Originally Posted by rdclark

First, let me repeat that the EPA does not do the testing. The manufacturers do, using the EPA methodology (and the EPA then double-checks 10-15% of the results, according to the site you linked).

Second, let me again repeat that the point here is not to compare EPA results with CU results.

The point is to compare CU results to other CU results. For example, no pre-2008 xB was ever tested using the new EPA method, so comparing EPA '08 results with EPA '05 results is just guesswork. *

But CU tested the 08 xB and the 05 xB using exactly the same methods, the same methods they use for every vehicle they test. These are a mix of road, test track, and lab tests. So you can reliably compare the results of an 08xB with an 05xB or an 04 Civic or whatever cars CU has tested, large or small.

This does not mean you will get the same efficiency CU got, obviously. It just means that the two cars got the stated results when subjected to the same tests.

Gen1 owners had the same reaction to CU's tests that xB2 orners are having: "their results are whack, my car is much [faster, more efficient, more comfortable, less noisy, whatever] than they said."

That's the sound of the point being missed. They're comparing cars to each other using common testing metrics, in order to determine relative performance.

They are not comparing their cars to our cars, or their driving techniques to our driving techniques, or their MPG results to the EPA's.

Bottom line: if CU's tests show that a given car's overall MPG is, say, 20% lower than another's, then it probably will be. A careful driver will probably get better numbers, and a leadfoot will probably get worse, but the relative efficiency of the two vehicles with the same driver under the same conditions will remain similar. That is the information the test is designed to yield.

*Yes, I know the EPA posted numbers estimating what older cars would have achieved in the revised test. Key word: estimate. They did not re-test the old cars. They just reduced the old results by a sliding percentage scale based on size, weight, and engine displacement. IOW, guesswork. CU's data through the years, on the other hand, is based on consistent and repeatable testing

Originally Posted by danos_XBox

I just hate to have people read this bad report and turn away from a great car

Originally Posted by rdclark

For the record, CU's results for the four cars they tested for the current report:

Elantra xB Lancer Sentra

City/Hwy/Overall 18/36/27 16/30/23 17/34/25 18/34/26

They've of course tested most other small cars on the market using the same methods, but I don't have back issues at hand.

The Elantra and xB were automatics. The Lancer and Sentra had CVTs.

The xB was the heaviest of the four cars, and also easily the fastest in all acceleration tests.

I don't see why these results would surprise anyone, taken in the context of CU's overall testing methodology.

Originally Posted by toronado

Exactly. I understand the CR MPG results to mean that the xB will get 2-4 MPG worse overall than those other three cars. But will also out-accelerate those other three cars. That seems like a perfectly understandable and reasonable report.

Regarding the EPA tests...
Didn't the EPA just start doing actual tests? Weren't the previous (pre-200 tests just theoretical tests based on emissions measurements and not actual road-course drving?