The Netflix prize was a famous early application of crowdsourcing to predictive modeling.  In 2006, Netflix published customer movie rating data and challenged analysts to come up with a predictive model that would improve Netflix’s prediction of what your rating would be for a given movie.  Various individuals and teams competed, and the winning group was actually an ensemble of various teams.  Although the winner did boost rating accuracy by the required 10%, Netflix ended up not using the algorithm.