In a recent Twitter post by Edmunds, the company said Tesla’s vehicles consistently fail to achieve EPA range estimates, while EVs from other manufacturers often exceed them.
The fact is every single Tesla we’ve tested has failed to hit its EPA range estimate: https://t.co/6ze4lqVBIy
— Edmunds (@edmunds) February 10, 2021
EPA efficiency estimates have always been a bone of contention in the automotive world. For MPG tests on gas cars, most people don’t achieve the EPA mileage. This has led car shoppers to conclude that the EPA estimate is unrealistic, or not “real world” enough to be of use to buyers except when comparing cars. It’s the old “YMMV” (Your Mileage May Vary).
Some brands have always done a little better than the others, though. If you google “EPA MPG Suit,” you’ll find various examples of customers who felt cheated by the range figures when vehicles fell far short of what was advertised. Volkswagens, on the other hand, has a reputation for getting around or a little better than the EPA estimates.
Why EPA estimates can be so good or bad is a complicated issue.
First off, yes, the EPA’s testing cycles are unrealistic. A testing cycle is a testing procedure of what speeds a vehicle needs to go over what time periods to complete the test. City tests have lots of stop-and-go and variable speeds. Highway tests have less variation and higher speeds, but they’re still pretty low at 50–60 MPH. Any vehicle is going to get better mileage at 55 MPH, but few drivers will actually go that speed.
Second, most of this testing is done by the manufacturer. Very little of it is done on tracks, though. They put a test car with a professional test driver on a dynamometer, or “dyno,” which is rolling wheels that the vehicle’s wheels sit on and pretend to drive while not moving. Basically, it’s a treadmill for cars. EPA-approved formulas are then used to add back in things like air resistance to make it more realistic.
The EPA has even accused some manufacturers of using drivers who are “too good” at running a car through the EPA test. The drivers go through the process very carefully to make sure the vehicle gets a really good MPG or range rating, and don’t drive through the cycle realistically.
This issue is compounded by just how terrible most drivers are at driving efficiently. Acceleration is the obvious culprit people blame for bad mileage (“heavy right foot”), but most people waste their gas when it comes to braking. Even if you take off like a grandma, you are still going to get poor mileage if you keep your foot gently on the accelerator until the very last second and brake hard at a red light. You can improve mileage quite a bit by just coasting up to stops as much as possible, and most people just don’t do that.
Between all of these things, there’s a lot of room for mileage to vary. Take a manufacturer that drives the cycles very carefully and then sell the car to a moron and the EPA mileage will look very high compared to the idiot’s “real world.” Take a manufacturer who comes up with more realistic numbers and sell the vehicle to a hypermiler and you’ll see the driver get far better numbers than the EPA says. Most of us lie somewhere in between.
Another thing to keep in mind is that most manufacturers aren’t hard up to get the best EPA range on their EVs. Yes, they get an eMPG rating that affects their fleet rating, but it’s already so high compared to a manufacturer’s gas offerings that eking out an extra few percent isn’t going to make a big difference in the grand scheme of fleet fuel economy. Simply being electric gives so many other emissions and tax advantages that even EVs that are inefficient turds compared to others won’t hurt the manufacturer much.
Tesla is unique in that it doesn’t offer any gas cars. It never struggles to meet fleet CAFE standards, and it sells other manufacturers a lot of credits. It would be a little odd for Tesla to be concerned with pushing range numbers higher with any kind of cheating because there’s just not much motivation to do so beyond possibly marketing. But even on marketing, would it be worth upsetting buyers just for a few more miles? Probably not.
Edmunds doesn’t use EPA test cycles when doing its testing. It takes actual road tests in the Los Angeles metro area, aiming for 60% city, 40% highway. Edmunds tells readers that it uses a vehicle’s most efficient settings, follows the speed limit within 5 MPH, and makes as much use of regenerative braking as possible. It also says, “Since no electric vehicle has exactly the same range, the route length is adapted to suit each vehicle.”
In other words, vehicles with longer ranges will drive a longer and different route than vehicles with shorter ranges.
At the bottom of the page, they make it clear that they don’t think their testing is more accurate than the EPA’s test cycles. There are simply too many factors that can affect EV range for any test to give someone an idea of what their “real world” is going to look like. Different drivers, different climates, and different cities all affect things, and Edmunds or anyone else can’t factor that all in.
With all of this in mind, let’s look at the numbers Edmunds presents.
Nobody else’s vehicles on the list have the range that most of Tesla’s vehicles do, with the exception of the standard range Model 3. None of the other manufacturers’ vehicles on the list exceed 300 miles. Thus, all but one of the Teslas got a different test than the other vehicles, because Edmunds gives them a longer test route. There’s not a correlation I could see that would prove that this affects the testing, but it’s something to consider.
The best vehicle on the list for beating EPA estimates was the Porsche Taycan. This is unsurprising because the Taycan has a two-speed transmission in the rear. From what I know, none of the other EVs on the list have a multi-speed gearbox. By keeping electric motor RPMs lower at highway speed, they can keep it in its most efficient powerband more often. EPA’s testing doesn’t go as fast as the speed limits on Los Angeles freeways, so this feature made a bigger difference in Edmunds’ tests.
Tesla’s poor showing may also be explained by test drivers. It’s possible that Tesla is using better drivers to finesse the test a bit. They’re not being dishonest, because they are legally required to get through the test the same as any other manufacturer, but a skilled driver can get better EPA numbers than a less skilled driver. This is especially true if you tell them to go for the max.
We don’t know whether Tesla is doing this, and it probably doesn’t matter unless you are an inefficient driver and you need to squeak out every last mile of range to get somewhere. Even then, Tesla’s route planner is quite good at telling you whether you’ll make it, so you’d be a fool to go get yourself stranded somewhere because the car didn’t make EPA range.
As I pointed out above, there’s not much financial incentive for Tesla to finesse the numbers with super careful driving, so this is probably not the case.
Things To Keep In Mind When Doing Research
If you’re researching a new car, especially an EV, you need to keep some things in mind.
First, avoid relying on the EPA’s figures except for comparing EVs. You’re unlikely to get exactly what they predict, and you’ll be disappointed if you let that give you wrong expectations. Even as a comparison tool, it can only give you a rough idea.
You’ll probably want to find several different efficiency tests to get a good idea of what you could realistically expect. With each one, look at the testing methodology to see how closely it matches your driving. If you live in the LA area, Edmunds is probably going to be pretty close. If you live in West Texas where the speed limit is 80 MPH, don’t expect either the EPA or Edmunds to give you a good estimate.
Finally, take advantage of tools like A Better Route Planner. In some places, their speed limit data could use some help, but for the most part you get a pretty good idea of what to expect with a vehicle if it’s been out a bit. By virtually testing a vehicle on YOUR routes that you drive, you’ll get the best estimate you can get short of testing for yourself.
In fact, self testing is probably as good as it gets. See if a dealer can let you use the car for a few days and see how it works for you. Nobody but you can do that.