Rainbow condition

[AmyNeudaiz]

Very low number for samples, ~100 fish across 6 data points, hence the "general guide, not a rule" part. Not going to put too much effort in this now with 2.0 on the horizon though.

What doesn't seem right to you btw?

I generally notice yellow having a slightly higher fail rate than white and from my experiences red is almost as bad as rainbow. I guess I bolded a bit too much there.
But this is just my experience talking, I have no solid data.