Image From the 4G press event (MPG and HP)

@fuzzy1 I think spwolf in this thread gives an answer to the riddle (at least partial) quoting from edmonds.com review of Gen 4:

"Official EPA fuel economy numbers are still forthcoming, but Toyota's own projections rate the new Prius at 52 mpg combined (54 city/50 highway), a 2-mpg improvement over the outgoing model. Note that the 2016 Prius' numbers were divined in adherence of the EPA's more rigorous testing guidelines that take effect for 2017 models."

So, it seems that the forthcoming official EPA ratings for Gen 4 will be measured by more strict ruler than Gen 3 and as such, much easier to beat, and the difference between the two is more than 2 MPG if measured with the same ruler.

 

And yet nobody has identified what these new mysterious EPA rules are other than the more detailed and specific standardized testing guidelines that I posted yesterday.

 

Yes. Toyota probably knows or taking extra cautions.

 

The EPA has always done random audits. Their budget let's them do about 10% of new vehicles. Some retests aren't so random, and the new Prius will likely be done, as was likely the new Volt. Mostly, preliminary MPG figures just means the EPA hasn't reviewed and signed off on the manufacturer's results.

These expected numbers are likely the low end of what Toyota got. So they can get another headline with slightly higher ones after the EPA gives their stamp of approval.

Easier to beat for those willing to try. The Fuelly average for the Prius tends to be lower lower than the EPA combined. Perhaps it will now be at the combined figure.

The 'road load' test going from being done at 50mph to a range of speeds, up to 70mph seems to be the biggest change. Of course we don't know what those range of speeds are, and if the average is taken, or if the dynamometer will adjust its resistance with the vehicle speed during a test cycle.

 

For reference, the Canadian testing regimen is explain in some detail here. For 2015 onward they've adopted a tougher "5-cycle" test, which yields numbers close to reality:

Fuel Consumption Testing | Natural Resources Canada

 

^^ The U.S. EPA scale has been 5-cycle since 2008. Though I believe there is also an adjusted 2-cycle option, at least for legacy designs.

 

When I first noticed ads with the new rating method, I thought "hey, I could actually get close to those numbers". The old numbers were absurd to the point of being meaningless.

 

Canada uses the EPA test cycles, and are now adopting the changes from 2008.
Detailed Test Information

It saves costs. Manufacturers only have to do a single test for both countries, and Canada didn't have to develop their own test.

 

This article from GreenCarReports says the opposite -- that load tests are changing from 70 mph to 50 mph.

EPA Issues New Guidelines For Automaker Fuel-Economy Testing

And, indeed, the new guidelines talk about coast down at 50 mph.

http://iaspub.epa.gov/otaqpub/display_file.jsp?docid=34102&flag=1

I just found this and haven't read it carefully but it appears to be the official regulations for EPA testing as of 2012:

http://www.gpo.gov/fdsys/pkg/CFR-2012-title40-vol34/pdf/CFR-2012-title40-vol34-part1066.pdf

 

Yes, EPA testing language spells out previously the authority to do random audits, but these don't usually sound random when we hear about them. Rather when buyers are groaning about unachievable EPA numbers, the EPA seems to come knocking. Maybe we just don't hear anything in the news when random audits are done and the conclusions were that all was on the up-and-up.

But if you see the articles Jeff N linked to, the EPA seems to be saying that starting this January they will finally press forward more aggressively with true random audits.

Here's another article I found:

http://www.nytimes.com/2015/02/24/business/energy-environment/epa-issues-stiffer-rules-on-vehicle-fuel-ratings.html

 

"The new rules apply to model-year 2017 vehicles, which go on sale next year."

what am I missing here?

 

The Gen 4 Prius looks like it is labeled by Toyota as MY-2016, but goes on sale for the first time in the U.S. in January, 2016. But it looks like the EPA considers all 2016 released models MY-2017 by convention.

 

We didn't here about them, because most of the time there isn't any cheating.

 

With no independent checking, how would you know?

 

How is this for a conspiracy theory:

Toyota releases conservative test figures to the EPA not just for the more challenging test cycle, but because the Plug-in version will be here in a year. This would give Toyota extra wiggle-room to ensure the Plug-in meets or exceeds either the core or "Eco" model mpg.

Let me adjust my aluminum hat and see if I can divine some more.

 

The EPA audits are the independent checks.
When one of the 10% that they check a year is found to not have cheated, it isn't news.