If anyone reading it didn't know, I've been pre-reading Jetfire's ongoing Dangerous Business sequel, Besides the Will of Evil. Now, if you want to go look through the comments section over there, you'll see that there's been some debate in the comments for a while now over the direction the story's taken. The short version, though, is that many of the commenters feel like the story's villain has been too successful, and the heroes too ineffectual.
Now, I'm not actually interested in talking about BtWoE itself, but the discussion there did get me thinking about the idea of over-effective antagonists--both how they can ruin a story, and how they can be used effectively. And specifically, how a particular piece of advice that I've seen offered a lot (not just on Jetfire's story, but on plenty of other long dark stories, both fanfic and non) isn't necessarily a good one. Some long-winded thoughts, below the break.
Conflict--at least, going by the broadest definition of the word--is the lifeblood of any story. That conflict can take a lot of forms, but in many stories it (or at least, an aspect of it) comes from a villain of some sort: a character whose wants/needs/beliefs bring him/her/it into conflict with the protagonist(s).
In most literature featuring a villainous character, that villain starts the story in a much more powerful position ("powerful," in this case, can mean a lot of things; in Rocky IV, Drago is shown to be more physically powerful than Rocky; in Robin Hood tales, King John and his cronies have money and military might countered only by Robin's poor, lightly-armed insurgency; in the major battles in The Lord of the Rings, Sauron's forces invariably have numerical superiority; etc.). Usually, this position is maintained at least until the climax of the story, since having a hero "triumph" over an obviously weaker foe is less dramatic. A weak(er than the protagonist) villain makes the hero look more like a bully than, well, a hero, and it's harder to see a clearly overmatched antagonist as posing a legitimate threat to the protagonist's aims--and thus, as being a source of conflict in the first place. That doesn't mean that villains can't have weaknesses, or even be portrayed as weaklings (the Emperor in Return of the Jedi is a wrinkly old guy who's clearly no physical threat to anyone), but they are almost always more powerful than their foes.
Of course, one can go too far when it comes to making a villain seem appropriately threatening; if one crafts a scenario where the villain is in such a position of relative strength that it would be impossible for the heroes to triumph against them, then has exactly that happen through some fancy handwaving unsubstantiated by the narrative to that point, the resolution comes off as unsatisfying and unbelievable. When a villain can only be overcome by a bad deus ex machina (as opposed to a good one--divine-or-equivalent intervention occurring in a story where that's a reasonable and narratively satisfying outcome is still a deus ex machina, after all), readers will cry foul. And when they do, there's one piece of advice I see, time and again: let the good guys win a round! Give the heroes a few more meaningful victories prior to the story's conclusion, to show that they aren't so badly overmatched overall; give us a powerful enemy, but one who we know the heroes can, ultimately, defeat.
That's not necessarily bad advice; a story specifically about a hero's rise to physical/military/political greatness needs to show that hero making incremental improvements, at least to some degree, lest their ultimate demonstration of superiority against the villain seem like it's come completely out of left field. But in many cases, I feel like that advice is missing the point: having identified a problem (the villain can't lose in any textually supported way, and yet does), it offers as a solution a change which doesn't meaningfully alter that part of the story's dynamic.
The problem, in such cases, usually isn't that the villain is too overwhelmingly powerful per se; one hardly needs to dig deep into literary and film history to find plenty of examples of absurd power differentials which nonetheless made for compelling stories (The Terminator gives us a couple of ordinary people against a humanoid killing machine that's essentially immune to conventional weaponry; The War of the Worlds pits 19th-century humanity against space invaders in a battle they literally cannot win; The Odyssey sees Odysseus and his crew confronted by a laundry list of impassable monsters and the wrath of Poseidon himself (you know, God of the sea, upon which Odysseus is sailing...)). The problem, rather, is that there's no good reason why the villain shouldn't win/have won in the end. And letting the hero score a few more points along the way does nothing to change that, by itself. At best, it teaches the reader that the villain's might is an informed attribute, and that the power they were supposed to have (the power, remember, on which the conflict, and thus the story, is premised) has been misrepresented.
At worst, it can completely undermine the moral of the story. Stories which are based upon good deus ex machinas, or those in which a (or even the) plot point is that evil cannot be defeated on its own terms (often murder; other times "its own terms" means waging war, physical conflict, acting alone, as revenge, and so on) can actually be significantly weakened by an attempt to put the hero and villain on more level footing. For example, let's look at Star Wars (i.e. "episode IV"). After the opening sequence, things go pretty damn poorly for the rebel alliance: Alderan is destroyed, Obi-wan is killed, and even seeming successes like Princess Leia's escape turn out to be all part of Tarkin and Vader's plan to locate the rebel base--a plan which succeeds perfectly. Beyond the setup, the imperial forces suffer no meaningful setbacks during the movie... until the climax, when Luke makes an impossible shot to destroy the Death Star ("use the force," incidentally, is an example of a good deus ex machina; Luke suddenly is able to do something that's well into no-meaningful-chance-of-success territory, but the sudden change fits the world that's been established perfectly). Beyond that, there's the simple fact that the Empire is clearly in a beyond-dominant position to begin with; the iconic first scene of that movie drives home the difference of scale as well as anything, and when the bad guys have a moon-sized battlestation that can blow up entire planets... well, that kind of speaks for itself.
And yet, however one feels about that movie, I think it's fair to say that it wouldn't have been improved by making the Empire weaker, the rebels stronger, or by giving the good guys some incidental victories (at least, victories that didn't ultimately turn out to have merely been ploys by the villains). Balancing the heroes and villains wouldn't do anything to address the movie's weaknesses, nor would it reinforce its strengths.
To sum it up: an overpowered villain is definitely a bad thing. But if that villain is "overpowered," the problem is rarely a severe mismatch in power with the hero, and the solution is equally rarely to try to balance that mismatch. Most of the time, an "overpowered" villain isn't one who seemingly can't be defeated by the good guys, but one who can't be overcome in a manner consistent with the world the story takes place in. It's an important distinction to make; there are few things more satisfying than seeing an impossibly mighty villain toppled in a satisfying way, after all!