Welcome to March Madness! I'm a huge sports fan, but I have to admit, I'm not a massive follower of college sports. I generally root for 3 teams. The first is UNLV. I adopted them when my parents retired here and my own alma mater was a Division 3 team (and not a very good one at that). The second is my alma mater, which somewhere along the way became Divsion 1. They aren't a powerhouse, but they are in the big tournament this year. Let's go SUNY Albany! If either the men's or women's team is still in it by the time you read this, the world may be ready to come to an end! The third team is whoever is playing Duke. It's a long story, but let's just say it goes back to the days when UNLV last played Duke in the Championship game.
I have no doubt that phrase madness mostly refers to the non-stop focus on the tournament for the next three weeks. But, it probably should refer to one thing in particular. That is, the probability of putting together a 'perfect' set of picks. There are 63 games in the tournament (once they get down to 64 teams). If each of the games were a coin toss (50% chance of either team winning), the odds of picking a perfect bracket would be a number that I don't even know how to name. It would be a 9 with EIGHTEEN zeros after it. The best I can describe it is 9223 Quadrillion (no clue what comes after Quadrillions!).
This kind of puts a new spin on how organizations can offer big prizes for a perfect set of picks. I wonder if they even bother to insure it? How much would an insurance company really charge for a 9223 Quadrillion for 1 chance that pays 1 million dollars? The expected value is 0.0000000000001.
Of course, the odds of each game is far from 50%. Even under the concept that on any given day, any given team can beat another, the probability is not 50%. When a 1 seed plays a 16 seed, it's probably a 90% chance that the number 1 seed wins. If I recall correctly, no #16 seed has ever made it out of the 1st round, so one could argue that the probability should be closer to 100%. Of course, when a number 9 plays a number 8, the 50/50 rule is more in play. The rest of the matchups will average something between 50 and the 90 (or 99%).
So, what if someone could accurately predict the winner 60% of the time? What would that do to our odds? That will bring it down to a number I can at least properly name. The odds would be 1 in 94.7 trillion. At 70%, it comes down to a 'realistic' 5.7 billion. I'm not sure the moneyline on very many of these games would reflect a 70% favorite, especially after the first round. These numbers all assume that the favorite wins or that someone out there has insight well beyond that which the guys who set the lines do. It's hard to see this happening.
As I write this column, the first day of games are wrapping up. Two 14 seeds, a 10 seed and an 11 seed have all already won in the first round. Some of these teams were serious longshots. More than likely, very few people had picked any of these teams. How many have picked all 4 of these upsets? From what I can see on Yahoo sports, it would appear that there are 25 people who still have a perfect bracket. I find even this amazing. More often than not, by the end of just the first weekend, there is no one left standing in the perfect category. With 32 games and even the ability to pick the winner 60% of the time, the odds are still more than 12.5 million to 1 of picking the winner in JUST the 1st round games, never mind the first two rounds that are played this weekend.
If you're a serious college basketball fan, however, you probably are in more than just one of those free contests on-line. You've probably picked in a pool with your friends or your office buddies where the person who gets the most points wins. Your odds of winning a pool like this is far greater than your chances to get a perfect set of picks. It kind of reminds me of that joke about a bunch of hunters who stumble upon a bear in the woods. As they stand frozen, one of them asks the others if they think they can outrun the bear. Another responds that he doesn't have to outrun the bear, he only needs to outrun the other hunters!