Where do you think Less Wrong is most wrong?

That it's not aimed at being "more right" -- which is not at all the same as being less wrong.

To be more right often requires you to first be more wrong. Whether you try something new or try to formulate a model or hypothesis, you must at minimum be prepared for the result to be more wrong at first.

In contrast, you can be "less wrong" just by doing nothing, or by being a critic of those who do something.' But in the real world (and even in science), you can never win BIG -- and it's often hard to win at all -- if you never place any bets.