The Surfacescapes team at Carnegie Mellon is doing some interesting work using Microsoft Surface to support tabletop D&D games. In some online discussion, someone claimed that having a computer adjudicate would be bad, since a GM couldn't "cheat" "in favor of a good story." The commenter claimed that if the GM were to fudge with such a system, the resulting override would be obvious and the atmosphere would be destroyed. There were unhealthy assumptions built into that claim, and it gelled into two Twitter posts. The first was:

If the GM needs to fudge die rolls for the sake of fun or story, you're using the wrong game system.

A bold, direct statement. A rule, if you will. Like any real rule, you need to know when to bend and break it. One of the strengths of a tabletop RPG is that rules are relatively safe and easy to bend. Problems that would hurt a computer game or strictly interpreted wargame can be easily patched over.

However, each time you do this, you're modifying the game, creating a new, slightly different game from what was published. And that's also okay. The hobby is rich in part because people have been modifying games since minute one. But in this case you're not agreeing as a group to a modified rule, you're sneaking one past the other players. That's not inherently wrong, but it's a dangerous, seductive tool. One little fudge here and a pointless, random death disappears. Another fudge there and a frustrated player becomes reengaged. Another fudge and an accidental weakness in your plot doesn't cause the entire thing to shatter when a PC missed a key roll. Without realizing it, you stopped playing D&D and now you're playing "Whatever the GM wants, happens." That's a fine game, but you didn't mention to your players that you've changed rules sets. The players are under the illusion that their character's fates are in the hands of themselves, the rules as published, and the dice, but you've lied to them.

The better solution is to find a game system that supports what you want. It might be an entirely different game. It might be an agreed upon modification to the game system you're already playing.

Now, this rule does assume an ideal world. It assumes that your group has similar enough tastes that there is a single system, or modifications to some system, that would match their desires. It assumes that your group is open to experimenting with modifications or other systems. It assumes that your group has the time for that search. That's a hell of a lot to ask. When you're in the middle of a climactic scene and the dice demand that you replace awesome with inane, it's hard to justify interrupting play to engage in a bit of improvisational game design. So in reality, you fudge.

I've fudged before. I'll fudge again. But I'm playing with a loaded bazooka and will do so cautiously and sparingly.

I almost sent out the above tweet as, "If the GM needs to fudge die rolls for the sake of fun or story, you're probably using the wrong game system," but decided against. In addition to not being as pithy, it weakened the rule, providing an unearned Presidential Pardon for fudging. Yes, there are exceptions, but like the exceptions that die fudging inherently is, it's best left unstated.

.

All of this lead into my next tweet, "If a possible die roll result is unequivocally bad for the game as a whole, why are you rolling? Why is it an option?" I hope that one stands well enough on its own. I erred on the side of caution and added "unequivocally," which I think fairly covers things.

(I cannot claim that these ideas are original to me. They almost certainly came from my reading into RPG game design, and I'm sure have been stated by others in other forms. I like to think that this specific phrasing is original to me.)