kigmatzomat
Hero
Though I agree with this in general (and thought your post was well explained) I sometimes think this is used to reduce detail below the level that's desirable in games.
There is a broad sweet spot for granularity in the market where desirability drops off very quickly on either side side. Venture beyond it and you become a dreaded "niche" game. Anathema to Hasbro Inc.
I realize some people don't care, but if you want, for example, playing a fencer to feel like a fencer, you need a system that drills down a bit farther than simple rolls to hit.
As fate would have it, I was in a "historic combat fencing" club in college (sponsor was in the local SCA). One of the fencers was ranked in the top twenty nationally in foil but peaked between Olympics, much to his coach's dismay. We had video footage (vhs) of the him and the next best SCA fencer using paired epees. In one leap attack, every couple of frames one of the four epees attacked and was parried. Between them, every sword attacked at least once and there were five separate attacks/blocks in around a second. (Fwiw, I was running the camera. I am a schlub, not a 0.1% level athlete)
I have zero idea how to model that without descending into "Champions-level of complexity" action phases or getting into d&d 3e + Tome of Battle rules where a fighter can get 5 attacks each round and do stunts.
Most people don't want that, simply based on market share. Plus most people already find d&d a "slow" game once you leave tier 2. Adding more granularity will slow the game further.