After reading time after time that people are dissapointed with their "real World MPG" compared with the official estimates, I thought I'd do a little homework and discover the real difference. I found 2 popular consumer companies that have either done their own tests by driving the actual cars on a route with the latest high-tech equipment rigged up, or have had the general public submit over 60,000 entries of their MPG and compiled the results.
I was actually quite surprised, despite what you might think, the cars in question get an average of 86% of the somewhat manipulated lab figures. The Top Ten all got 58 MPG or above, and before I get accused of miss leading people, that's UK MPG folks!
A lot of cars even get slightly higher than the average figures.
I found 2 popular consumer companies that have either done their own tests by driving the actual cars on a route with the latest high-tech equipment rigged up, or have had the general public submit over 60,000 entries of their MPG and compiled the results.
Well neither of those take into account the variances in the driver.
Put it on a dyno and report the consumption at say a steady 50, 100 and 140 KPH.
Then all you have is the drag coefficient to take into account, but I'd sooner them just report the numbers without it since all the math in the world won't give you an accurate answer.
I think that's why they did the drives in real World conditions, different traffic, weather etc. In the labs, the EU test cycle is a tad silly, the room is between 20 and 30 celcius, and the average speed for the entire test is just 29 MPH yes 29! A tad ridiculous when you think people travel between 70 and 80 on the highway. It's true there are many variations and factors that will affect the results, but getting the car to actualy drive on a road, by an actual human is a good start!
The U.S. EPA tests are the same way - they run through a specific "course" and give the MPG ratings based on the results. They updated it a few years ago to behave more like current driving, but there's still a lot of room for variance in individual results.
In what I assume is meant to prevent misleading consumers, it's actually illegal to advertise MPG numbers other than the official EPA ratings. There was one person who was talking about suing Toyota because she didn't personally get the EPA numbers in her own Prius. Assuming the model was properly rated, those numbers are the only thing Toyota is allowed to put on the window sticker. I'm not in support of misleading or irrelevant (or flat-out fraudulent) ratings, but I'm also against people being stupid. If you don't understand when buying a new car that the ratings on the sticker are for a specific test, and not a guarantee of exactly what every person will get in regular usage (combined with the fact that they aren't allowed to put any other numbers on it), you deserve what you get. Sorry, I'm a jerk that way.
Rather than suing car companies, we need to be complaining that the government testing isn't relevant. I think I'd rather see a few ratings at different speeds and maybe something stop and go, as opposed to the current system of trying to map out an average drive. Again, you're never going to get a rating that perfectly matches everyone's driving experience, but if I know what to expect when cruising at 55 or 70, or when going 25 with a stop every block, I think I'd be lot less likely to be unhappy with the ratings. I could estimate a mile of stop & go at each end of my commute plus 10 miles of highway in the middle, and come up with a pretty good idea of what my overall gas usage would be.
As I have said before, the EPA test is NOT designed to measure fuel economy. It is designed to make sure the vehicle meets emission standards. Fuel economy measurements are a byproduct of emissions testing. I don't know about the testing in other countries, but it most likely is the same except for using CO2 emissions for tax purposes.
There has been a similar story Here with a lady who bought a Golf TDI, advertised as getting 80+ MPG, she was only averaging 65 ish and threatened to sue VW. and then I read that someone else went on a 995 mile road trip in the same car and averaged 97 MPG! There's a lot of conflicting data out there.
But this is becoming an increasing concern for motorists now, I have read that the EU test cycle will be changed to obtain realistic results. Charon is right, the tests actually measure C02 and then the MPG's are calculated using these figures. Hybrids seem to get even more ridiculous results, with some getting just 30% in the real World of the claimed results! The Chevy Volt sold as the Vauxhall Ampera here, has now stopped being sold in the UK due to poor sales, I'm guessing one of the reasons is this, claimed to get 235 MPG but averages 70 ish, big difference.
In relation to plug in hybrids, the EU uses a test and reporting method in which the car does the test for a certain distance, starting out with a fully charged pack. Then the fuel efficiency is calculated for that distance while ignoring the electricity consumed from the wall. It is the same problem with entering PHVs into Fuelly.
GM released a 230mpg figure during Volt development derived from a similiar proposed EPA test method. It was never used officially, because it was too dependent on the users trip and gave no indication of electric consumption. After the first year PHVs were available, the combined mpge for gas and electirc was dropped from the window sticker, and just the separate consumption figures while using electric and while using gasoline were reported.
Poor sales of the Ampera/Volt, and also the Prius PHV, are mostly due to their high price in the UK. The Volt is being cancelled because of those sales, but there may be other factors. The new Volt comes out next year, and it is also excluded from Thailand and Australia. GM may not want, or can't, produce a RHD model at this time.
The Volt is an interesting paradox in assigning an FE figure. Most people whose round-trip commutes are within te battery pack's range can see 100+mpg, but it's mileage tanks on roadtrips. That makes it exceptionally hard to put a number on the car with any accuracy whatsoever.
Matt Farah, a brash young autojourno, who owns a Corvette and typically samples high-end sports and luxury machines, actually leased a Volt. He, as you can imagine, doesn't exactly drive for high FE and averaged 115mpg over his first several months of driving the car, according to his ELR vs. Volt test.
The case of PHVs with their varying EV ranges really makes assigning some value that is a mix of grid and gasoline power impossible. The variance between individuals' daily drives has a larger impact than even with a non-plug in cars.
The best that can be done is to do what the EPA has settled on; report electric economy on the grid supplied electric, the EV range, and the fuel economy in hybrid mode. That's all the info the individual needs to figure out what to expect from a model with their drive. There is likely online calculators to help with it.
It is nice when the posted test numbers are close to what people get in the real world, but that isn't actually the important part. The important part is that it is done with a standardized test under standardized conditions. This allows a level field for comparing the fuel efficiency between different models. reported numbers from magazines and an individual are near worthless because of the variables that likely changed between the testing of two different models.
I think showing separate electric range vs gas mpg is the way to classify PHV's. I really love the idea of PHV, each potential buyer of a PHV would have to look at their own individual commute to determine if it would make good sense for them. The Chevy Volt has such a large range electric only that most drivers would never use the gas engine. If I were in the market for a new car, I would without a doubt go for a PHV.