I do not think we need to provide evidence that a developers time cost money (ie, to code/acquire DRM software). Second, I don't think we need to provide evidence that running a server and having bandwidth to run authentication servers cost money. Last, I doubt we need to provide evidence that extra support calls thanks to DRM issues cost money.

Each three of those are what I call common sense, ie, we know them all to be true. Those are also the only assumptions I included in my "I relied on common sense and observations". The rest is observations, which is the first step in a honest and scientific discussion about the effect of DRM. First you do observations, second you collect evidence, third you do analysis and last you create theory. Only once you done each step and reached theory should economic decision be made in the case of including DRM to increase revenue. Before that, only the cost of DRM is know and everything else is speculations. Those speculations has both those that say its great and those that say its crap and there is nothing in the world that will convince them beyond a scientific approach.