Reply #38: Well, it's a small number of races

[Febble]

but it's interesting that the distribution is normal - compatible with a pervasive factor (whether fraud or bias) rather than a discrete factor only operating in selected states. If we take absolute deviation from zero as criterion for suspicion, if we regard Montana as suspicious, then suspicion should also fall on Maine, Massachusetts, Minnesota, and Connecticut, as well as pro-Democratic fraud in Nebraska. If we regard Virginia as suspicious (not sure where TIA got his Virginia number) then, as well as suspecting pro-Republican fraud in Wyoming, Arizona, Washington, Ohio, Utah, Florida and California, we should suspect pro-Democratic fraud in Maryland, New Jersey and New Mexico.

Alternatively we posit either pervasive fraud, or pervasive bias.

But, either way, the exit poll give very little, if any, indication of where to sensibly look for fraud. But there was almost certainly voter-suppression fraud in Virginia, and I'd like to know about those undervotes in certain Florida counties.