Saturday, February 6, 2010

Let's reason about significance of Vegas for just moment

I happen to know a lot of betting men at work and through friends. They, like most others, have been swept up in the hoopla about the Colts during this past week. They have contributed towards that strange figure everybody is talking about: 68.5% of all SB44 betting money has been placed on the Colts.

I am proud to say that I have deeply troubled these betting men with my assessment of two great computer simulations of the game.

  1. The EA sports Madden 2010 simulation which picks the Saints to win 35-31
  2. The PredictionMachine.com simulation which says that there is a 79.4% chance that the Colts will not cover the 4.5 point spread. {Remember, the Colts have to win by 5 or more points, or you loose money.}
What ever in the world do these two points mean? First, let's recognize the most obvious fact: If Colts should happen to win by 5 or 6 points, there will be a financial bloodbath in Las Vegas. 68.5% of the betting shares will have to be paid out. This is at a time when Vegas is deeply injured by recession.

Fortunately for Vegas, the odds of a financial Bloodbath are very low. According to the most Colt-friendly computer projections, there is a 79.4% chance that that Colts will not cover the spread. Let's not forget that the Madden 2010 simulation picks the Saints to win straight up.

Just to refresh you memory a bit, PredictionMachine.com stated that the Colts won only approximately 51% of the matches in their 50,000 match series. The Saints won approximately 49% of the time. Less than 1 point per game differentiated these two teams, when all scores were averaged. In 60% of their simulated victories, the Colts failed to cover the spread of 4.5 points. I remind you that this is the most Colt-friendly interpretation of the performance data you will find.

The correct interpretation of this data is as follows: This is an unbelievably tight match up of equally great, and similar, football teams. There is a razor thin difference between them. This match could go either way, depending on the breaks. The odds of a laugh-in or a blow-out are extremely low. There are really only two ways we can see a blow-out in this game:
  1. Serious injuries would have to be sustained early by one side.
  2. One side would have to come out and lay an egg, really stinking up the joint with serious errors.
Most experts and the software do not predict either event. PredictionMachine.com says that there is a 90% chance that the Colts won't win by 14. It also says that there is a 90% chance that the Saints won't win by 14. Remember, there is only a 20.6% chance that Colts can even cover the spread.

Vegas knows this. Vegas has intentionally set the spread so that they have an overwhelming (at least 79.4%) chance of collecting that 68.5% of betting money that went down on the Colts. They are in this business to make money, not to lose money. It does not pay Vegas to pick the winner. Vegas wins when you lose. The spread has been set to pretty well ensure you will lose your bet on the Colts. Vegas wasn't built on winners. Vegas was built on losers.

With that said, it should be noted that former Colts coach Tony Dungy has stated publicly that he does not believe SB44 will be close. He says it will be a snoozer with the Colts winning by 2 TDs. The Colts will hold the lead all game long.

When asked why he thinks that, Dungy states that Vikings laid an egg with 7 turnovers, and the Saints still only won the game by a field goal in overtime. The Colts will not turn over the football 7 times. The Colts, on the other hand, wasted the most powerful defense in the league. Ergo the Colts deserve to be there and the Saints don't?

You'll pardon me for saying so, but this is the purest example of human bias I have ever seen. Not only does his paternal pride show through, but he hyper-extends just one point of data when are 17 others to consider for each team. There is a way to hyper-extend one point of data such that you break your neck and your back. This is what human biases do to you. Don't buy off on it.

Why I am not concerned about the Championship games and their outcomes? I'll tell you why:
  1. The Jets sustained serious injuries in their secondary during the course of the AFC champions game. This is why their coverage broke down, and they were forced to stop blitzing.
  2. The Jets may have been the #1 defense in the league this season, but that isn't saying a whole lot. There was no legendary defense in the NFL during 2009. The Jets' defensive statistics do not compare favorably with those of the great defenses of the past.
  3. We knew the Jets did not have the offensive fire-power to keep pace with the Colts.
  4. The Saints began to play in a very odd, and conservative manner once the Vikings began to turn over the football. It is as if their confidence rose to a point where they did not feel they needed to be aggressive or take risks. They were confident that the Vikings were self-destructing. They took it easy, presuming the win was in the basket. This resulted in a much closer match than I expected.
  5. I expect to see the same Saint team we saw against Arizona. This is a Saint team that is greatly concerned about the scoring prowess of their rival. I expect the Saint offense to swing for the fences and attack ruthlessly all game long. They know they are going to have to outscore the Colts in a scoring match. Even if the Colts begin to drop the ball, I expect the Saints to attack viciously and relentlessly.
So there you have it. If you took the Colts and the points, you are a stupid sucker. The most likely outcome of this game is that the last team to have possession of the football will probably win by a last-second field goal in regulation. This scenario cannot result in the spread being covered.