Nutshell Sports r a t i n g s  i n  s l o w  m o t i o n
 

PicoSearch
   Help
Site Search by PicoSearch  
Be notified of
page updates
it's private - Powered by ChangeDetection

 

          Site Guide

Find more sports ratings and other features

My New Project: The Real Happiness Guide

Locations of visitors to this page
(These stats do not include David's Directory)

 

Explanation for Dumb Sportscasters

18 September 2007

Okay, guys, listen up. I know you are experts and you have your opinions but you're a bunch of goofballs. I always hear one or more of you saying something like: How can Buffalo be ahead of UCLA? Well, what you're looking at are probably retrodictive ratings. These generally start from scratch at the beginning of the season. They link games together according to who beat whom and who beat them and whom they beat. It is a logical way of stacking the teams.

As an example, David Wilson's retrodictive ratings on the week I am writing this have as the top ten: Alabama, Kentucky, Florida, Arkansas, South Carolina, Ohio State, Sioux Falls, Nebraska-Omaha, Kent State, Cincinnati. The only reason he publishes them this early is that he assumes everyone will read how they are calculated. Unfortunately, I constantly hear sportscasters who don't do their "expert" research but just sit around and say: Can you believe this guy? Yet, I defy any one of you to prove him wrong. It can't be done because at this early stage every team they beat is listed below them and every team that beat them is listed above them. There is no way to dispute their ranking.

Of course, you guys, being experts, are going to wag your heads and say: What an idiot! That is because you morons have no idea how retrodictive ratings work. Why don't you do a little reading so you won't continue to sound like lamebrains to us, who actually do know what's going on?

Would you like me to educate you? Then here's the first fact you need to know: Retrodictive ratings do not normally use a margin of victory. This is another thing I hear you complaining about. You say how can team A who beat the number 100 team 14-13 be ahead of team B who beat the number ten team 14,000-3? It's because they don't consider margin of victory. Then I hear you complaining about teams running up scores. Which way do you numbskulls want it?

A retrodictive system gets more accurate each time a game is played. Early in the season is not a time to be watching and complaining about retrodictive systems. Their very nature is to be most accurate at the end of the season when they are actually used to pick bowl games - for example, Wes Colley's ColleyMatrix.

Even my retrodictive system, which does not start from scratch and is very similar to David's, gets out of whack early in the season simply because the right teams don't always win when they're supposed to. The week I am writing this, my top ten teams are: Ohio State, Washington, Alabama, Boise State, South Carolina, Florida, Arkansas, Cincinnati, California, Kentucky. No, some of these teams won't be near the top ten at the end of the season. But, again, you cannot provide evidence that I am wrong because the teams that beat a team are above it and the teams that the team beat are below it. There is no way to disprove me.

Now, if you think you can understand that, let me explain how my predictive systems work. My systems are in the manner of predictive systems. These are systems that project how a team will play next week by using the season's information to make a projection. My system, like retrodictives, is designed to be accurate after ten weeks of games have been played. That is why some desperate teams still have high ratings. The top ten for college are: LSU, USC, Ohio State, Oklahoma, Florida, Texas, Oregon, Michigan, California, West Virginia. The obvious question here is: Why is Michigan (1-2) still in the top ten? The answer is that they have seven more weeks to find their correct position. Another part of the overall answer is that they just beat Notre Dame 38-0. These are two teams that nobody understands yet. My system lets teams find their places gradually over a long period of time. If you do not understand this, you will think something is terribly wrong with my system.

Anyway, you experts go back to your sportscasting. Perhaps after a few of these sessions, you will learn enough to sound as if you know what you are talking about to us, who do understand how things work. Better still, just give us your opinions on the games and forget about trying to understand the ratings.

 

Copyright Nutshell Sports 1998-2010
Contact Webmaster