Can you create a method to predict future geopolitical outcomes that is better than current state-of-the-art methods?