This is a combination question and commentary. Here is the issue:
Using this site, I have been tracking my mileage the old fashion way - miles driven / gas in. I compare that to the MFD, understanding that the dealer probably didn't fill it the way I do on that first tank.
Anyway, here is the skinny from MFD: Tank 1, 38 mpg (dealer short?) Tank 2, 56 mpg, Tank 3 55.6 mpg. At this point my calculation was 51.9 versus 51.3 on the MFD.
Then my wife borrowed the car at about 1300 miles on the odometer. She drove ~40 miles, and pulled the MFD average down to 50.9. By my calculations based on my odometer reading before she borrowed it, she got about 26.5 mpg over that 40 miles. Seriously, you have to try to get that low!

The problem is that my wife thinks that the Prius will get better gas mileage than the other car in any case, so why should she even worry about how to drive it. It's a car she says - just drive it.
Even so, it is bugging me to no end. I feel like the MFD has been tainted. I'm smart enough to know that this is a temporary thing, that at 30,000 miles that extra gallon of consumption will have little impact on the overall mpg. But it bothers me now. This all got me to thinking....
When I reset the trip meter each tank, the # miles on the MFD resets to 0, but the overall fuel economy calculation does not. It has a separate reset on the MFD. I started wondering what everyone else did. Does everyone leave that as a lifetime stat, or do you all reset it as well each tank and let the tracking website or spreadsheet calculate your lifetime?
Because if most people reset it each week, then next tank my "tainted" numbers will go away when I reset it. Thanks for any thoughts you share.
--John