The delivered temperature can pretty easily be the air temperature and so selling gas at 90 F can happen whereas you suggest it can't.

There are very good reasons not to deliver gasoline at 90 F. That would make for a blending nightmare as butane flashed off.

But I agree that you could deliver gasoline at 90. But when the NIST did their sampling, that's not what they found. I have heard that the average in California was higher than for the rest of the country, but that the sample size was very small. But I have never been able to dig up the raw data. That would be quite interesting.

Hi Robert,

I think the raw data is needed to make a better estimate. One want's to look at the temperature together with the selling price to see if prices are elevated together with temperature. If not, then the effect will average out at the level you cite, but if so, then the effect will end up higher. One also wants to look at the skew since if all the below average numbers are closely clustered but the above average numbers have a large range you can get oportunities for manipulation as well.