ESPN is a great example. This year no one predicted even the first round correctly, much less the first two. In fact, only 3 people out of 5 million got only on

ESPN is a great example. This year no one predicted even the first round correctly, much less the first two. In fact, only 3 people out of 5 million got only one game wrong in the first round. The main problem with your logic is that most people make the same choices. For example, only 1% of ESPN brackets had Cornell surviving the first 2 rounds this year, while 73% thought Kansas would make the championship. To my knowledge, no one on ESPN.com has ever predicted all 65 games in the first 2 rounds correctly.