I have to admit that I have for the most part I have been more of an NFL fan than a College Football fan. But thanks to my good friend Donnie, who got me hooked into a college confidence level pool I have become a big college football fan. I mean nothing like getting up on Saturday to catch the 9AM PST Purdue-Northwestern game – watching 12 more hours of football ending it watching the 7PM PST Arizona-Oregon State game.
For those of you who are not familiar with this type of pool, confidence points are used to give different weight to each of the games during a given week. You assign a point value (from 1 to 20 if there are 20 games) to each game based on how strongly you feel about the accuracy of your pick. If your selection is correct, you'll receive points equal to the confidence point value you associated with that game. If you lose, you lose those points. The person(s) with the most points win the pool that week.
The pool we are in has 360 players from all over the country. The annual entry fee is $100. Do the math. That means there is $36,000 to be won throughout the year both on weekly and annual basis. In August, I send a check into some guy in Florida who I have never met in my life. In fact I only know two other guys in the pool, Don, and my brother in law Jeff.
In our pool, you do not have to worry about the point spread – you just need to pick the outright winner.
Well this last week North Carolina goes into Virginia Tech as 16.5 point dogs. Out of the 360 players in the pool, 244 picked VT and as their 20 point game. Only 11 players picked North Carolina –most of them as their only their 1 or 2 point game. One guy picked North Carolina assigning 16 points. And what happens? North Carolina pulls the upset winning the game outright!
So how do you pick and assign points. Good question - as I obviously have no clue. Currently, I am in 73rd place which that plus a $1.50 will get you a cup of coffee at Starbucks (only the top 25 each week and the top 25 at the end of the year are in the money). I have finished on a weekly basis anywhere from 35th to 206th.
As I do not have time to watch film on over 40 teams each week, I have used a balance of technical modeling analysis, point spreads, and looking at some of the sites that provide Monte Carlo simulations to predict the winner of each game.
For example, I use Jeff Sagarin’s, of USA Today, college football ratings. Let me quote Sagarin here describing his method of assigning a rating for each team and how to use it to “predict” the outcome of a game.
“In ELO-CHESS, only winning and losing matters; the score margin is of no consequence, which makes it very "politically correct". However it is less accurate in its predictions for upcoming games than is the PURE POINTS, in which the score margin is the only thing that matters. PURE POINTS is also known as PREDICTOR, BALLANTINE, RHEINGOLD, WHITE OWL and is the best single PREDICTOR of future games. The ELO-CHESS will be utilized by the Bowl Championship Series(BCS).
The overall RATING is a synthesis of the two diametrical opposites, ELO-CHESS and PURE POINTS (PREDICTOR).”
I use this predictor as a guide along with the Vegas spreads and the Monte Carlo simulations to not only pick the games but assign the values to each of the game. Sounds pretty reasonable right?
Well the North Carolina-Virginia Tech is only one example of how it is basically impossible to pick these games using any type of data analysis, using Vegas line as a guide, or any other analytical model. There is no possible (at least rational) model that would have told that someone would pick North Carolina and assign it 16 points.
So I have decided to change my approach this week.
The picture at the top of the article is my cat JoJo. JoJo is going to make my picks for me this week. Here is how he is going to do it.
I am not going to feed JoJo for a day (to all you cat lovers out there; he is way overweight, as you can see, so a day of food won’t hurt him). I have printed out all the games on a big sheet of paper. I am going to take two pieces of cat food kibble and put one on the home team and one on the visiting team for each game. Whichever piece of kibble Joho eats is the team I am going to pick.
I will then take each of JoJo’s pick and put them in a hat. I will pull them out of the hat to determine what points to assign to each game. The first one out of the hat will be my one point game; the second one will be my two point game, etc.
I will report back next week how I did.