[time-nuts] How good is my T-bolt...??

[Magnus Danielson]

Mike Naruta AA8K skrev:
> What if all three are different from each other?
> 
> Or, if two agree, how do you know that the
> two are not both wrong?
> 
> If you have 30 clocks and 20 say one time while
> 10 say another time, do you go with the majority?
> 
> Is there not a small probability that the 10 are correct?
> 
> 
> :)

Actually, you can't say that. If you have 20 clocks saying 12:05 and 10 
clocks saying 12:10 and you see the sun setting (assuming we are 
sufficiently distant from polar circles, which most people are most of 
the time), then clearly all 30 clocks are wrong. Statistics only 
indicate the mean value for your set of clocks, not which one of them is 
right. Your lack of external reference prohibits you from making such 
conclusions.

Cheers,
Magnus