No that's not what I'm asking at all you snarky ass. The other answer is probably the correct one, that being the average for that day historically.

I was assuming a cumulative average or even a moving average was being used and if that were the case then there should not be a scenario where adding a current measurement that is lower than the average would actually raise the average. If I need to point you to the actually moment where this happens, I will be happy to.