I think there is a valid point to be made that as the difficulty of something goes up the number of people capable of producing good results with it goes down.

I think there is a valid point to be made that as the difficulty of something goes up the number of people capable of producing good results with it goes down. That leaves all the other people who try to use and wind up producing poor results. If the difficulty is such that there is a greater chance you will produce a poor result with it than a good one then it might not be worth taking such a gamble but rather sticking to a simpler methodology capable of producing good results a greater portion of the time even though the best result it can possibly produce is less good than that of DDD. I really think there is an argument to be made for that. It's not that people are stupid, it's just that things are hard. Doing simpler things helps communication scale as it's easier to get everyone on the same page and engineering failures are almost always the result of communication failures across the engineering org.