Honestly, for me, I find fortune in the middle makes the most sense. Or, perhaps a better way to put it, it's the most versatile. If you put fortune at either end, then you can only have one explanation ever for any event. To me, that's not what good resolution mechanics should do. Good resolution mechanics should never dictate the fiction beforehand.
While I agree with your analysis, I don't agree with this. Fortune at the End can produce different explanations for any Intent, based on success/failure of said actions, though that comes down to specific implementations of resolution systems.
What I'm really interested in, though, is
good mechanics. I can't argue with your preferences; all I can do is share mine. What I believe good resolution mechanics do is
change the situation. I believe that, when you enter a conflict-charged situation, engaging with the mechanics should produce an interesting change to the situation without creating a conflict of interest for players.
(Leaving aside the aspect of creating a conflict-charged situation for the moment...)
The
interesting change is often the most difficult part to get right, because it relies on many different factors: suspension of disbelief, genre conventions, and others. What good mechanics should do is allow players agency in how those factors tie into the reward system, and have the
interesting change reflect changes to the characters and how they interact with the game world; and furthermore, by those changes make the characters more interesting and their relationship to the game world more complex: i.e. a reward system.
I believe that engaging a resolution system should engage the players with the reward system; and a reward system should factor in how the game deals with suspension of disbelief, genre conventions, etc. (And, of course, what the interesting material in the game world is; that is, Creative Agenda/GNS.)
The
conflict of interest for players is pretty well covered by the "Czege principle"; I'll let Eero take it for me
here:
...a proposition by Paul Czege that it’s not exciting to play a roleplaying game if the rules require one player to both introduce and resolve a conflict. It’s not a theorem but rather an observation; where and how and why it holds true is an ongoing question of some particular interest.
*
Now, all that said, when one considers a design with Fortune in the Middle or at the Beginning, I think you have to look at these things: how the Situation is set up (Characters + Setting, and how they interact); how player agency interacts with the resolution system; and how the reward system responds.
Either method can produce the results that I'm looking for: as long as you get players influencing the reward system through their choices (assuming you have a decent reward system!), FitM vs. FatE is a difference of technique; either one suits some designs and not others.
Fortune at the Beginning is an interesting aside: I'm thinking of Wandering Monsters + Reaction Rolls in B/X. Monsters roam around, the
mythic underworld is full of them; they don't have much treasure (so little XP) or much else to offer PCs. The Reaction Roll offers PCs a way out of this
dangerous situation, since the Hostile result is rare; and, of course, players can reduce the number of Wandering Monster checks through efficient dungeon exploration. This seems to set up a situation where the player's choices suddenly and radically change, and one where player choice can make the outcome work for the PCs (savvy diplomacy or e.g. leading unintelligent monsters into a fight with the monsters in another room) or lead to disaster (1d8 HP for Fighters!).
So what you have there is player agency leading to an interesting change in the situation, and player agency feeding back into the reward cycle. Well done, and that's FitB.