The reason I haven't been updating my fuel logs:

[mhmitszach]

I quit keeping track of my gas mileage for about 3-4 tanks just because I had car troubles and couldn't actually properly fill it up and drive. It was stuck back home and such.

Going to keep track again even if I'm not really hypermiling so much.

[Mustang Dave]

Originally Posted by Frank Lee

My take on fuel logs is they should be all or none. It ain't a contest where we win by picking out the "bad" entries.

I agree.

The only fill-up I didn't post for my "White Pony" was the first one after the dealership filled it during a test drive. They didn't reset the trip odometer when they filled it, so I didn't have an accurate "miles traveled" number on my first fill. So I didn't post that one. I DID post my worst fill-up ever, and explained the reason.

I omitted the first 6+ years on "Big Red"'s fuel log because I didn't have the inclination and enthusiasm to post another 100+ fill-ups. And it would have just been "more of the same"...

In my humble opinion, I've managed to squeeze some fairly impressive fuel economy out of my 4-liter Mustang. My next fill-up won't be as impressive, due to colder weather, but I believe I can break 32 MPG next year.

I promise to keep my fuel logs current and accurate.

[bestclimb]

There is a difference between cherry picking data and expunging bad data.

[Christ]

Originally Posted by bestclimb

There is a difference between cherry picking data and expunging bad data.

Well, that's how I feel, but apparently, you and I are in a league all our own.

[Frank Lee]

Bad data is corrupted with errors or equipment malfunctions... and then there's data you just don't like.

[Old Tele man]

...as the 19th century British Prime Minister Disraeli said: "...Lies, damn lies and statistics!"

...bad data = "outliers" ie: beyond the normal distribution.

[Christ]

Originally Posted by Old Tele man

...as the 19th century British Prime Minister Disraeli said: "...Lies, damn lies and statistics!"

...bad data = "outliers" ie: beyond the normal distribution.

And, of course, statistically, hauling loads heavier than Oprah are outside the realm of "normal" for a minivan, especially my minivan.

Technically, I'm still following scientific method by eliminating those entries that would corrupt the basic data stream that is required in the fuel log, if one were to consider the fuel log a series of test data used to obtain a viable average.

Regardless, I'm not going to discuss it further, because it's pointless.

[Old Tele man]

...actually, I've found a PROBLEM with entering data here! It won't let me enter all three decimal places for gallons used, ie:

actual gallons: 11.236
entered gallons: 11.23_ <--last decimal digit gets truncated, not accepted!

...this makes the MPG either slightly too HIGH or slightly too LOW. And, as Goldy-Locks would say: "...definitely NOT just right!"

[Christ]

Originally Posted by Old Tele man

...actually, I've found a PROBLEM with entering data here! It won't let me enter all three decimal places for gallons used, ie:

actual gallons: 11.236
entered gallons: 11.23_ <--last decimal digit gets truncated, not accepted!

...this makes the MPG either slightly too HIGH or slightly too LOW. And, as Goldy-Locks would say: "...definitely NOT just right!"

I just use place-rounding for that. It's funny, though, that you can enter 3 decimal places for the cost, but not for the fuel reading.

In the case of your entry above, I'd have rounded to 11.24, more likely to 11.25.

Usually, I round the mileage down, and the gallons per fill up. It's only ever by 1/10 or 1/100, so the variance is well within noise.

[cfg83]

Christ -

Originally Posted by Christ

I just use place-rounding for that. It's funny, though, that you can enter 3 decimal places for the cost, but not for the fuel reading.

In the case of your entry above, I'd have rounded to 11.24, more likely to 11.25.

Usually, I round the mileage down, and the gallons per fill up. It's only ever by 1/10 or 1/100, so the variance is well within noise.

I note the third digit in the details section and round up/down as necessary. That way the data is all there.

CarloSW2