Speed differential all the way. Somebody noted on that article that deaths actually dropped in the 65mph zones, possibly because it still allowed grandad to do 55 and not get plowed by the impatient who habitually do 10 over, but probably figured they may as well go way over on 55 because it was so "unreasonable"

Modern vehicles seem actually to have a dip in the curve for wind resistance, the "always less resistance at 55" is true for bricks, but for aerodynamic shapes it depends where they have been designed to be most efficient, and whether they are shaped for turbulent or laminar flow... mostly for safety modern cars seemed to be shaped for good turbulent flow.... which can mean that just into the turbulent region they are more efficient than the high end of the laminar region... which happens in the 50-75 range for most cars. Then also BSFC and gearing is a factor. So the argument for 55 on energy efficiency grounds is rather shaky. Trucks, due to their size, also operate in a different aerodynamic range than typically sized cars and may not be as efficient at 55 as they are a bit faster. You'd actually be surprised that the cD of some trucks is quite low, it's the length that does it.

The distractions argument is valid also, I nearly got hit the other day, at about 10mph, the driver had a cell in one hand and an ice cream in the other...

Well, it's the frontal area really. The length of the object in question would affect the Cd. Simplified, total drag = frontal area * Cd. So even if you can build a semi tractor with a lower Cd than an Insight or Prius, it's going to have way more drag as a result of it's much higher frontal area.

But it's the cD that determines how much effect the frontal area has, hence the popularity of the cDxA figure to determine a factor representing true drag force. Something with a cD of a brick, .60 say, with a 10sqft frontal area has a cDA of 6, a vehicle with a .2 cD and a frontal area three times as great, 30sqft, also has a cDA of 6, indicating the drag force on those two vehicles would be the same.

Hmmm, okay I'm off on the semi's having good cD due to length, I just realised the cD doesn't drop as a function of length/breadth appreciably beyond a ratio of about 4, I was assuming it kept dropping, I was also assuming that a long enough truck got towards approximating a flat plate... I guess they're not long enough.

I'm calling bull on the study.

There were over 205 million registered vehicles on the road in 1995. In 2004, there were over 243 million. That's a larger percentage increase in the population of available units to crash compared to the increases noted.

useless study. pointless to argue. Some will believe it, others won't. There are too many variables that have changed in a decade to draw a substantial conclusion. The fact that physics says speed kills doesn't mean speed was the determining factor.

This is as pointless as cash for clunkers. More stupid from the top.

dir

Such an insightful, stimulating response. :/