It's never okay to go through an intersection when the light is red...your analogies are quite weak here.
It is always okay to go through the intersection when the light is green; it is never okay to go through the intersection when the light is red. (Well, actually both of those statements are false -- there are circumstances under which it is not okay to proceed through the green, and where it is okay to go through the red....) Likewise, it is always okay to devise the set-up as you will, and never okay to fudge die rolls.
My analogy is weak only because it is a rebuttal to a weak argument, intended specifically to make that weakness obvious.
Yes, because there is a defined goal in chess: to win, using the specific rules of the game. It makes a terribly poor analogy for D&D, of course, which neither is competitive nor does it have a single defined goal.
I can play chess to try to engineer your win; I could then claim that chess is neither competitive (as I am trying to make you win), and that it doesn't have a single defined goal (as I have just made a new one).
Again, neither would make my behaviour acceptable.
Even Gygax suggested that fudging may be called for at times.
I know. In this instance, IMHO, Gygax was wrong.
Are you suggesting that the DM should be able to predict a bad run of luck? Since it's luck, it can change at any time. Consecutive natural 1s can be followed by consecutive natural 20s.
There's no reason to think that a bad run of luck will continue, and lead to a PC's death. How is either the player or the DM supposed to predict that? Your "proper" solution involves the DM predicting the future. That's not a viable solution.
In any edition of D&D, a "run of bad luck" means a reduction in survivability. When setting up a newbie area, the DM should make certain that the creatures/hazards encountered therein are of the kind where there are multiple decision points prior to death.
So, in this case, a "run of bad luck" need not be predictable, prior to character death, to be responded to.
This was the situation I was dealing with. Player stupidity and GM stupidity both shouldn't result in fudging.
Because they'll both be there in normal play.
Well, we agree here.
But in this case: Okay, they have a run of bad luck, and die. You say "wow, you were really unlucky"
They go "uh-huh". They don't realise just how different the game normally plays out, because they have no frame of reference, they're a newb. To them, this is how the game plays out.
And this is sometimes true.
There is a good reason, in D&D, as the game normally plays out, to get used to the idea that you will not always win.
You will note that I recommended that the GM then provides "normal" context.
I will further note, as I did above, that the GM did a poor job of setting up a newbie area if the challenges encountered killed the newb due to a run of bad luck, and withdrawal prior to death could not have been suggested.
And you can't really know they're losing until they actually do so. My first 4e GMing experience, with new players, they thought they were dead meat in the boss fight. Some of them were preparing to flee. Then, they hit the dragon with an immobilising attack, ongoing damage, and a forced move. Boom, it was dead before it recharged it's breath weapon.
I think it is pretty fair to say that you can tell whether the PCs are:
1. In good condition to go on,
2. Hanging on a thread, or
3. Done like dinner.
I accept that this might be harder to know in 4e, because of the way the resource management paradigm has changed. However, I do not accept that, even in 4e, it is impossible to set up a newbie area that allows you to know these things.
AFAICT, that is what D&D Encounters is all about.
I think the core foundation to your opposition to fudging and "saving" the PCs, is that you think damage is being done.
I do.
Nor does something have to give you cancer to do you harm.
Nor does my opinion alter how you play your game, unless you should happen to decide that you agree with me. If you do not agree with me, then does my opinion do you harm?
Since we're talking about a game of make-believe, I submit that anything like this must necessarily be a matter of personal preference.
Every game is a game of make believe. Including chess, baseball, whathaveyou. If you accept that my tricking you into thinking you are great in chess is harmful, then the "game of make-believe" argument is inconsistent, at the very least.
To say it's wrong is disrespectful. To say it's actually harmful goes beyond that.
Pawsplay already did a much better job of answering this than I am capable of doing.
RC