I figure every great action game has a villain, somebody the player is supposed to absolutely hate. Thinking about the different villains throughout time, genres, and different platforms, I've realized a couple universal statements:

Defeating the villain will give the main character something they want, something of value. This is my "umbrella theory" because all other theories are just extensions of this one. For future reference, I will call this the "motivation," as it motivates the player to dislike the villain.

Each of the above motivations has its own place, and only works in specific situations.

For the motivations to be effective, they must be universally evil, or understood.

Every villain has dialogue, or some other form of expression, that (s)he/it uses throughout the game to enrage the player.

<wall of text>
Regarding the first theory, "Defeating the villain will give the main character something they want, something of value," I have determined there are infinite posibilities for the motivation. For instance, the motivation might be money, power, revenge, or my personal favorite, a tangible item (like kirby's star-rod). It is your job to plan and create this motivation. This brings us to the 2nd point:

Like anything important, there is no "right" motivation, generally speaking of course. If a death metal version of Kirby's Adventure were released, King Dedede would have to do a lot more than steal a star rod to become a villain. For those who have no idea what that meant, how could you hate the Germans in Call of Duty if only pastel colors were used? The idea of shooting somebody in a happy environment goes against human nature.
Basically, the motivation MUST make sense, given the situation. This includes every aspect of gameplay from graphics, to music, even genre. This deals mostly with setting the tone or mood of the game, so the player can properly reflect and react.

Graphics: As far as graphics go, I find the less evil the motivation, the brighter the colors. Obviously, this rule cannot apply to all situations, nor can it represent a linear graph. There is a limit to brightness of colors, but there are other articles for that. This rule, along with others is only a guideline, not a DO or DON'T.

Music: I'm not much of a musician, but generally the more evil the motivation, the lower the notes should be and the slower the pace should be. The same can be said in relation to graphics: The darker the graphics, the lower the notes should be and the slower the pace should be. (Please excuse my lack or misuse of any musical terms and/or definitions)

Genre: This one is a bit tricky. All aspects mentioned previously can (and have been) totally ignored by games, and still be successful. However, genre is especially flexible. To make it easier to analyze, I will henceforth reference to genre instead by the game speed. The speed of gameplay has a large part in setting a mood or tone. Obviously, a puzzle game will be slower than a top-down shooter, and an RPG might be somewhere in between. Generally, a more evil motivation means a slower speed, darker graphics, and slower music.
Note: These pseudo-ratios are not meant to be law, nor are they meant to limit you, but are meant to act as a guide for the confused. Many great game designers have disregarded these suggestions and have been very successful.

Right. Now for the third:

This one is pretty self explanatory. The more people who understand why they are supposed to hate the villain, the more evil (s)he/it appears, and more people will find the villain appropriate. Hey! you're still reading this? Thanks a lot for your support. As a show of thanks, if you post "The rooster has awoken the farmer" then a series of five digits whose sum adds up to a prime number, in a comment I will give you my eternal respect! Ah-hem...back to the article: Think of it this way. If my friend died of food poisoning, and I made a villain E.Coli, it would be a truly evil villain to me, but what about you. You would see a little blob that doesn't talk, and hardly moves, then you would say "LAME" (hopefully out-loud to provoke embarrassment) and place the game in the recycling bin.

One of the most important advantages of being an indie developer is that you don't have thousands of dollars depending on how well the game does. This will allow you to make you game...unorthodox. Humor is often where this takes people. Their villains are often villains because they do something "not nice." For example, Lyle in Cube Sector is a pretty successful game entirely based around a hoodlum who stole a cat. I'd like to see Master Chief rescue a cat in his next game!

Finally, the last (and one of the most important) part:

Constant interaction with the player is crucial in developing a villain. Of course, this could go either way. If your villain is a hoodlum, it might be best to save their unveiling until the wedding final boss sequence. However, if the villain is apparent from the start, interacting with them could add that extra touch to your game. This goes hand-in-hand with the genre you choose. Puzzle games or other genres with regular breaks from gameplay have to most potential for benefiting from interaction. Likewise, faster paced genres like top-down shooters require more gameplay than cut-scenes.

As far as actual dialogue, the style is entirely up to you.

If you have a less serious game, less serious dialogue should result. Do not take this too far though. The last thing you want is for your audience to make friends with the villain. Instead, the humor should be indirect (like if a villain bosses around a minion, and the minion talks back).

If you have a more serious game, a more serious villain should result. They should never mess up (outside of gameplay, mind you. No boss should be invincible), or be self-centered. Simple character development is the key here.

</wall of text>

All in all, these theories are highly flexible, and only my thoughts. If you think you are fairly decent at understanding a character's roles and character development, chances are you're right. With thought and a little self confidence, your villain (and other characters for that matter) will turn out fine. On that thought, some of the worst villains I've seen are only horrible because the villain style shifts. Like a good argument, choose one type of villain, and stick with it.

Phew, that's it I guess. Sorry if I've digressed from the topic. Thanks for reading!