Let's create a new fallacy!
In discussions and debates about character optimization in D&D, one side tends to associate optimization with its most extreme form: that of a character who effortlessly wipes out all challenges without the help of other PCs.
Their argument against optimization is that D&D is a cooperative, teamwork-based game, and that trying to "compete" to be the most powerful character is against the spirit of the game. This is not a bad idea in and of itself, but it's often used as justification against making powerful characters on any scale at all. At worst, the speaker contradicts himself and says that incredibly gimped characters are more in tune with a cooperative game because other players can fill in for the flaws. A gimped character who cannot do anything competently can be more detrimental to a party than an overpowered character because the party's chances of survival decrease, and everyone else needs to use resources to ensure his survival.
But why is it viewed this way? Is it because full casters can step on the toes of noncasters so often that they're encouraged to "reign it in" for the fun of others? Is it because the original munchkins were powergamers who screwed over other PCs? Is it because people associate "optimization and min-maxing" with its greatest extremes of overpowered cheese?
Example story:
Several years ago, I used to have this mindset. One of the players in my games often played min-maxed characters. There were times when he went "overboard," but many times his character wasn't all-powerful. I once got into an argument with him, and I pretty much presented the points above relating to "teamwork and cooperation."
He argued that an effective character (a sorcerer with Save or Suck spells, in his case) was an effective team player: the longer a dangerous monster or opponent remained on the battlefield, the higher the chance it would kill a PC. Reviving a PC cost money, which cut into the group's gp and into the wealth-by-level guidelines. He also pointed out that he only dropped Save or Lose spells on opponents which were giving the party Barbarian/Rogue/noncaster a hard time. He was gauging the threat assessment of encounters and planning accordingly by reserving powerful spells for difficult opponents.
Granted, he was a sorcerer and most of his spells were combat-focused, so he didn't step on the toes of the skill-users. And the melee guys didn't mind because he hung back until things got really bad to break out the big guns. This saved the party's bacon several times.
These points got me thinking, and I realized that my preconceived notions of optimization and cooperative playing were wrong. Although his spells often made a cakewalk of powerful opponents, he was not doing it in a competitive manner, putting his own needs above the group, or otherwise acting like a grognard's view of a "powergaming, rollplaying munchkin."
P.S. Similar points were addressed in the Munchkinmaxer Fallacy thread, but this conversation is meant to be a discussion covering why some gamers view optimization and and teamwork as mutually exclusive.