What are we measuring to determine if this or any change “improved” and worked?
The sheer volume of data here admittedly does not transfer well to a hunting forum, but I found some interesting reading here:
Performance Report
Reading through this a bit, it occurs to me how difficult it is to objectively interpret this type of data while also successfully avoid several data fallacy pitfalls (McNamara, P-Hacking, Cherry-Picking etc.). In other words, you’re asking a very important question that I’m not sure there is a good answer to.
Last edited: