I don't understand. Are you saying that sometimes the GM has to ignore successful checks and treat them as failures because otherwise the players will win the game unfairly or too easily? That's a strange assertion, if it's the one you're making.Which means a player rolling a hot die can - and IME almost invariably would - have her PC bypass any and all obstacles the setting wants to throw in its way, and sail through the story/adventure/mission/whatever without any delays or frustrations or, dare I say, effort...with the one exception being any combats that are unavoidable.pemerton said:when the check succeeds the player decides, when the check fails the GM decides. It's so simple it's elegant! And it doesn't exclude any possibilities - the players are free to declare the full range of possible actions, the GM is free to narrate the full range of possible failures.
The setting, and by extension the GM, exist in part to oppose and-or challenge the PCs and by extension the players; meaning that whether you like it or not there's always going to be that element of adversarialness (yeah, new word there) in their relationship. If the players are given free rein to narrate their successes then most if not all players IME would take that as license to run roughshod over the principles of the game.
I also don't understand what "combats that are unavoidable" has to do with anything. That's just more checks. If the player's dice are "hot" (as you put it) then the player can "bypass" the combat also.
Even within the framework of AD&D I don't really know what you're envisaging here. For instance, nothing in Gygax's AD&D books suggests that a GM can ignore a successful check to find secret doors or to disarm a trap because allowing the success would make things too easy for the players.