My best villains are based off the what the players care about. I'll talk about methods and some recent (in the last 3-4 years) examples. You may want to check out my 7 Types of Antagonists as well.
What the players care about - Flags
So first off, I tend to play games with explicit mechanics for the players to tell me what kinds of conflicts they're into. That makes it easy for me to figure out what kinds of villains will press on those conflicts in interesting ways. As we play, a key point is finding out what exactly hits a player's hot button which is usually near, but not quite on the Flag they've given you.
As you play, pay attention when the players are really serious about winning conflicts, or when they're going above and beyond to protect something or accomplish something - all the physical cues (body language, voice) that tells you they're feeling suspense are key to getting more on point, though games where you can spend hero points can also give you a mechanical cue as well.
The General
About a year ago, I ran a game of Tenra Bansho Zero - the player had a
character who was basically an adopted outsider to the daimyo - he was
close to the family but still politically a second class citizen.
Emotionally, he's like a step-brother to the daughter. In TBZ, mecha
can only be piloted by children and adolescents - so usually daughters
are put up to this task while sons are raised to be leaders.
Long story short - the country ends up having to fight a defensive
war, and a general is pretty eager to send her out on the front line.
The player character speaks way out of line at court, and gets exiled
by the general's pressure.
The general has hammered on the fact the player character is an
outsider, separated him from one of the NPCs he cares a lot about, is
going to send her into serious danger, and also put the player
character out to the border where he could possibly die.
This only worked because I had successfully played up the daughter as
being sort of a great younger sister character - a nice kid trying to
do her best - so the player CARED about her, and now the general was
basically threatening all of that. The thing is, the general actually
isn't entirely wrong about wanting to use the daughter to defend the
nation - they're outmatched and a giant robot is actually one of the
few things they've got to equalize. Because of this, it makes it
harder for the players to argue against it. What finally made him
absolutely hateworthy was that the general had nothing but trash
talking to the player on TOP of everything else. (The general falls
under "The Hater" category).
AU Star Wars
5 years ago, I ran a Primetime Adventures game in an alternate
universe Star Wars (clones were not an army, but rather a way for the
government to replace troublesome individuals in positions of
power...)
We had several villians who were all very good and emotionally charged:
A senator - the father of one of the Jedi. He was kind of an ass and
ended up getting in the way of their plans, though I made sure his
motivations were clear. He was hated because it hit on the player's
actual personal issues with his own father.
A clone of one of the player characters - absolutely there to take
over someone's life, and threatened to pretty much do the opposite of
everything the player character wanted to do. The player found themselves having to
kill the clone and feeling terrible about it - it forced them to cross
their own moral line in the process (see below for more on that).
A Jedi who was on the same side as the player characters, but was just a little too
overzealous and violent. He took the player's goals but twisted the
reasons for doing so, even though he often used the same rhetoric. He
made the players question the very cause they were fighting for.
The moment of moral bankruptcy
What usually makes the villain click for players is the point when the villain crosses a single line that makes the players angry. What people often confuse is the idea that this moral line has to hurt the player characters, or their allies - it doesn't. This also can be how a character who is relatively neutral or even supportive of the player characters becomes a villain in their eyes.
This works best if the goals are understandable, but the methods are extreme or twisted along the way.
Transhumanism whether you want it or not
I recently ran a sci-fi game where a roboticist found out one of the
AIs he had built had survived the destruction of his lab, and in fact,
was still out there, doing stuff. Since the lab and it's fellow
robots were destroyed by angry rampaging humans out of ignorance, it
decided the solution was to improve and fix humanity. To snatch up
souls from the dead and give them immortal bodies, better bodies, and
more time to learn - eternity.
Of course, no one was asked if they wanted this. And the AI simply
treated them as objects, to be turned on or off at will. ("Don't
worry, if this unit bothers me too much I'll just reset it's memory.
It's easier that way.") And the AI decided that in order to show how
much better this new path was, it would destabilize the religion on the
planet by supporting terrorist violence along the way - "If they die,
it doesn't matter, I can simply bring people back. This is why my
path is superior to religion. I can bring people to life, eternally,
now. So who cares if they die in a bomb blast? They'll be back soon
enough."
The casual disregard for human life, or human will was pretty much
what made this AI a great villian. It stepped on the roboticist's
sense of responsibility for his creations, and another character's
sense of obligation to the dead soldiers he once led.
Drifter's Escape - casual racism
A friend of mine played a game of Drifter's Escape and relayed this
event in play. The main character, a drifter, is picked up in a truck
along with several other folks to do day labor. The character is
white, but the other men are Latino. The farm owner picking them up,
also white, says to the drifter, "Hey, why don't you sit up front, in
the cab? We'll have some good work for you." and throughout the
session basically offered the character preferential treatment to the
other workers throughout the session.
For everyone at the table, it was a great example of a villain who
didn't have to do anything overtly violent or say anything ugly - in
fact, he simply offered the drifter good things - but everyone could
see that part of that price was to participate in putting down/cutting
out the other workers from a fair shake too.
A simple, moral line.
Justice delayed is justice denied
Along with the specifics of a villain being either an obstacle to goals or twisting ideals, etc. a key part of how you play that out is having a villain the players cannot pay back or overcome immediately. The frustration makes you hate them more, and you really can't wait to finally see them get theirs. Of course, when that time comes, it might cost enough that you have to really decide if it's worth going for.
A good extra is your allies or friends also like this terrible character. It turns the battleground into fighting for their hearts - convincing your allies of your cause and avoiding social fallout for taking action.
Troublesome Villains > Fiat Villains
In the games I run, I don't have villains who simply auto-succeed at doing terrible things - they don't just manage to kidnap your loved ones, or get you ousted without the players getting a chance to prevent it. I don't fudge the dice or simply play a "gotcha" on the players.
What this does is establish a position of trust for us as a group - the players know if they protect or save something, they really did it, I didn't hand the victory to them. They also know if they fail or things go wrong, they COULD have possibly succeeding and stopped the villain, which means they really rack their brains over "what could I have done better?"
In comparison, when you have a game where the villains do their evil by GM Fiat, the players don't feel like fighting harder, they feel cheated and angry at the GM, not the villains. I've played those games, they're not fun.
