Cars you've driven and your MPG vs EPA MPG?

[ketchup]

How has your MPG compared to the EPA ratings for your car/cars you've driven long enough to notice trends?

(I'm using the 2008+ "revised" numbers referenced from FuelEconomy.Gov for this.)

First, my parents' old cars I drove ~5 years ago and before:

2003 Saturn VUE.  Rated 19/26.  I averaged 27MPG in a roughly 50/50 city/highway commute.

1994 Buick LeSabre.  Rated 17/26.  I averaged 23MPG in the same 50/50 city/highway commute.

Cars I've owned:

1988 Chevrolet Sprint.  Rated 44/51.  I averaged 53MPG no matter how/where I drove it.

1996 Volvo 850 Wagon.  Rated 17/24.  I averaged about 24MPG city, 27MPG highway.

1999 Chevrolet Metro.  Rated 34/42.  I averaged 45MPG no matter how/where I drove it.

1992 Buick Roadmaster Wagon.  Rated 15/23.  I average about 15 city, 20 highway.

Other cars I've driven:

2009 Kia Spectra.  Rated 24/32.  Only drove it on the highway and averaged 34MPG.

2013 Toyota Prius c.  Rated 53/46.  Averaged about 60MPG city, 52MPG highway.

So for me at least, it seems like most of the EPA ratings are pretty conservative, but not too far off (except for a few).  The only one I undershoot the highway rating on is the 1992 Buick Roadmaster Wagon and that might have something to do with the fact that our Roadmaster highway miles tend to be accompanied by a lot of cargo, and with the fact that we're getting close to rolling it past 200k.  Also, the car is a greenhouse and needs AC on constantly if it's sunny out and above 65F outside.

How do your numbers compare?

[onlykelsey]

Hm.  The only car I have data on is my 2007 Honda Fit (manual 5 speed) Sport.

Apparently EPA thinks 28 city and 34 highway.  I definitely moved cross country (fully loaded with a 95 lb dog as well) and got JUST north of 40, but I sort of made it my goal to save gas money.  Overwhelmingly highway miles, and I didn't go over 60 unless the traffic demanded it.  Tried to keep AC off or on low despite it being summer in Texas. 

[neo von retorch]

2006 Nissan 350Z (22.2 mpg) EPA 17/23/19
2008 Honda Fit Sport (34.4 mpg) EPA 28/34/31
2013 Mazda CX-5 (31.2 mpg) EPA 26/35/29
2008 Acura TL (23.4 mpg) EPA 18/27/21
2007 Honda Fit Sport (32.5 mpg) EPA 28/34/31
1999 Chevrolet Tahoe (14.3 mpg) EPA 11/15/13
1994 Dodge Stealth (25.0 mpg) EPA 17/22/19
1999 Subaru Impreza (26.7 mpg) EPA 19/26/22
1985 Honda Prelude (27.2 mpg) EPA 22/29/24

[gggggg]

2004 Honda CRV.... EPA: 22 mixed  Me: 24 mixed
2015 VW GTI, manual trans.....EPA: 28 mixed  Me: 32 mixed (I've gotten it insanely high on long hwy cruises though)

[onlykelsey]

2004 Honda CRV.... EPA: 22 mixed  Me: 24 mixed
2015 VW GTI, manual trans.....EPA: 28 mixed  Me: 32 mixed (I've gotten it insanely high on long hwy cruises though)

I think "manual" may say it all.  I don't know if I can ever buy an automatic, although I see the appeal.

[SoccerLounge]

I'm too lazy to type all the data in for every car I've owned, but I agree that beating EPA - especially the current way they calculate, assuming lead-footed drivers - is fairly easy.

1st gen Honda Insight: EPA 48 city / 59 hwy / 53 combined. I see about 55-60 city, 80-85 hwy, about 68-70 combined. Pretty good for a vehicle that is still primarily gasoline powered (the electric assist system has a relatively limited role compared to later hybrids). There are Insight owners who do far better than that, but I don't really want to do things like never run my AC, or drive at 45mph on the Interstate, so... :)

[Paul der Krake]

About 2-3 MPG higher than the EPA rating across 3 vehicles.

[Mikila]

Fun thread!

Unless I am in a hurry, I strategically drive to minimize fuel usage.

2002 Toyota Echo manual. EPA MPG 29/37/32. My mpg was 45 all day long.  Never got less than 40, and maximum was 59 at a steady 65 mph, no a.c. and no stops.

2010 Toyota Yaris manual EPA MPG 29/36/32.  My mpg is 38-40 mpg, depending on passengers.

1991 Chevrolet S-10 auto.  EPA MPG 16/17/17. My mpg  was 19, driving four miles at a time.

1994 Toyota pickup 4 cylinder 2wd. Auto.  EPA MPG 19/23/21.  My mpg was 21.  This one had some mechanical problems that lowered fuel mileage.

1996 Toyota Camry 6 cylinder auto.  EPA MPG 18/26.  Most driving was short trips- a couple miles at most at a time, and then my mpg was 19.  Highway driving I could attain 31 mpg. 

2004 Pontiac Grand Prix 6 cylinder auto.  EPA MPG 18/28/21.  My mpg ranged from 24 to 31, between town and highway driving.

1979 Datsun/ Nissan 280z.  We were lucky to get 18 mpg, but in it was a fun ride.  No epa estimate for it that I found.

1996 Toyota Avalon 6 cylinder auto.  EPA MPG 18/26/21.  Consistent mpg over years of tracking was 32.


If you are a careful, strategic driver, it is not difficult at all to exceed the EPA estimates. 
 

[BudgetSlasher]

What numbers is everyone using for actual? I ask because when I have attempted to track (usually a specific trip, not a cars lifetime) I've noticed across multiple cars that the in car computer is 2-5 MPG higher than miles traveled divided by gallons put into tank.

That being said when I have tracked a trip I usually beat the EPA rating . . . unless I ended up hitting a major city's ~3-hour rush hour traffic.

[Paul der Krake]

What numbers is everyone using for actual? I ask because when I have attempted to track (usually a specific trip, not a cars lifetime) I've noticed across multiple cars that the in car computer is 2-5 MPG higher than miles traveled divided by gallons put into tank.

The only way of doing this reliably is what you outlined: divide mileage by gallons pumped. Instant MPG reading from the car computer is worthless.

I have a spreadsheet that gets updated everytime I stop to get gas. You can definitely see the trends, for example I get a 3-4 MPG better numbers in the warmer months than in winter. It can also help spot something wrong with the car if you are suddenly experiencing a reduction in efficiency.

[Mikila]

Yes, it is miles driven divided by gallons.  You do this by:
1. Filling your tank.  ( and note how full- 1st click? 2nd? )
2.  Reset your trip meter or note your mileage.
3.  Drive at least a half tank' s worth of miles to get an average.
4.  Fill up to the approximate same level of fullness as before.
5. Divide miles driven by gallons. 

[BudgetSlasher]

What numbers is everyone using for actual? I ask because when I have attempted to track (usually a specific trip, not a cars lifetime) I've noticed across multiple cars that the in car computer is 2-5 MPG higher than miles traveled divided by gallons put into tank.

The only way of doing this reliably is what you outlined: divide mileage by gallons pumped. Instant MPG reading from the car computer is worthless.

I have a spreadsheet that gets updated everytime I stop to get gas. You can definitely see the trends, for example I get a 3-4 MPG better numbers in the warmer months than in winter. It can also help spot something wrong with the car if you are suddenly experiencing a reduction in efficiency.

On the warmer vs colder months: in my state there is a change in fuel types (dependent on county). The specific fuel types has change, but there are different formulas at different times of year to control for ozone. . . as of this year our seven southern counties are in the federal reformulated gasoline program (winter/summer blends).

[VladTheImpaler]

I average 46-49 MPG in my '08 Prius.
Last month I learned how to "glide" it under 40 mph, so numbers should improve.

[SoccerLounge]

Fun thread!

Unless I am in a hurry, I strategically drive to minimize fuel usage.

2002 Toyota Echo manual. EPA MPG 29/37/32. My mpg was 45 all day long.  Never got less than 40, and maximum was 59 at a steady 65 mph, no a.c. and no stops.

That's that Echo performance right there! The echo is a true sleeper in the world of fuel-sippers. Many people don't realize how fuel-efficient they were. If I hadn't found an Insight for the right price, I'd have picked up an Echo.

[etotheix]

2012 Manual Impreza Hatch: 25/34/28
I do average 28 MPG, but my commute is significantly more highway than city driving.

1929 Manual Ford A Pickup: EPA Didn't Exist
I average about 17 MPG on 100+ mile trips cruising at 45-50 MPH.  I have never bothered to check my fuel mileage around town.

[neo von retorch]

What numbers is everyone using for actual? I ask because when I have attempted to track (usually a specific trip, not a cars lifetime) I've noticed across multiple cars that the in car computer is 2-5 MPG higher than miles traveled divided by gallons put into tank.

That being said when I have tracked a trip I usually beat the EPA rating . . . unless I ended up hitting a major city's ~3-hour rush hour traffic.

I kept a notepad in my glovebox, and I'd track my odometer, and always fill the tank and log the gallons. Then I started transferring this information to Excel. I'd just take the receipt and write the odo or trip odo (which I reset with each fill) on that receipt. Then I transferred everything to fuelly.com. Now, I take a picture of the gas pump and a picture of my trip odo, and later I copy it to Fuelly. (No more paper harmed in this process.)

[ketchup]

What numbers is everyone using for actual? I ask because when I have attempted to track (usually a specific trip, not a cars lifetime) I've noticed across multiple cars that the in car computer is 2-5 MPG higher than miles traveled divided by gallons put into tank.

In the past I kept a little memo pad in the glovebox of the car and write down each fill-up with odometer readings and do the math.  I've since graduated to doing the same but in a Google Docs spreadsheet.

I, too, have found in-car MPG output to be a bit optimistic.  Probably because people would be upset when their 35mpg car gets only 5mpg when you start it, idle for 10 minutes, and then drive half a mile.

[Laserjet3051]

For the past 16 years I've driven a  1999 Jeep Cherokee 4.0L, M/T, 4x4 and get 17.5 mpg combined highway/city and ~19-20 mpg highway.  EPA lists this car at 17.0 combined and 20 mpg highway. Despite my continuous attempts at hypermiling, I'm dead on the EPA estimate.

[SoccerLounge]

I, too, have found in-car MPG output to be a bit optimistic.  Probably because people would be upset when their 35mpg car gets only 5mpg when you start it, idle for 10 minutes, and then drive half a mile.

My current car's onboard MPG is relatively accurate, because the car was intended as a "mileage enthusiast's" car. As such, the MPGs can be pretty harsh; the car really doesn't sugar-coat it if you drive like an idiot. To go with seeing 85ish MPG on the highway (and higher if it stays in lean burn mode for extended periods), I've seen almost SUV-like mileages around town if I had short trips up big hills, I was in a rush, and the battery assist was off, whether disabled or due to low battery. The car is pretty uncompromising about showing you just how badly you missed the point of owning it, if it has to!

(I think one of the reasons a lot of cars only have a totalizing MPG vs. a continual instant MPG readout is precisely what you cite, too.)

[johnny847]

Relevant link

They're not EPA estimates, they're car manufacturer claims of EPA procedure compliant testing.

The EPA establishes the tests that yield the fuel economy figures, but for the most part it doesn't conduct the tests itself. It doesn't have the budget, equipment or manpower to test the hundreds of individual models with unique engine and transmission combinations that automakers produce each year.

Instead, the agency gives its test protocols to the auto companies and lets each test its own cars and trucks. It accepts as true the "EPA estimated" fuel-efficiency numbers each car company submits. To keep the industry honest, the agency runs scores of spot checks each year.

[Crusader]

I drive a 2011 Mazda 6. It stays well within the EPA estimate of 25/30. So everytime I crunch the numbers at the pump it is within that range.

[clarkfan1979]

My 2003 Pontiac Vibe (automatic) has an estimated MGP of 22/29.

I drive about 80-90% highway in Colorado during the summer. My summer tanks average 32.5 mph. I use the cruise control, but I don't really draft people. If I drafted semi-trucks in the slow lane I could easily average 40 mph per tank.

I was driving our 22 ft. RV and my wife drove behind me with the Pontiac Vibe on a 55 mile trip. She was drafting me, but not on purpose. Our ultra-gauge showed 47 mph for that short trip.

[NoStacheOhio]

I don't have exact numbers, but my experience:

13 Civic Si 32-35 actual combined vs. 22/31 EPA
14 Mazda3i 39-45 actual combined vs. 30/40 EPA

Both 6MT

On the other hand I can't get my wife's 14 CRV (5AT) over 30 mpg no matter what I do.

[kitkat]

Just wanted to highlight a new DOE project will hopefully be the next-generation of EPA mileage estimates. Essentially you let the app run for however long you'd like (a month, say) and it tracks YOUR driving behavior to tell you what YOU would spend on a variety of combustion and electric vehicles.

http://www.popsci.com/virtual-test-drive-for-green-cars

They are using very detailed powertrain models (Saxena is a mechanical engineer) for over 5000 vehicles, and collect 1-second interval GPS data from your phone as you drive to calculate speed/acceleration and elevation (through mapping your location to terrain maps). It is not live yet, but their results so far are amazing. They walked me through some plots yesterday that show that based on their field tests (75k miles of driving on a variety of vehicles so far) they can model fuel consumption within 5% accuracy. In comparison, they've found the EPA estimates to be off by +/- 50% much of the time. Another huge benefit is showing drivers hypothetical SOC of a variety of EVs so they can understand what sort of range they would need to meet their driving needs (its always smaller than you'd think!).

Not sure when they will be able to release to the public. The beta version is beautiful, but they need to scale up the computational resources and funding is *slim*.

[Kriegsspiel]

2006 Scion XA - EPA 30 - I got ~48 (mixed highway/city)
2008 Subaru Forester - EPA 20 - I got 32 (mixed highway/city/off-road)