This is why I strongly dislike L/100K Its annoying. LOL

MPG or LPK are so much simpler and make more sense. Whoever came up with that method of determining mileage should have been shot.

I think he is over complicating it. Lets come at it from this direction:

You buy a car with X number of miles on it.

The first time you put gas in it (lets assume its a full tank fill up), you create your entry here and record the fill up. So you START with a full tank.

According to fuelly you have gone 0 miles and have used x number of gallons/liters of fuel at x cost already.

Fuelly of course will not give you a mileage yet, as you have not recorded a full tank to full tank usage.

Once you have your second fill up (assuming again its a full tank) you will get your first idea of your mileage. But according to the averages, you will have bought gas twice, yet drove the distance for one.

My suggestion there PB is this. Only drop the first fill up volume/cost recorded from the averaging of both volume and cost. This sets the BASE value (A baseline if you will) for your vehicle. But you can't use it for the calculating math. If you eliminate the first fill-up of values, you eliminate the discrepancy for the averages.

But I will point out, after hundreds of fill ups, the current method is not going to make nearly as much impact as it does with a few the way it is now.

He is saying that, because we are counting all the fuel, dividing that into the total distance driven, we are getting an incorrect value. Now, while I understand his thinking in dropping the last fill up from the calculation, here is the problem with that:

The last fill up, represents how much fuel you used for that previous tank and the distance you recorded. So that value HAS to be used to be correct. This is why the first tank value has to be dropped. Make sense yet?

So you do not need to drop the last and count the first, just simply don't count the first entry at all, even in the averaging. Then you will be all good. I'm assuming your overall average math is counting all the volumes/costs then dividing by all the mileage or something to this effect.

This is what I did with my excel sheets. I do not factor the volume or cost of the first fill up at all for my averaging.

That should fix this.

Now the only other problems are, using partials a LOT, or missing fuel ups, those will REALLY screw with the averages. If people want accurate tracking, they will be best served by ALWAYS filling their tank full, and always recording the fill up. I have so far for nearly 2 years done this without fail. All of my entries are present, and I have no partials. I will take that back, there was one partial that someone else did, which I corrected to a full tank to keep my figures straight by adding it to the next fill up

Hope the book helps you two..