[JacobAziza]

After hearing the MythBuster's episode referenced for the nth time, I decided it was time to test it out myself, on my own particular vehicle.

Results:
Inconclusive.

I ran several tests.
For the first, I held a constant speed (55mph) until one exact spot on the freeway, then held in the clutch and saw how many miles I got before slowing to 50mph.
After 2 runs each, the difference over about 2 miles was only .1 (cover on was better by .2 once, exactly the same once)

Then I tried accelerating to 75 on level ground, and then coasting back down to 50 and timing how long it took. Did this one in both direction so that any effect from prevailing winds would be neutral, and started at the same point on the highway in case of any slight road gradients. Cover on won again, but only by an average of 1 second (over 20 seconds - possibly significant as a %, but well under the margin of error introduced by the guy pushing the stop watch buttons).

So, I figured I had to bite the bullet and go ahead and waste a bunch of gas running a long test.

I went about 80 miles, out and back on the same roads, at a higher speed than usual (since aero differences should be more marked at higher speed), starting and ending at the same fueling station.

On the first run, between it being an extremely hot day, having to go through some hills, running faster than usual, and having just replaced the stock fan with an electric, it began to overheat so I had to stop and remove the grill blocking. I made sure to exit and stop the engine at the same point on run 2, but of course that leaves a lot of variation. Also, on run 2, judging by the angle of the trees, there was a substantial tail wind on the home stretch.

Between those two, the accuracy of the entire test is suspect, but I wasn't up for spending the 2nd half of my day repeating them.

In the end, cover off won, by a total of .1 gallons less fuel used (the minimum unit the pump measures, and most likely within the variability of when the pump clicks off).

Looks like its gonna take a much longer road trip to test it properly.
In the mean time I will assume that if it makes a difference, its a small one, and that if there is stuff in the back (which there usually is) it will smooth airflow by having it on.

One bright point of the failed tests:
my average mileage, even though I was running at a higher speed than usual) was my highest yet, 24.08mpg
(this is the first mpg check since adding belly pan, and removing wipers, stock fan, mechanical vacuum pump, and alternator belt)