I would agree with this.This is going to sound kind of odd, but I think it depends on sequence of events.
1. Party is wandering around in the forest. Picks one path in a split up ahead. GM decides its about time for something to happen, drops an encounter they made up right then and there. Not really illusionism (though can make it questionable whether the players' choice at the fork was meaningful).
I have issues with this. I would agree that, if the DM drops an encounter on the party irrespective of the parties efforts to avoid one then the DM is negating the parties choices. Unless the party clearly fails in the scouting/evading efforts.2. Party is wondering in the forest. Decides the next time the players pick a path, that a particular encounter they've thought of will happen. Players try to listen and maybe scout the paths to avoid hitting any trouble. Encounter happens anyway. Very much illusionism (and probably railroading).
I maintain: that there is no functional difference between content generated by an algorithm (i.e., random tables) and content on a stack.
By the stack, if the DM has a list of prepared encounters (similar to what one would get from a table) and just picks the next one on the stack when it is appropriate in the fiction to present an encounter.