Will the Wisdom of the Crowds compete with Polling?

There was a very interesting political experiment initiated before the election by a group in the San Francisco bay area to tap what has been labeled “the wisdom of the crowds,” after the bestseller by James Surowiecki.

Well in advance of the election, Predict06 asked the general web community to make educated predictions about who would win any of the US House and Senate seats that truly were in play. It started out tilting towards the Republicans because some conservative bloggers and online groups first began to take part. But as the progressive blogosphere and netroots types learned about it, the predictions evened up. On the eve of the election, it was looking pretty interesting…..

The results? A full 84,501 predictions were made and the predicted results were amazingly close to the actual results. The predicted Senate was 50/50. The actual Senate, 51 Dems to 49 Republicans. The predicted House: 231 Dems to 204 Republicans, with the D’s picking up 28 seats The actual was: 229 to 196, with the D’s taking 30.

You have to wonder whether this experiment could be refined in successive cycles to evolve into a pretty accurate tool that might compete with the current generation of public opinion polling. At the very least, this is a tool that is worth playing around with. Check it out:

Peter Leyden