Winterknight, sorry but I cannot agree with some of your points:
It would be a dull and quiet world, let alone forum, if we all agreed with one another.
What I'm saying is that we should move this element of risk from "uninteresting" parts of the game to more interesting ones.
We should ask ourselves: "why do players and GMs cheat?"
And the answer IMHO is: to avoid boring/uninteresting situations in their game.
Here's my opportunity to disagree with you. I think people cheat, fudge, tweak for as many reasons as there are for gaming.
Why do poker players cheat, even when money isn't at risk? Why does my uncle stack the deck in a friendly game of cribbage, even though he rarely benefits from it? Is it to feel an adrenaline rush? To win at all costs? To get something over on another person? It's rarely, IMHO, because the situation would otherwise be boring.
In other words you spend a Fate Point in that combat against a orc because you think that the death of your character in that situation would be boring.
Or, because you don't want to spend an hour making a new one. Or, because you're deeply attached to the character. Or any number of reasons.
I once had a player commit suicide (I did not require a roll) because he had blundered into a trap and his foot was cut off. Despite the fact that the world was fairly magic rich, and prosthetics were abundant, and with the party's healer, he would have been back to nearly fit form almost immediately. No, in his opinion, I had destroyed what had been the "closest thing to a perfect character" he'd ever had.
Many people have complex relationships with their characters, even the guys you think wouldn't. It's easy to see why they'd be loathe to lose the character, as an investment of psyche. In the above case, the player couldn't even stand to have his character marred, and would RATHER lose him that see him flawed. As I said, complex.
Is it possible to build a set of rules that will help players and GMs to avoid boring situations and stress interesting ones without having to cheat?
IMHO yes... not only it is desirable to do it as games should be fun!
Second part first. Absolutely. Games
should be fun. Sometimes they aren't. I can play Guitar Hero with my wife, and it's a great time. We help each other out, compliment one another, and enjoy the music. When I play with my sixteen-year-old son, I can't take more than 15-20 minutes of it. It's not that he pwns me (I'm old, fat, and slow- he should), its that he gloats, dances around, rubs my nose in it. Same game, the only changed variable is the player mix of personality.
I've played RM with groups of players that were absolutely fantastic. I've played with groups that stunk. I've played where the change of a single player made all the difference. I've also played about 20 other game systems over the years, and about 3 times that number of board and card games. Yes, some of the systems stink, whether from poorly worded rules to lackluster mechanics. But overwhelmingly, the majority of the truly unenjoyable moments come from the human characteristics, and have nothing to do with the game.
As to whether you can build a cheat-proof set of rules, well, I have to say that I don't think it can be done. I don't know that it should be done. I think it overlooks the more basic issue: the player-GM covenant, and the degree of trust associated with it.
In my opinion, a rules-light story-sharing structure turns over a certain degree of control to the players. This might be great if you have players who all care equally about the story, and about one another. If, however, you have an individual who wants to be (and is capable of being) the center or attention, that person can make the game boring or uncomfortable for everyone else. If you have a group that enjoys that kind of sparring, then it's a wonderful experience. If not...not so much.
When you negotiate what you do, in a narrative fashion, you alter the outcome. When I change a 99 critical roll to a 33, I alter the outcome. I consider both to be equally valid, and neither is wrong.
My campaigns tend to be role-played novels. Not the written works of others, but a novel-sized adventure that builds from the shared stories of all the characters. A few die along the way, but their deaths are rarely futile. How a character faces death can be much more important that where or how he finds his end. A death in a bar room brawl, particularly one precipitated by the player, can be as significant as a leap from the highest tower to land a killing blow on a demon with an enchanted sword.
You'll hear me say things like "Make a bribe-the-guard-without-getting-caught-by-his-supervisor roll." Or "Make a crawl-through-the-bushes-under-her-window-silently roll." If it's not important, it usually doesn't get a roll. If it is important, the players know what is at stake, and may themselves suggest alterations.
"Hey, I should get a +20 to that bribe-the-guard roll, because I spent yesterday afternoon playing dice with him in the tavern."
"Yes, and you won...poorly. I'll give you a +10, no more."
These kinds of negotiations and interactions are developed because the players know and trust the GM, and because the parties involved are imaginative. Can this work just as well in another game system? Certainly. Will another game system DESIGNED to have this kind of interaction ENSURE this kind of interaction? No way. Unless the players know and trust one another, no system will force them to take a more controlling role. If they do have that level of trust and imagination, Rolemaster can (IMO) fulfill those needs as well as any system, and better than most.