User blog comment:Kitahara/Recording drum expedition success rates/@comment-25779449-20150129053933/@comment-25832576-20150129063649

I did read a comment on one of your posts regarding sample size, I really want to set the sample size higher but these expeditions take too damn long. It might be a possibility in the future but at the moment the goal is just to roughly gauge where base rates, % boosts per sparkle, and a possible cap may lie, before trying to fine it out.

Ideally if we could determine that levels have no impact, then we'd be able to combine our results for a pretty reliable sample size. But that's going to take some time anyway.