If not comparing to another sample or being done in a controlled setting then we want a huge sample.
My MPG on 0w20
- Thread starter Sienna dude
- Start date
Really good way to blow it up and have an excuse to get a car that doesn’t drink like crazy
That’s not chaotic and a wide sample is a law of averages. That’s the entire point of a large sample base.
You’re the physicist here, not me. Can you explain how a large sample base is less accurate?
BMW saw a 3% increase going from LL01 to their FE oils. You’re not going to see anything with that change and could damage things. A few psi more in your tires would handle that. Don’t drive with your tailgate down and get a bed cover but then again you’re already spending money to try and save a small amount. A tune would help as well but you’re talking 1-2mpg after doing everything.
it’s entirely dependent on the significance of the variable. Insignificant variables do not become more significant with increased sample size.
How have you been? Glad to see you’re still reading all my posts.
"Chaotic" definition in mathematics: having outcomes that can vary widely due to extremely small changes in initial conditions. Yep, an MPG value is chaotic when not done under very controlled conditions. I didn't mention variations of tank fill, variations in temperature of the fuel when onloaded (and therefore volume pumped), variations in pump accuracy, and many, many more small changes in initial conditions on each and every calculation of MPG. It's chaotic.
The weather will affect your fuel economy far far more than what oil viscosity will. Or your mood. Or what tires you put on it. Or the road surface. Or if Van Halen is playing on the radio 
And immediately the Charity Brown song from 1974 began playing in my head.
If a variable is insignificant how could it ever contribute to an inaccuracy in a larger sample pool? Just looking for some logic here.
Last edited:
- Joined
- Apr 12, 2025
- Messages
- 867
Tip generator
Is that what you’re looking for? Good to know.
The way you follow me around like a puppy dog I knew it had to be something.
Last edited:
The most significant variable in fuel mileage variability is totally random and one you can’t measure - wind. My Ridgeline displayed fuel mileage on cruise at 70 mph through the flat delta lands of MO and ARK has ranged at the extremes from 23 to 32 depending on whether i have a head wind or tail wind. Three days apart, same oil.
Were you writing posts expecting no one would read them? Yes, I am looking for your logic. So what is it? Please reference your two contradicting posts above.
Last edited:
Yes but a large sample size negates that because it takes all of those unmentioned factors into account. MPG in a frictionless environment or wind tunnel doesn't matter. Pumps are held to a recorded standard, it's labeled on the pump. Fuel temps don't really matter to be honest. However, real world large samples encompass all of that and the larger the sample the more accurate the result. I can look at all my fill ups that I've recorded with the mileage from my odometer and after 2 years it's taken into account all the seasons, highway/city, oil changes, etc and gives the averages. That's why it's called an average. The variables can change but they're taken into account.
In this system, what people are looking for in this thread is an insignificant variable.
But you do know this I’m sure of it.
It's all but guaranteed anytime you're asked to make sense of your post you change lanes.
AZjeff
$50 Site Donor 2023
Please tell us you don’t drive with your seat in the gangsta layback.