User blog comment:Kitahara/Recording drum expedition success rates/@comment-25730832-20150201174159/@comment-25730832-20150202195941

Yep, the 50% is what I'm trying to test out. I'm also in the same shoe as you since I did not record those resources at all too but just started 5 runs ago. It's not the main experiment and truthfully, it can be done with any other expedition but since it can be included I might as well.

My take on it is currently 2 tables for 1 test, 1 for Normal with 2 columns for get and not get and another one for Great Success with the same format.

The rambling at above aside; Wizon made a blog that is useful if we got lazy and don't want to look up CI stuffs.

http://kancolle.wikia.com/wiki/User_blog:Wizongod/Calculating_Confidence_Intervals_for_Underlying_Probabilities