Stephen Bainbridge has a post about crowds and experts. He wonders if there has been a study on whether prediction markets limited to experts in the field do better than prediction markets open to any and all comers.
Here’s my, umm…, prediction: individually, the experts would do better, but the conventional wisdom of the all comers group would outperform most of the individual experts.
Christine Hurt follows up on that thought in the context of the Battle of the Encyclopedia Britannicas and the Wikipedias.
The answer, as far as I am concerned, is that crowds do fine as long as you remember to trust, verify and triangulate. That post by Jim McGee, which I have linked to before, is a compelling argument, at least to me, for the benefits of multiple data points.
And in the blogosphere, multiple data points requires a crowd.