eyebeams said:
Well randomness doesn't "spike." It's more that it is not just a remote possibility that even a well designed encounter will not have the anticipated effect. While thousands of dice rolls will usually end up averaging the way you'd expect, single dice rolls usually don't.
Now you can say, "Let the chips fall where they may!" but when you do that, this kind of silences you as an effective critic of a system's results. At that point, you don't care about the system -- you care about game mechanic results. So to my mind, it's incoherent to be concerned with both balance and an "automated" way of getting results, with one exception: instances where rules lack clarity. But that instance is also interesting in terms of this discussion.
RPGs are caught in a trap of subjectivity that only seems to increase with attempts to stamp it out. ...Of course, if you stick with a loose system you end up with fudging being made "official."
Does this mean that rolling in public, etc. is bad? No, of course not! But there's a difference between saying that GMs who feel it's *necessary* to fudge are engaging with a "broken" system, or that systems that encourage GM moderation are also "broken."
People should also recognize that when they do let the dice fall where they made, they are *also* engaging in a form of social and system manipulation. There is really no such thing as a truly "hands off" style, as much as social strategies that suit different groups.
First, a disclaimer: I strongly suspect that my idea of a "fun game" would not match yours, eyebeams, so we might have to just agree to disagree. That said, I am puzzled by your central premise, as I understand it.
You seem to be saying that while an average result (in micro for a particular attack roll, or in macro for a particular encounter) might predict a positive outcome for the game-- e.g. party beats up group of orcs-- that statistics cannot predict each outlier result. So, carrying the example encounter, the orcs' CR was appropriate to the party, but still leads to a TPK. To avoid a negative outcome for the game (in the sense of non-fun, frustration, etc.) then you advocate GMs fudging the rolls to move the results back to the positive outcomes that the average results would have predicted.
Is that right? If so...
1. If fudging rolls is required to re-adjust back to the baseline average result that would be expected, then why roll at all? Why not just use average dice rolls? Or, if you still wanted some level of randomness, but more controlled towards the norm, then use average dice rolls for some plus a smaller random factor (e.g. 11d6 vs. 35+1d6). I have played a few HERO games using that method, and really found no real difference from just rolling dice because...
2. Even though each roll cannot be predicted to fall within the norm, given the number of rolls in each encounter, to say nothing of a full session or campaign, it seems extraordinarily unlikely that bad rolls are not balanced by good rolls, and so on. Sure, you might miss your attack while the orc crits, but it is likely that you will hit on the next attack-- odds predict that result, certainly.
3. Either way, if reducing negative outcomes due to unexpected randomness is your goal, would it not be consistent to apply that to both enemies and allies? Otherwise, the players would be protected from fate, but monsters would not. That just seems unsatisfying to me.