This depends, I think, on one's definition of "better". The designers (and many fans) of 4e pretty clearly consider "stories" with as much misfortune (especially sudden PC death) in them as in old D&D unsatisfactory. Reducing the probability of dying from a failed save versus poison from so many chances in 20 to so many in 1000 may make the outcome seem only the more arbitrary.
So your argument is that poison that has a good chance to kill a character makes for a better story.
All mechanics support some agenda behind what the story should work like. The old D&D attack tables support the idea that an experienced swordsman should hit a foe more often.
Mechanisms explicitly directed at "authorial control" can enable participants to avoid such derailing of whatever stories they have in mind. Heroes and villains alike can have "plot protection" to preserve dramatic structure when the dice are heedless of such niceties.
Yes, and? I don't see how random dice determining an outcome is different in principle than some limited resource. Or some combination of the two. You're talking as if the players were all GMs, which is not the case in any game I've seen.
That is simply not a consideration in most games. There is no plot to preserve in Chess, Backgammon or Parcheesi! A historical wargame may constrain players from departing too much from history (using anachronistic tactics, making implausible alliances, refusing the very battle that is the scenario's subject, etc.) -- but it is central to the game concept that the course and outcome are dependent on a combination of player skill and luck.
Roleplaying games are very different from chess and backgammon. Neither chess nor backgammon create stories.
D&D -- and thereby the whole RPG field -- originated with that same expectation. Finding out not merely how but whether the characters survive and succeed, fail or perish was rather the point of the game! If Ulrica the Unready met a quick and ignominious end, one rolled up a successor and the game continued: an endless game, an ever-emerging story.
The assumption a priori that Ulrica is the heroine of a great saga is a marked departure.
Socialite: Mr. Churchill, what kind of woman do you think I am?!
Churchill: Madam, we’ve already established that. Now we are haggling about the price.
We have already established what kind of character Ulrica is, although we may haggle over the scope of her entitlement to survival and success.
In the old game, Ulrica's story was simply whatever happened in play. That might turn out to be an epic tale, a cautionary anecdote, or a mere footnote in the rolls of the dead ("eaten by owls"; "set on fire and drowned"; "petrified and sold as an objet d'art").
Yes. Different game rules tend to create different kinds of stories. That's why we have different rulesets, and the idea of GM fudging results and making spot rulings - they eliminate these border cases where an unsatisfactory result is created.
That is another benefit IME to metagame mechanics. They tend to eliminate GM fudging. There's literally no need, that is put into the players' hands. Heck, I don't even roll dice in BtVS. I couldn't fudge a roll if I wanted to!