I wasn't aware that Drummond's baskets and rebounds don't count.
20 years from now, when people look back on this era, they will be amazed at how much the coaches got wrong in interpreting the analytics. I have said it before, but the big difference between basketball vs. baseball and football when it comes to analytics is that the events in basketball are not independent of the events that occurred immediately before. Therefore, the events in basketball are not independent variables, but rather very dependent on a whole range of other events that just occurred and are occurring at that moment.
Football and baseball stop and reset between each play. Basketball, for the most part, does not. Basketball also has fewer rules where everyone has to be on the field of play, and the transition from offense to defense is instant. All of these factors mean that the statistical regressions currently being done are only about halfway as sophisticated as they need to be. And as a result, most coaches are still getting it very wrong.
Just because more 3's are generally a good thing doesn't mean that more 3 point shooters are a good thing. Teams should be working an inside/out game to get their best 3 point shooters more 3 point shots. Instead, teams stockpile undersized forwards and guards because their 3 point shooting percentages are a tiny bit higher, and then playing small lineups for extended periods of time. This only works if the other team plays exactly the same style. Pounding the ball inside to centers will get more high quality shots inside, more free throw attempts, more offensive rebounds, and better 3 point shots for the shooters on the court. It will also result in better rim protection, more defensive rebounds, and better defense overall. Analytics has not figured this out yet, but it will.