/23 "2015-11-29T00:12:13Z")
> [@zombywoof](#): > > Well, of course it doesn’t matter whether the tank is ever even “full” or not, never mind what that means - perhaps it would be less confusing for you to think in terms of “gas purchases” rather than “fill ups”. > > I know how many miles I’ve actually driven, and how many gallons I’ve actually pumped - tracking over several weeks and dividing the sum of the former by the sum of the latter gives you your real-world mpg. Yeah, I have the most data for my motorcycle. Maybe seventy five consecutive fill-up. The onboard computer is consistently in the neighborhood of 10% optimistic as compared to fuel added vs. miles driven. I gave up recoding mileage for my car, a BMW, a while back. But over a dozen+ fill ups, it was closer but still consistently optimistic. I get that the computer _can be_ very accurate, but my experience is that it is not. Unless the gas pumps are consistently incorrectly reporting fuel dispensed, which I doubt given the regulations and inspections involved. ---
**Author:** ![Habeed](https://avatars.discourse-cdn.com/v4/letter/h/a587f6/32.png) [@Habeed](https://boards.straightdope.com/u/Habeed)\ **Post date:** [November 29, 2015, 1:38am UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/24 "2015-11-29T01:38:50Z")
> [@jz78817](#): > > the “trip computer” gets its information from the PCM, which by its very job needs to know miles driven and gallons consumed. It knows at all times how fast the car is moving, and how much fuel it is injecting into the engine. the trivial calculation is “hmm, I’m going X miles/hour, I’m injecting Y gallons/hr of fuel, divide the two, hours cancel so I’ve got X/Y miles/gallon!” > > the real reason people doubt the accuracy of the trip computer is because it averages/integrates over a time interval which hides tank-to-tank variations. OP is a great example. he is getting consistently lower mpg than the previous driver, so it stands to reason he’s driving it harder. Nope. Nope nope nope. Totally wrong. Every injector pulse a tiny amount of fuel flows into the engine. You \*can \*take injector timings and multiply time \* pressure \* orifice size factor. But you don’t really know exactly. The feedback system is "ok there’s a bunch of air flowing in, more fuel needed (manifold air system), and “too much oxygen in the exhaust, more fuel needed” and so on. It’s basically marginal feedback. The gas tank sensor may also be an input but that float/etc is also inaccurate. The car does not know exactly how much volume is missing. Eventually, an unknown (to the car) amount of liquid is used to refill the tank. What you are trying to do here is very very similar to the problem if you[piecewise integrate your GPS coordinates over time.](http://www.i-programmer.info/news/145-mapping-a-gis/9164-gps-always-over-estimates-distances.html) Every coordinate you add in is off by an unknown, but bounded percentage. You use that to measure speed. Much less accurate than timing how long it takes to traverse a known distance, and use the average speed theorem. Anyways, the gas station measurement method is better. Allegedly, though, gas pump meters do not compensate for volumetric changes due to temperature, so even this method isn’t perfect. ---
**Author:** ![Johnny\_L.A](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/johnny_l.a/32/1084_2.png) [@Johnny\_L.A](https://boards.straightdope.com/u/Johnny_L.A)\ **Post date:** [November 29, 2015, 3:01am UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/25 "2015-11-29T03:01:45Z")
> [@zombywoof](#): > > Just to be clear, have you manually calculated your actual average mileage over several tank fillups? > [@Pork\_Rind](#): > > Yeah, I have the most data for my motorcycle. Maybe seventy five consecutive fill-up… > > I gave up recoding mileage for my car, a BMW, a while back. But over a dozen+ fill ups… My spreadsheets have 681 fill-ups for the previous car, and 207 for the current one. ---
**Author:** ![Roderick\_Femm](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/roderick_femm/32/14875_2.png) [@Roderick\_Femm](https://boards.straightdope.com/u/Roderick_Femm)\ **Post date:** [November 29, 2015, 4:53am UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/26 "2015-11-29T04:53:34Z")
> [@jz78817](#): > > People in general are terrible at evaluating their own driving. Just because your tires aren’t squealing every time you take off from a stop, it doesn’t mean you’re driving “conservatively.” I’ve been along on rides with a few people who drive Ecoboost Fords, and the ones who complain about their gas mileage invariably accelerate briskly and wait until the last possible moment to brake. but they inevitably claim they drive “gently.” And yet you can tell how I drive having never seen it. I guess your experience with your “few friends” overrides my ability to evaluate my own driving. Interesting that you put two spaces between each sentence. I’m guessing you’re over 60 and learned to type on a typewriter. I have more evidence for that than you have for your opinions about my driving. ---
**Author:** ![Rick](https://avatars.discourse-cdn.com/v4/letter/r/9f8e36/32.png) [@Rick](https://boards.straightdope.com/u/Rick)\ **Post date:** [November 29, 2015, 6:22am UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/27 "2015-11-29T06:22:19Z")
> [@Habeed](#): > > Nope. Nope nope nope. Totally wrong. > > Every injector pulse a tiny amount of fuel flows into the engine. You \*can \*take injector timings and multiply time \* pressure \* orifice size factor. But you don’t really know exactly. The feedback system is "ok there’s a bunch of air flowing in, more fuel needed (manifold air system), and “too much oxygen in the exhaust, more fuel needed” and so on. It’s basically marginal feedback. > > The gas tank sensor may also be an input but that float/etc is also inaccurate. The car does not know exactly how much volume is missing. Eventually, an unknown (to the car) amount of liquid is used to refill the tank. > > What you are trying to do here is very very similar to the problem if you[piecewise integrate your GPS coordinates over time.](http://www.i-programmer.info/news/145-mapping-a-gis/9164-gps-always-over-estimates-distances.html) Every coordinate you add in is off by an unknown, but bounded percentage. You use that to measure speed. Much less accurate than timing how long it takes to traverse a known distance, and use the average speed theorem. > > Anyways, the gas station measurement method is better. Allegedly, though, gas pump meters do not compensate for volumetric changes due to temperature, so even this method isn’t perfect. Allow me to borrow from your post Nope. Nope nope nope. Totally wrong. The ECM know exactly how much fuel is injected. If the oxygen sensors call for richer or lean it knows how much more or less is injected. As I mentioned the data read out on a Toyota is in micro liters. That is one one millionth of a liter. And you are telling me that isn’t accurate? Or are you telling me that that the shutoff on the nozzle is more accurate than 1/1,000,000 of a liter? ---
**Author:** ![Hilarity\_N.Suze](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/hilarity_n.suze/32/6223_2.png) [@Hilarity\_N.Suze](https://boards.straightdope.com/u/Hilarity_N.Suze)\ **Post date:** [November 29, 2015, 8:59am UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/28 "2015-11-29T08:59:48Z")
I routinely get at least 5 more mpg in the car than my husband driving the same car. We don’t know why–we’re both good drivers. I think I might be a little gentler, and I also think possibly I take my foot off the throttle a little sooner when I see that I might have to brake. He also outweighs me by 60 pounds or so. I have no idea if that’s a factor. Sometimes we’re both in the car and I think it probably averages out. Now, I can really push it, and I have, when driving through an area where I thought there would be a gas station and it was closed. Oops. 30 or so miles to the next one. So, going up hills, no extra acceleration, going down hills, no extra acceleration. (It helped that there was exactly zero traffic besides me.) I am not sure he could have gotten us to the next gas station, and he agrees, and we would have been SOL because there was not a lot of cell phone reception out there either. I also think (and this is not really backed up by any data) that I get better mileage when the tank is over 1/4 full. That is to say, if I’m driving it, that’s when I fill it up. But if I’m driving it it’s getting better mileage anyway. ---
**Author:** ![california\_jobcase](https://avatars.discourse-cdn.com/v4/letter/c/9d8465/32.png) [@california\_jobcase](https://boards.straightdope.com/u/california_jobcase)\ **Post date:** [November 29, 2015, 3:31pm UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/29 "2015-11-29T15:31:28Z")
Fueling time of day and weather are mentioned as factors in fuel density. However, in the U.S., all gas tanks are well underground. I bet there is very little, if any, variance in fuel density in those tanks due to time of day or weather. However, I have measured significant MPG drops with winter fuel in Indiana compared with gas in S. Georgia (yes, the topography is similar). ---
**Author:** ![kopek](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/kopek/32/337_2.png) [@kopek](https://boards.straightdope.com/u/kopek)\ **Post date:** [November 29, 2015, 4:30pm UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/30 "2015-11-29T16:30:24Z")
> [@Roderick\_Femm](#): > > I have said twice now that I drive pretty conservatively, meaning that I don’t rabbit from stops and race to the next stoplight, or that sort of thing. Given that, maybe you (and **Rick** ) have something concrete to offer in the way of driving advice. Conservative could be the issue; to really tell anything for sure I would need to visit and ride with you several assorted trips. Admitting up front we’re using standards and not automatics, I drive “harder” and always get better mpg than the Old Wench who is much closer to the speed limit and slower off the starts. The problem is that she spends more time outside the engine’s designed power band making her conservative driving a little less efficient than mine. Not tons different but enough that when we chart it on paper we get say a 2-3mpg difference. ---
**Author:** ![Roderick\_Femm](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/roderick_femm/32/14875_2.png) [@Roderick\_Femm](https://boards.straightdope.com/u/Roderick_Femm)\ **Post date:** [November 29, 2015, 4:37pm UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/31 "2015-11-29T16:37:04Z")
> [@kopek](#): > > Conservative could be the issue; to really tell anything for sure I would need to visit and ride with you several assorted trips. Admitting up front we’re using standards and not automatics, I drive “harder” and always get better mpg than the Old Wench who is much closer to the speed limit and slower off the starts. The problem is that she spends more time outside the engine’s designed power band making her conservative driving a little less efficient than mine. Not tons different but enough that when we chart it on paper we get say a 2-3mpg difference. How does one tell what is the engine’s designed power band (I don’t even know what that means)? The car has a little gauge that is supposed to show the current mpg efficiency, so to speak - it’s a needle on a dial that goes down when I accelerate and goes up when I’m coasting. I do pay attention to that somewhat, where it’s safe to do so (it does require taking my eyes off the road though). ---
**Author:** ![HoneyBadgerDC](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/honeybadgerdc/32/1033_2.png) [@HoneyBadgerDC](https://boards.straightdope.com/u/HoneyBadgerDC)\ **Post date:** [November 29, 2015, 5:11pm UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/32 "2015-11-29T17:11:19Z")
My brother was having problems with his mileage. It turns out he used a program he bought to set his truck for towing specs. Once he put it back to normal driving settings the mileage came back. ---
**Author:** ![kopek](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/kopek/32/337_2.png) [@kopek](https://boards.straightdope.com/u/kopek)\ **Post date:** [November 29, 2015, 8:04pm UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/33 "2015-11-29T20:04:25Z")
> [@Roderick\_Femm](#): > > How does one tell what is the engine’s designed power band (I don’t even know what that means)? > > The car has a little gauge that is supposed to show the current mpg efficiency, so to speak - it’s a needle on a dial that goes down when I accelerate and goes up when I’m coasting. I do pay attention to that somewhat, where it’s safe to do so (it does require taking my eyes off the road though). It’s possible its in the owners manual or somewhere like that. I’m not sure how those current gauges operate or how accurate they are but I do know people who followed their tachs even when driving automatics to max their mileage. My Dad’s old Tempo did best when you kept it between 2600 and 4000 or so as I recall. (Dad was a serious nut about maxing MPG) ---
**Author:** ![Roderick\_Femm](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/roderick_femm/32/14875_2.png) [@Roderick\_Femm](https://boards.straightdope.com/u/Roderick_Femm)\ **Post date:** [November 29, 2015, 8:27pm UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/34 "2015-11-29T20:27:42Z")
> [@HoneyBadgerDC](#): > > My brother was having problems with his mileage. It turns out he used a program he bought to set his truck for towing specs. Once he put it back to normal driving settings the mileage came back. I have this nagging suspicion that the Subaru dealership I bought it from changed something, with I’m sure the best of intentions (:rolleyes:) that is having this effect. Like maybe they tightened up the parking brake so that it’s always on some even after I release it, or something like that. I may bite the bullet and take it to a mechanic (not the dealership though) to see if they can find anything. ---
**Author:** ![TSBG](https://avatars.discourse-cdn.com/v4/letter/t/77aa72/32.png) [@TSBG](https://boards.straightdope.com/u/TSBG)\ **Post date:** [November 29, 2015, 8:40pm UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/35 "2015-11-29T20:40:20Z")
Sorry if I missed this in the thread, but have you tried contacting the previous owner to ask how he got 25.5 mpg? The MPG calculator on my Prius is consistently wrong, by the way–says I’m doing about 10% better than revealed by mileage and gas consumption. ---
**Author:** ![GreasyJack](https://avatars.discourse-cdn.com/v4/letter/g/df705f/32.png) [@GreasyJack](https://boards.straightdope.com/u/GreasyJack)\ **Post date:** [November 30, 2015, 4:34pm UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/36 "2015-11-30T16:34:54Z")
I think the obvious question is assuming you did the same general driving cycle on your previous car, how did you do with it mileage-wise compared to its EPA rating? The EPA ratings might not be all that relevant to your own personal driving cycle, which is particularly true with city driving since your “city” driving might have many more stops, idling, etc than the EPA one. If you’re that worried about the car itself, the thing to do is go for a weekend trip somewhere that’s all highway driving (preferably like steady 55-65 highway driving). If that’s nowhere near the highway rating, then I might start complaining about it. On a more specific note, this car has a CVT which does tend to increase the difference in mileage you see depending on your drive cycle. They also render the stuff **kopek** mentioned about power bands somewhat irrelevant because unlike a conventional automatic (or obviously stick) the CVT will generally always keep the engine at a good RPM to produce power. On a CVT it really is just “less gas pedal application = better fuel economy.” Also, how many miles are on it now? I’ve got basically the same car and my mileage went up a bit (and oil consumption went down) after the first few thousand miles as the engine broke in. ---
**Author:** ![GreasyJack](https://avatars.discourse-cdn.com/v4/letter/g/df705f/32.png) [@GreasyJack](https://boards.straightdope.com/u/GreasyJack)\ **Post date:** [November 30, 2015, 5:29pm UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/37 "2015-11-30T17:29:55Z")
> [@Roderick\_Femm](#): > > I have this nagging suspicion that the Subaru dealership I bought it from changed something, with I’m sure the best of intentions (:rolleyes:) that is having this effect. Like maybe they tightened up the parking brake so that it’s always on some even after I release it, or something like that. I may bite the bullet and take it to a mechanic (not the dealership though) to see if they can find anything. Now that you mention it… was there ever any explanation given for why the car was traded in with so few miles? Subaru has replaced quite a few engines in these due to oil consumption, mostly in the 2012 and 2013 MY (although maybe a few 2014’s too). It could be that this car was one of them, but the original owner decided he didn’t want to deal with it. If that’s the case, the car might have had a nearly brand-new engine when you bought it and so you could still be in the break-in period and the cumulative mileage includes a bunch of driving on the old (broken-in) engine. ---
**Author:** ![PastTense](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/pasttense/32/14550_2.png) [@PastTense](https://boards.straightdope.com/u/PastTense)\ **Post date:** [November 30, 2015, 5:51pm UTC](https://boards.straightdope.com/t/auto-question-re-mileage/738807/38 "2015-11-30T17:51:36Z")
There are sites which collect this data from drivers, for example: > **[Fuelly - Track and Compare your MPG](https://www.fuelly.com/)** > > Fuelly allows you to tracks your gas mileage over time, helping you calculate fuel expenses as you drive. Here is your car: > **[Subaru Impreza MPG - Actual MPG from 5,014 Subaru Impreza owners](https://www.fuelly.com/car/subaru/impreza)** > > The most accurate Subaru Impreza MPG estimates based on real world results of 108.3 million miles driven in 5,014 Subaru Imprezas [Previous page](https://boards.straightdope.com/t/auto-question-re-mileage/738807.md?page=1)