Both UAB and Troy ST are up and comers in I-A and will be playing very good footbal soon. Troy State has played some close games and won some big games last year. I disagree with using computer rankings anways because they never give the true story when it comes to how good teams are. This goes for all of NCAA teams.
I agree that Troy St did very well. I was quite surprized, especially with last years results (I live in Nebraska, so when Troy State played so well I was watching intensely). With the money Troy State earned by playing big name teams, they should be able to build on their great 1AA record. But until they win this year against a 1A team, they will be in the bottom 25. Note they are far ahead of many 1AA teams - it isn't like Troy State is a Canisius or Valparaiso.
How do you rank teams like Troy State without a computer? When you play half of your schedule in 1A and half in 1AA it becomes really hard to judge a team by any other method. When it comes to the Coach/Press voting, there is no way to differentiate between all the teams with 0 votes. How would voters even know if beating Austin Peay was a good feat or a piece of cake? They aren't in a conference so they can't be ranked by confernce wins. I just don't know of any other way of doing it. What is your suggestion?
No ranking is ever perfect. At least I openly admit my error margins. If I optomize for the best prediction of a game winner, I'll get it right 84% of the time but the score predictions suck. If I optomize for the best prediction of the game score, I'll be off by 9.5 points but the winner prediction plummets. Instead I go in between: I'll pick the winner right 80% of the time and the score will be +- 10 points. So Troy State will be beaten 80% of the time by the teams ranked above it. Yes 20% of the time Troy State might win. There is no perfect ranking system - with teams playing each other twice each winning once, there is no possibility of a 100% accurate ranking. This is just the best way I know how to get it done.