To quote the front page
> Less Wrong users aim to develop accurate predictive models of the world, and change their mind when they find evidence disconfirming those models, instead of being able to explain anything.
So, by that logic, one interesting metric of the forum quality would be how often what is posted here makes people change their minds. Of course, most of us change their minds almost all the time, but mostly on some mundane topics and in very small amounts, probably too small to pay attention too. But if something comes to mind, feel free to link a thread or two. Depending on the response, we can even try to measure how influential newer posts are vs. older ones.
EDIT: Feel free to mention the Sequence posts, as well, could be a useful benchmark.
EDIT2: Why specifically changing your mind and not just learning something new? Because unlearning is much harder than initial learning, and we, to generalize from one example, tend to forget the unlearned and relapsed into old ways of thinking and doing. (Links welcome). Probably because the patterns etched in the System 1 are not easily erased, and just knowing something intellectually does not remove the old habits. So, successfully unlearning something and internalizing a different view or concept or a way of doing things is indicative of a much more significant impact than "just" learning something for the first time.