# Electric-Car Efficiency: Forget MPGe, It Should Be Miles/kWh

2011 Nissan Leaf plugged into an EVgo quick-charging station, Texas

Enlarge Photo

## Related Photo Galleries

See more photos »

Two days ago, we ran down all the rated ranges of every plug-in electric car offered for sale in the U.S. market this year.

We did not include ratings for efficiency.

In part, that's because the unit the EPA uses doesn't actually have much to do with the electricity owners use to recharge the battery packs that power their electric cars.

Instead, the EPA has chosen to use an artificially created metric called Miles Per Gallon Equivalent, or MPGe.

Setting aside the issue that when a plug-in car is traveling on battery power, there are no gallons of gasoline involved anywhere in the process, here's how it works.

The efficiency rating measures how far an electric car can travel on the amount of electricity contained in 1 gallon of gasoline (which, for the record, it says is 33.7 kilowatt-hours).

Note that the gallon of gasoline costs \$3.50 to \$4.75--while the equivalent energy content delivered over the electric grid to your car costs from \$1 to \$8, depending on your local electricity rates.

But it's impossible to buy a MPGe, and so the efficiency measure is useless for actually calculating the cost of operating an electric car.

Instead, electric-car drivers buy kilowatt-hours of electricity, most of which goes to power their cars (plus some for overhead like keeping the pack at the optimal temperature, in many cars).

Of course, the non-linear Miles-Per-Gallon measure isn't very helpful in calculating how much it'll cost to go a given distance.

To get the cost of going 100 miles, for instance, you have to divide by gas mileage (whether it's 10 mpg, 25 mpg, or 50 mpg) to get the number of gallons, then multiply that by the cost of gas.

So why not measure an electric car's efficiency in what the owner actually buys?

If you like the miles-per-gallon equivalent, we'd suggest Miles Per Kilowatt-Hour, or MPKwh.

2012 Toyota Prius Plug-In Hybrid - production model

Enlarge Photo

If you live outside the U.S., however, you don't use MPG at all. Instead, you use a measure of consumption: the gasoline used to travel a given distance.

That's usually liters per 100 km or, for non-metric North Americans, gallons per 100 miles.

To parallel that, we'd suggest kilowatt-hours per 100 miles, which can then be multiplied by your electricity cost to get the total cost of running your car for some useful distance.

For a 2012 Nissan Leaf, its average rated efficiency of 99 MPGe translates to 34 kilowatt-hours per 100 miles. Just multiply that by your electric cost.

So if you pay the U.S. average of 12 cents/kWh, the Leaf will cost you \$4.08 to go 100 miles (versus \$16 in a 25-mpg car with gas at \$4/gallon).

Tell us what you think is the most useful way to measure the efficiency and running costs of a plug-in electric car: Does MPGe work for you? Or would you prefer a measure using kilowatt-hours?

+++++++++++

Posted in:

### Have an opinion?

• Notify me when there are more comments
1. Excellent idea. I am not sure what the EPA was smoking.

I think it is late in the day to make Kwh/100mi. The Europeans will do Kwh/100km so then we are all set.

2. kwh/mi makes more sense to me than anything else. Its easy to calculate your cost, your hypothetical range, and compare to an gas burning car. kwh/100mi just add one more (albeit easy) step for each of these.

3. If you really want to change it - it should be Wh/mile (or kWh/100-mi as Tim Meyer suggests), not mi/kWh or MPGe.

4. Absolutely, yes.
To allow for easy comparisons between vehicles using different fuels or energy sources however, two other numbers would be useful:

1) Fuel cost per mile or other distance, in actual cents/dollars. This, everyone understands.
Unfortunately this will change a bit every year, or even between areas.

2) Scientific measure of the energy used per standard unit of distance => (kilo)joules per meter.

E.g: a Nissan Leaf sips very little energy: ~0.7 kJ/m.
A Hummer H2 burns over 10x more, or ~8.4 kJ/m.

5. I've (independently) proposed 2) above in a later post - but J/m can be simplified to N (Newtons), a force. And extending Just O's point in 1), I do think we should take the opportunity to move away from measures that vary with time and place - in particular, avoid 'gallon' because of its variable definition; for liquid fuels, mass is better related to energy content, but if it has to be volume, then at least use litres (which is at least the same size as liters!)

6. I disagree that we should use cost per mile only because it varies constantly with utility/fuel costs. It would be misleading to say the Model S has a zero cost per mile (free superchargers), as opposed to saying it gets so many miles per KWH.

7. MGPe is good for comparison to MPG liquid fueled vehicles but for real life driving a kWh burn rate is more practical. Knowing approximate kWh burn rates of accessories is also useful knowledge.

In addition to EPA testing/publishing charging rates for electric vehicles, it would be educational if they rated ReGen* efficiency (& max rate).

* ReGen is main reason Hybrids have higher MPG that legacy gas vehicles, but there is no way to separate motor effiency from braking regen effiency. Differences between EPA cycle tests & real-world MPG values are becoming more evident.

8. The EPA uses both units--the author really needs to look at a fuel economy label for a PHEV or EV at fueleconomy.gov. So there's that.

The reason why the MPGe unit is useful is that without it, a prospective buyer cannot compare the efficiencies of a conventional or hybrid vehicle with a PHEV or EV or NGV or FCV. As more vehicles with different fuels are added to the market, it will be more and more difficult to compare across the various technologies. I can see why people would want to know the miles/kWh so that they can easily calculate costs but that is why having both is such a good idea.

Having the mpg of a car "isn't very helpful" in calculating distance? It's a two-step calculation, so that's hardly what I would call unhelpful!

9. How fast are the electric cars going when US Govt measures how many miles per Kwh? With ICE cars you've got City and Hwy numbers so you can kind of extrapolate. I'd like to know which BEV's do the best at 75 mph.

10. Looking at city vs. highway test graphs provides insight into speeds & amount of stop/go for each EPA test cycle.
http://www.fueleconomy.gov/feg/fe_test_schedules.shtml

The Highway test peaks at 60mph, and Highspeed test peaks at 70mph. BUT note for both Highway & Highspeed cycles, the average is ~48mph!

I wish all manufactures (EPA) provided a graph of Energy Use vs. Constant Speed, like Tesla posted for Model S: http://www.teslamotors.com/blog/model-s-efficiency-and-range (note: graph uses Range for fixed 85 kWh, but can be easily relabeled to display kWh vs. Const. Speed). The graph clearly shows the accusative effect of drag vs. speed^2 above 50mph. (mosly air & wheel resistance)

11. We have a Leaf and a 2005 Prius. To convey the cost savings of the Leaf I just tell people that the Leaf is three times cheaper to run than the Prius or the equivalent of 140mpg. I spent just under \$200 to drive my Leaf for one year.

On the sticker for an EV, they could have a silhouette of a Toyota Prius with only 1/3 of it blacked out, the rest grey. This would say that this EV will cost only 1/3rd as much to operate than the benchmark Toyota Prius. There, problem solved:)

12. I agree MPGe is insane, and irrelevant.

miles/kwh is the most familiar and useful, and the closest analogy to the already familiar MPG.

- Bigger number is better, same as MPG.
- Concept is the same: How far can I go on a unit of fuel?
- The numbers are currently in the range of 2-4, so the they are easy to remember and the math is simple.

I'm all for mi/kwh...

13. I think miles per KWh is the best way to show how efficient an EV is.

Now, let us be very carful with it. Apparently, Honda and Toyota have figured out a way to CHEAT the game. Prius Plugin and Accord Plugin seem to have some really high efficiency in its EV mode. That is b/c they limit their EV mode to lower power and lower speed. So, it cheats the number by limited the EV mode to the "low usage" mode and then public that number. But in reality, all other BEVs/PHEV/EREV can get even better number than that if they limit their cars to the same mode.

So, buyers beware of the game. EPA should extend the test to 40 miles in length and then increase the top speed to 70mph...

14. I personally find mi/Kwh useful when considering how my my car is performing, but this measure is useless when comparing cars with an ICE driver; kWh is not a unit that most people grasp intuitively. I usually convert to cost per mile. People can quickly divide the cost per gallon they pay by their car's mileage. I especially like this metric because I pay just over a penny a mile to charge off-peak. That delivers the electric car message very nicely.

15. Bravo John.

As a two car EV family,
I always use mpkwh or sometimes shortened to mpk. MPGe has too many variables.
Then you can directly compare to gasoline and MPG as I have done in this post.

http://electric-bmw.blogspot.com/2013/02/the-cost-of-fuel-for-two-cars.html

16. First, it can't be MPKwh. That's because it is not Kwh. It's kWh. The W is capitalized after the originator of the figure of measurement - Watt.

So MPKWhor MpkWh or MPKWH but not MPKwh.

Second, there is nothing non-linear about it and it is an easy conversion.

MPGe/33.7 IS MPKWH - there being by the EPA definition 33.7 kWh per gallon as pointed out.

We don't actually find MPKWHr particularly useful. We like Watt hours per Mile. And so if you have 95 MPGe = 95/33.7 or 2.82 miles per kWh. This works out to 355 Wh/M.

This is fun because 10x Wh per mile roughly give you to the weight of the car.
355 Wh/M = 3550 lb car. If it is a HEAVIER car, it is quite efficient. If lighter, not so efficient.

Jack Rickard
http://www.EVtv.me

17. I'd also like to suggest that the battery information provided for a new EV should primarily show available energy, (not 100% - to 0% discharge that isn't allowed or recommended). For instance, I believe the Volt advertises a 16kwh pack but only allows 10.3kwh of use. Both numbers are important to the buyer.

18. kWh per 100 miles is mixing metric and english measurements. Wh per mile seems to be the simplest all english unit. EVs use between 150Wh and 300Wh per mile, with lower numbers better than higher numbers.

19. Energy per unit distance boils down to force, an effective thrust. So if we're going for a change, I would vote for Newtons, N, as the measure of efficiency. Maybe strange to some, but would soon become familiar. Incidently, given the energy content of fuels, the same measure would work for all vehicles, and the relative inefficiency of ic engines would not make for a pretty comparison! (1kWh/100miles = 22.37N very nearly.)

20. Our 2012 Leaf gets 4.5 miles per KWH. Our KWH cost is .16/KWH which calculates to a cost of \$3.55/100 miles.
We received our 2013 Tesla Models S late week. Both cars and our home is powered by a 48 panel solar array producing more than 19,000 KWH/Year. We have 11,000 miles on the Leaf at an estimated cost savings of \$2,000.00 in fuel.

21. The best argument for using MPGe is that most people are familiar with MPG and won't have a feel for MPkWh until there are a lot more EV's on the road.

Why does it have to be an either or choice?

2013 Leaf 130/City, 102/Highway, for 116 MPGe Combined

is a lot more meaningful than:

2013 Leaf 3.86/City, 3.02/Highway, for 3.44 MPkWh Combined

It just doesn't have the same ring...

Neil

22. I meant to say that I prefer Wh/mile where lower is better.

In my estimation, the most important factor for lowering the consumption of an EV is lowering the aerodynamic drag. The Leaf and the EV1 are pretty close to the same weight (3,354 vs 2,908), but the EV1 uses almost half the Wh/mile (340Wh/mile vs ~160Wh/mile). The EV1 has a much lower Cd (0.19-0.20 vs 0.29) and it has a smaller frontal area.

From this CdA list: http://ecomodder.com/wiki/index.php/Vehicle_Coefficient_of_Drag_List

EV1 CdA is 3.96 sq ft
Leaf CdA is 6.94 sq ft

Most important factor for efficiency is the drivetrain.
Second most important is CdA.
Third most important is weight.
Fourth is rolling resistance and accessory power use, especially heat and A/C.

Neil

23. A component of the drivetrain revolutionizing efficiency is Regenerative Braking. For example 2013 LEAF can now recover a higher percentage of energy; 94 percent compared to 88 percent with the prior model.

24. Where did you get those numbers from?

In my experience, the previous generation Leaf's regen is very "weak", especially when you aren't in ECO mode.

@ 94%, you are telling me that the generator is better than 97% efficient and the inverters/battery is also 97% efficient in recovering and storing those kinetic energy... That is a "far stretch" from reality...

25. I agree. MPGe is a construct trying to fit a round peg in a square hole. M/kWh, or kWh/m, makes much more sense (I prefer the former) and allows the consumer to better predict their costs.

26. kWh per 100 miles is a meaningful figure as something that enables a potential buyer to evaluate how much energy he/she will use. As an energy geek, I wish it was easier to calculate what the primary energy use in kBtu is per 100 miles and compare that with gasoline, but that is more meaningless and convoluted to most folks than MPGe.

I was really impressed by the Tesla Model S I got a chance to check out last week and wish there was a more affordable EV that had the features of the Tesla but in a hatchback form factor like my Ford C-Max. http://www.greenlifestylechanges.com/my-dream-family-car-tesla-model-s/

27. Should be kwats not Gallons but they doing this so people would stick with gas they dont want electric cars revolution, they pretend electric cars dont exists

28. Been saying and writing about it for a long time. MPG and even MPGe makes ZERO sense. We need something more tangible.

29. The EV conversion community tend to use Wh/Mi or Wh/Km. Which is easily converted to kWh/100mi or kWh/100Km for larger packs and longer ranges.

30. We already use Watt-hours/mile or Watt-hours/km.
Please see http://www.evalbum.com/794 as an exemple.
A unit should be unitarian: so many watt-hours for ONE mile or ONE km.

31. I just Googled a cost calculator for BEV's anc came up with this:

http://www.dteenergy.com/residentialCustomers/productsPrograms/electricVehicles/eVCalculator.html

It nailed my actual yearly cost to the dollar which is pretty remarkable. If you had a bar code on the window sticker that told a smart phone what parameters to load, you could just let a smart phone app be the interface to get a clear picture of what your cost savings could be given your current ICE usage per day. Of course, you would need to know how many miles/day or miles/yr you drive (or use defaults). The local gas price and prevailing local kWh electric rate could be derived via GPS in the phone. To us EV geeks, wh/mile makes sese but to the general public, not so much.

32. Cost per mile, backed into via consumption per mile. My tdi Jetta costs 11 cpm, my electric GEM costs 2 cents.

Consumption per 100 miles is a good metric because liquid fuel units end up being easily conceptualized units.

33. Absolutely, mileage should be reported as miles/kwh.

I didn't get serious about buying an EV until I read a blog where someone reported his mileage in miles per kWh. My eyes almost popped out when I saw a real number like that, and I was sold.

People understand mi/kwh. But how does someone connect with a bogus MPGe figure? They don't!

34. I still prefer the admittedly clunky MPGe because one can make direct comparisons with ICE cars. Further, everyone knows what MPG is, while non-savvy people (most everyone) can't equate kWh with MPG.

35. Exactly...

36. I like miles per dollar as an easily grasped metric:

The 2005 Prius:
At 43mpg it costs \$.0837 per mile to drive our 2005 Prius
\$1 divided by .0837 = 11.94 miles per dollar spent

The 2011 Leaf:
Our electricity currently costs \$.10 per kWh
\$1 divided by .10 = 10.0 kWh per dollar
Our Leaf travels approximately 3.8 miles per kWh
So, 10.0 x 3.8 = 38 miles per dollar

The 2011 Nissan Leaf goes more than three times farther per dollar than the 2005 Toyota Prius.

37. Miles per kiloWatt-hour has just as good of a chance of catching on with average drivers as meters per Joule (or Joules per meter) will catch on with EV enthusiasts. Keep in mind miles per kiloWatt-hour is also a bastardized unit. Miles and hours are not SI units. Meters and Joules are SI units.

38. There is one thing I really like about MPGe. It is so much higher than a gas car's MPG (even a hybrid) that it makes people notice plug-ins. If it were a different unit, you could not compare them at a glance.

The thing that I'd like to see is all the PEVs to offer both miles/kWh and Wh/mile data. Today they tend to only have one or the other.

39. \$/mile would equally measure both ICE vehicles and EVs equally.

40. you need easy number 99.99% of general public who drive gas can understand. say \$35 to drive 1,000 miles or cut your fuel cost to 1/6th of the cost to drive a 22mpg.
it must be extremely simple and almost a virtual meme.
difficult as it is to get folks to look at TCO, stress gas price up no matter what the cause. it is a difficult sell until you hit pocketbook

41. The mphkw measurement is the way to go as this tells you what distance you can travel on an understandable format. I have a 2011 Nissan Leaf. The dash shows I get 5.1 mpkwh average at this time. This tells me I am doing okay based on the electricity I use. Calculating the mpkwh average should be relatively correct, compared to the Leaf's ability to calculate the range estimate.

42. Finally, I've been saying mile per kwh for years... eventually, it will change..

MrEnergyCzar

43. Keep the wording simple for me and others. "Watts per mile" (dropping the hour) as it is just a comparison tool for the EV's travel efficiency and the less energy you use the better the EV.

More »

Go!