Thursday was a sad day for journalism thanks to this PolitiFact entry. Angie Drobnic Holan wrote and researched it. Scott Montgomery did the editing. Both dropped the ball.
On to the claim at issue:
"The problem is that, for decades, we have avoided doing what must be done as a nation to turn challenge into opportunity. As a consequence, we import more oil today than we did on 9/11. The 1908 Model T earned better gas mileage than a typical SUV sold in 2008. And even as our economy has been transformed by new forms of technology, our electric grid looks largely the same as it did half a century ago."The Model T earned better gas mileage than a typical SUV sold in 2008. Supposedly.
(U.S. Department of Energy)
Drobnic starts with the obvious, averaging the mileage for SUVs from the 2008 model year (see the afterword for a deeper critique of this aspect of the piece):
(T)he Environmental Protection Agency found that light trucks — the class of vehicle to which SUVs belong — averaged 18.1 miles per gallon for model year 2008. The most efficient SUVs do much better than that — a Jeep Compass gets 23 mpg in the city and 28 on the highway — but Obama said "typical."So is "typical" the average SUV sold or the average for all the distinct SUV models plus other "light duty trucks"? Drobnic apparently opts for the latter and turns to the point of comparison:
To find out what a Model T averaged, we consulted Bob Casey, author of The Model T: A Centennial History , and the curator of transportation at the Henry Ford Museum in Dearborn, Mich. (The museum is named for Ford but independent of the Ford Motor Co.)The wording suggests that Casey made no attempt to estimate how the Model T might have rated using the EPA's methodology. Drobnic should have asked a question focused on that comparison to avoid the old apples and oranges problem. Perhaps that is what she did, but the story gives no good indication of it.He said the best estimate for a Model T's mileage is 20 miles per gallon, though it might be able to get 25 under the right conditions.
And it would not have been a bad idea to consult more than one expert.
The Model T was introduced on Oct. 1, 1908. It had a 20-horsepower, four-cylinder engine, reached a top speed of about 45 miles per hour, got about 13 to 21 miles per gallon of gasoline and weighed 1,200 pounds. It was the ninth of Henry Ford's production cars.
 |
Gas mileage estimates for the Model T vary widely, however. The uncertainty comes from many factors, including low-quality gasoline when the vehicle was introduced, bad roads, and no odometer.
So technically Obama is right.Perhaps Obama is technically right. PolitiFact hasn't made the case.
And now Drobnic allows her readers to again experience the traditional PolitiFact torture of inconsistent distinctions between technical accuracy and the nature of the implicit argument attached to the claim:
But his implication is that we haven't gotten more fuel efficient in 100 years. And that's a reach.Uh, no, it is not a "reach." It is a blatant falsehood. The model T weighed about 1200 lbs. For comparison, one of the poorest-mileage SUVs, the Range Rover, has a gross weight of over 6,000 pounds. Put those 6,000 pounds on the Model T and the number of miles per gallon will drop precipitously--guaranteed.
Avoid keeping unnecessary items in your vehicle, especially heavy ones. An extra 100 pounds in your vehicle could reduce your MPG by up to 2 percent. The reduction is based on the percentage of extra weight relative to the vehicle's weight and affects smaller vehicles more than larger ones.
 |
Just for fun, let's do an estimate using a lowball number. An extra 100 pounds decreases the Model T's mileage by just one percent. Adding a mere 3,000 pounds of the Range Rover's curb weight, we get 100 extra pounds thirty times. That comes to a 30 percent drag on the Model T's fuel economy--down to 17.5 mpg from Casey's loftiest estimate.
But the Range Rover received an EPA estimate of 19 mpg for the highway. And that was under new EPA standards that take into account gas-draining factors such as the use of air conditioning. The Model T had neither air conditioning nor electric headlights.
For a more complete examination of the relationship between fuel economy and vehicle weight, give a listen to an NPR show from late last year featuring an editor from Popular Mechanics.
For Drobnic it's just "a reach," but her final quotation of Casey provided the most accurate judgment of Obama's claim:
"The government would not allow anyone to sell Model Ts today because they're so unsafe," Casey said. "It's a car that no one would use on a regular basis today. It's not a fair comparison."Not a fair comparison, eh? That put only the slightest damper on Drobnic's party.
We agree that the two cars are totally different. But Obama was careful in the way he phrased his statement: "The 1908 Model T earned better gas mileage than a typical SUV sold in 2008." As long as you don't consider any factors other than mileage, he's right. We rate his statement Mostly True.That rating fails to take into account the deeply misleading nature of the president's statement. His very obvious point was a general failure to significantly advance the fuel efficiency of production automobiles, using that supposed failure to suggest the need to start doing a decent job of improving fuel efficiency. "We have avoided doing what must be done as a nation to turn challenge into opportunity," he said. The rating should have been "barely true," and Drobnic may well realize it now, as an AP fact check rated Obama much more severely on this issue.
The grade: F
Afterword
What is "a typical SUV sold in 2008?
As we saw above, Angie Drobnic Holan apparently used the assumption that the average gas mileage for trucks such as the Ford F-150, the best of which gets 16 mpg. On the other hand, as I understand it, it may have excluded SUVs like the Ford Excusion. On balance, the inclusion of pickup trucks and vans probably lowered the mileage estimate significantly.
Unfortunately for Holan's assumptions, SUV sales were not necessarily proportional. During 2008, SUVs with good gas mileage tended to sell better as illustrated by the following charts. Note that crossover vehicles significantly outsold SUVs. The CUVs generally got better mileage.
|
|
While these data are not complete, we can still use them to obtain a useful estimate. And the right way to determine the gas mileage for the typical SUV was to weight each model according to its sales in 2008. That means that the mileage for the top-selling SUV should be weighted proportionally according to its sales. I'm not going to try to complete that daunting project since the PolitiFact story is a shambles even without going through all that trouble. But doing an estimate isn't that tough. I sifted out the twenty best-selling SUVs/CUVs and found that they accounted for over half of all SUV sales. If the average for the best-selling models turns out higher than Drobnic's estimate, then we have a reasonable indication that her estimate was too low.
A few caveats about the mileage figures: Though most of the models listed come in a number of variations, I always went with the smallest engine (better mileage, generally) and an automatic transmission (worse mileage, generally). I also stuck with 2wd and FWD models exclusively (4WD and AWD models tend to get worse mileage)
It turned out that when the top twenty best-selling SUVs from the two lists were weighted according to sales, the typical SUV in that select group averaged 20 mpg (rounded up from 19.99 if anyone is curious).
That was using the first column of mileage figures, however. The first column represents the EPA's new method of estimating vehicle mileage, and, as noted above, that involves things like running the air conditioner.
Starting in model year 2008, estimates reflect the
effects of
- Faster Speeds & Acceleration
- Air Conditioner Use
- Colder Outside Temperatures
 |
The Model T did not have an air conditioner and did not operate at "faster speeds" (topping out around 45 mph). As for colder temperatures, getting an older car to start was always a challenge. It simply isn't at all likely that Model T mileage has ever taken cold starts into account.
That means that the old system of EPA mileage estimation, though still an apples-and-oranges affair, likely provides a more realistic comparison.
Using that column of mileage estimates, the best-selling group of SUVs averaged 22 mpg (rounded down from 22.12)--about 10 percent better than the estimate used by Drobnic for the Model T's gas mileage. In short, Drobnic's conclusion is subject to reasonable doubt.
So what does it all mean? Essentially this: Not only was the PolitiFact treatment soft on Obama for the "reach" of his implied argument, it was also soft in accepting the raw claim that the Model T earned better gas mileage than the average (new) SUV sold in 2008.
Mar 22, 2009: Added a clarifying clause in the next to last paragraph.
No comments:
Post a Comment
Please remain on topic and keep coarse language to an absolute minimum. Comments in a language other than English will be assumed off topic.