The basis for Free Kregspeil was mentioned earlier in the thread. It relies on an expert to make adjudications based on their expertise. IMO, the DM being the creator of the setting/dungeon/world places him in that expert position for D&D.
Personally I don't see it.
In Free Kriegspiel, there is a shared goal, in a teaching/training context, of doing a thing well. An expert in that thing helps with the teaching. In some ways it's like when I judge a moot, and I use my knowledge of law and legal argument to pretend to be a judge and put the student advocates on the spot.
The students can get better by going off and studying more law and more advocacy. And I'm a better moot judge now than I was fifteen years ago because I've had more experience and training of my own.
Likewise, the junior Prussian officers can go off and study more, speak to other more experienced soldiers about their experiences, etc; and then they will do better at Kriegspiel.
Both the wargaming and the mooting can work like this because there is an external reference for adequacy: the real worlds, respectively, of warfare and of legal practice.
In the D&D case, what is the GM an expert in? What is the external reference that establishes criteria of adequacy? How does the GM know better than the players what will happen if you strike a bronze statute with a war hammer, or if you try to seek an audience with a noble, or if you prey to Corellon in Orcish?
All we really have is the assertion that the GM should enjoy authority over these things. Cloaking it in the language of expertise is just misleading.
I already quoted, upthread, the text of the Hermit background feature which tells the player to work with their GM to establish the details and implications of that feature. So 5e D&D
in its core rules rejects the proposition that the GM should enjoy sole authority over setting. Clearly some people don't follow the rules of 5e D&D in that respect.
Of course that's their prerogative to do so. But the notion of expertise, which does a lot of work in the context of Free Kriegspiel, does none in this case. Its
authority over the fiction that is at issue.
It is a bit odd in structure that the DM controls both the opposition's behavior/numbers/terrain advantage/social advantages (the game isn't just combat), etc, while simultaneously determining how well the PC's tactics work against that opposition.
In old school play this oddness was avoided by focusing on designing the whole dungeon before play, then attempting to stay neutral during play as the PC's tried to overcome it. This is why I don't think the notion of MMI really came up much in old school play (i'm sure someone will correct me if I'm wrong). But now, unlike with a dungeon, the whole world in all it's detail cannot be designed ahead of time. So sometimes the DM is simultaneously creating content and adjudicating player actions upon it - or at least nearly so.
To me that's the area that's causing the real contention, and the area that really could have the most structured improvement for D&D. Thoughts?
For many years I've posted on these boards that the idea of map-and-key adjudication isn't able to be operationalised outside of the artificially sparse, and highly conventionally governed, domain of classic dungeon-crawling. (Here's just one example of the conventions I have in mind: it's mandatory for a classic dungeon key to give you room shape and size, but not for it to give you the colour of walls or the number of pebbles on the ground.)
I'm currently sitting at a desk where it would be a real pain to itemise every item within reach, let alone every item in the room, let alone everything I can see through the window.
It would be, in practical terms, impossible for me to list every person I know, and everything I know about them and how they respond to various things, what they do and don't like, who their friends are (some of whom I know, some of whom I've only heard of), etc.
I think it's obvious that map-and-key resolution can't possibly work for a "living, breathing" world.
EDIT: Map-and-key resolution, and authority over the fiction, are related in this way (maybe other ways also, but this is at least one salient one):
The players are expected to declare actions for their PCs that will engage with the mapped and keyed fiction - moving objects, poking surfaces, opening doors and lids, peeking around corners, etc. And the GM uses their knowledge of the map-and-key, which is itself treated as authoritative, to extrapolate what happens as a result of those actions.
Map-and-key resolution will break down if the the participants don't do their bit. For instance, if the players start focusing on the colours of floors and ceilings, and the GM hasn't got those things in the key, and so just makes up answers on the spot
without telling the players that this is what is going on, play will become inane - because the players think they're doing their thing of "cracking the GM's code" by gradually collecting information from the authoritative key; but in fact it's all just sound and fury signifying nothing.
This is a small illustration of the reason why
@Campbell posted, way upthread, that one of the key things for successful exploration-focused play is GM care in designing and communicating the setting/situation.