Indeed... the guy said he lost mileage on his Neon... he said nothing of testing the stuff in a rig like ours... Your point is moot though... when one "fills" his tank when one is "half full", one would write down the gallons of fuel added, and the mileage run since the last fillup... giving a very accurate representation of overall fuel mileage. If I filled my tank every time it got to "half full", I would still have all the same data as I did if I were to run the thing out of gas each time before filling it up.
The EPA estimate of a Neon is something like 30-33mpg on the highway... I don't know the exact number, but if he is getting 61mpg or even 52mpg in that same 33mpg EPA rated Neon, more power to him. The guy has done his homework... maybe it's time for the unbelievers to do theirs...
I have an idea... your dad seems like a smart guy... I'm sure he has access to all sorts of nifty toys in his job. Have him pick up a 4 cycle motor or even a small block, put it on a stand, and attach a 10 gallon fuel cell. Run the engine on pure fuel until it runs out of gas... leave it at idle the entire time... record the amount of time the engine runs on that single tank... do the test at least 3 times to get an average, then run at least 3 tanks of an acetone mixed tank in the same engine and record the results... post them here, and we can see for sure. Scientific study is the only thing that can discount this idea, since many people say it works from experience, and those who say it doesn't just haven't tried it yet or say it doesn't work because of the chemical properties that SHOULD be there and SHOULDN'T be there...
If your dad's testing shows that there is no improvement in "mileage", then I will stop using acetone in my tank... until then, I'm enjoying my 15mpg in a truck that never got more than 10-12mpg.