Odd and even App.number get (unwantedly) associated to different on-peak probability distributions, due to a poor mathematical formulation of the proability function. This could be improved/fixed based on the suggestions from @wc12864.
The discussion with @wc12864 occurred mainly via email between me and him, so the information needs to be retrieved from the email exchange and condensed in a way that can fit the documentation.
Odd and even App.number get (unwantedly) associated to different on-peak probability distributions, due to a poor mathematical formulation of the proability function. This could be improved/fixed based on the suggestions from @wc12864.