Re: TEB: Explain average of ratios/ratio of averages better

[Chris Warburton]

I've kept the terminology, but also:

 - Mentioned that ratio of averages is a pooled average (accounting for
   duplicates)
 - Given equations for both forms of average
 - Mentioned that the ratio of averages is the average numerator and
   denominator
 - Mentioned that neither tells us the absolute number of conjectures
 - Mentioned that we can take the mean of the cardinalities to predict
   this
 - Included F-score, along with equation
 - Added true/false positive terminology to precision/recall definition
 - Added no negative examples as a limitation