Richard’s Ramblings .........

August 16, 2010.

For the good times..... 2010 marks my 41st season ranking college football teams. I remember those early days fondly. Although I’m nearing 60 now, I still have that youthful enthusiasm about college football today that I held when I was a teenager. I find it quite interesting that some of the same teams that battled for National Championships in the late 60’s and early 70’s, Alabama, Texas, Oklahoma, Southern Cal and Nebraska are still contending for top honors today. Maybe it’s true, the more things change, the more they stay the same. I must say though, it is refreshing to see teams like TCU, Boise State, Utah, Brigham Young, and Cincinnati mentioned in the same breath with those traditional powers.

Why in the world did I start this mess in the first place?.......In the beginning, for the most part, I always felt the wire service polls (as they were called back in the old days) did an adequate job in ranking teams, but there were obvious flaws. Opponent strength was never given enough credit, and margin of victory was always given too much. Those flaws are what propelled me on my journey to improve college football rankings. It’s been recorded in many interviews over the years that the Billingsley Report is “an improved AP”. Whether or not it is an improvement is up to the individual, but it is true that my rankings have many of the great qualities of a human poll, like a proven starting position and more emphasis on most recent performance, and it does balance out those obvious flaws by making opponent strength a top priority and not considering margin of victory at all. (The use of margin of victory is a separate discussion all together. One which I’ll get into at another time). But most importantly, as can be said of all the BCS computer rankings, there are no biases in the rankings by team or geography. My computer does not care about the name of a team, or the conference they play in. I wish I could say that about human polls, but unfortunately it’s human nature to consider those things. I’ve said it thousands of times, but I’ll say it again today, the best college football ranking is a combination of human and computer. In that sense, the BCS got it right. Coaches and sportswriters are absolutely needed to analyze the performance of a game based on rivalries, weather, penalties, turnovers, and momentum, but only the computers can fully analyze opponent strength and take those common biases completely out of the equation.

The Billingsley Report is about as close as you will ever find in regards to a combination of  human and computer rankings. That doesn’t mean it’s “better” than other systems, it simply means I was successful in what I set out to do, create a computer generated formula for ranking college football teams that is about as “human” as you can get.  The Billingsley Report correlates much better to the AP, Harris and Coaches Poll than it does to the other BCS computer rankings. Some people appreciate this, others think I should be kicked out of the BCS because of it. The rankings correlate so well in fact that one complaint at times received by BCS headquarters is that “Billingsley should be removed because his rankings are too human”. Go figure. I thought that was a good thing. All I know, is that of the 12 year history of the BCS, the #1 and #2 teams in the Billingsley Report went on to play for the BCS National Championship 11 times. I’m very proud of that. Certainly not a bad record if you consider human polls the gold standard.

Oh those pesky preseason polls......Human Polls may be the standard in ranking college football teams, but at times, their preseason polls do quite a disservice to the sport. Ranking teams based on preseason hype is an unwise (and unfair) thing to do. The Billingsley Report carry’s over a teams rank from one season to the next, fairly, I think, establishing a proven ranking that the teams earned themselves. Many will say personnel changes from one season to the next undermines the effectiveness of this procedure, but history proves me right. Teams do not change in strength from one season to the next as drastically as you would think, and even when they do, it’s often due to a coaching change, not a dramtic change in personnel. Either way, the system quickly adjusts to their new found strength. Check out this list of the ranking history of teams during the BCS era and you will see, it’s not often that a team changes more than 20 positions from one season to the next.  The problem with preseason human polls is not that they have a starting position, it’s that they are “inflexible” in their minds about moving teams up and down once they establish that starting position. My rules of “head to head competition”, the core of my system, gives lower ranked teams ample opportunity to move up by beating higher ranked competition, and starting high in the Billingsley Report gives no advantage unless a team goes undefeated, and in that case, they certainly proved their worthiness of starting high.

If you’ve read this far, you can tell why I titled it “RAMBLING”. Opening game is only 18 days away. I can’t believe after all these years I still get butterflies in the pit of my stomach and lose sleep in anticipation. What a joy. By the way, if you have not signed up yet for the game on “Stock Market Sports”, you should. It’s a lot of fun and a great way to test your knowledge of my system, and your upset picking skills!