In the abstract, avoiding having "nothing happen" is a good thing because it has the decided benefit of helping prevent play bogging down.
Game mechanics and/or GMing techniques that prevent a "nothing happens" outcome, however, do work at cross purposes with the players' in-character efforts (if any) at risk-mitigation. Specifically, players trying to maximize the chances of their characters succeeding at their longer-term goals have an incentive to try to limit the potential in-fiction consequences of any risky actions they elect to take along the way--in other words, such players are actively trying to drive the downside risks as close as possible to "nothing happens." That doesn't (or possibly can't) play nicely with game mechanics and GMing techniques that seek to prevent that outcome.
Of course, in those games where setting longer-term goals and strategizing to maximize the odds of achieving those goals isn't a part of play, this tension is irrelevant (assuming everyone is on board with the goals of play, anyway!) But it's definitely a part of play at some traditional tables. And I would suggest that goal setting and strategizing (and the resulting player efforts at risk mitigation) is a defining part of play in some styles of sandbox campaigns.
From my perspective it would seem to be a rational choice for any GM to "outright reject" any technique that works at cross-purposes to a defining part of play at their table. Would you agree? If so, it would seem to me that GMs whose campaigns are defined by longer-term PC goal setting, strategizing, and attendant efforts at risk mitigation are making a rational choice to categorically avoid techniques that prevent "nothing happens" outcomes, and to instead find some other method of avoiding play bogging down.
At a higher level of generality, it's reasonable when playing a game with well-definied purposes of play to reject using incompatible GMing techniques, yes? Traditional games don't have such clear-cut or uniform purposes of play, but each individual campaign presumably has some (hard to categorize though they might be), and it would seem to me to be equally reasonable for each such campaign's GM to also reject incompatible GMing techniques.
I would suggest that GMs who reject allegedly incompatible techniques may simply know their campaigns well enough to be making an informed choice, rather than engaging in reflexive conservativism, even if the difficulty of analyzing the diverse playstyles that make up "traditional" campaigns makes it challenging to tell the one from the other.