So, If You Lose Early...or Late...What is Going On Here?

1. 2008 is over and done with. let’s stop talking about it
2. if we win all of our games, nothing else matters. just win our games.

I think it is unavoidable to take into consideration “potential” and to speculate based on talent level. We just have too small a sample size. Furthermore, while most of the time, the better team wins, this is not always the case (so Matt Leinart was only wrong in that one instance). Fluke wins and upset wins happen. I don’t think anyone would disagree that USC would beat Washington, and probably badly, 19 out of 20 times on a neutral field and if both teams were 100% healthy. Good teams can have bad luck and bad games against inferior opponents and lose.

Further along in the season, I agree that results should start becoming more and more important as we have more data to look at, as opposed to voters stubbornly sticking as close as possible to their preseason list. However, even there, things get difficult. Let’s say Oklahoma beats Miami this week. We then have this:

BYU > OU
FSU > BYU
Miami > FSU
OU > Miami

Who wants to figure that out?

It’s an imperfect system and always will be. That’s not to say that it shouldn’t be better; there’s no question that the voting system needs an overhaul. However, it will always have a lot of subjective analysis in it, which is why I wish we had a dedicated body of voters who watched every game of the Top 30 teams or so (in other words, it’s practically their jobs) and justified their picks with some arguments every week.

A fun resume ranker, always good for a laugh for me but I respect him for thinking outside the box. He just tries to rank off on field results without preconceived notions like you say, though of course the a > b > c > a chains make that kinda ridiculous.

His poll this week

it’s not possible to list all of them. Note: I’m not blaming the POLLS — I’m blaming the human idiots who blindly fill in weekly ballots.

Head to head play is ignored. Home vs. road venue often is. Strength of schedule is overlooked. You lose in the first two weeks, it’s like it never happened. Lose in the last two weeks and you’re sentenced to oblivion. Worst of all, many voters (obviously) fear they’ll be blackballed if they rank a one-loss BCS team ahead of an unbeaten BCS team. SoS be damned. And every WRONG vote is carried forward, since it’s hard to overcome a poll decision that puts a legit Top 10 team at 18 midway through the season.

There almost a formula to filling out the ballot: Win, the team moves up (but only if somebody less than 10 spots in front of you gets beat). Lose and you fall; 3-5 spots if it’s a close loss to a good team, 5-10 if it’s a decent showing against a respected opponent.

That’s the formula; deviating from it gets your license revoked. Or so it would seem.

1. Problem at this point in the season is that everything being done still reflects the preseason polls, which are so meaningless as to be absurd.
2. As everyone with a brain knows, there isn’t enough relevant data this early to rationally rank most teams. So, some are ranked OK, many are way over- or underrated.
3. The AP poll has often been bad because of its year-to-year sameness. Now, there is some thought put into the ranking process by some voters. But still not enough.
4. The coaches’ votes are a joke. Most don’t even do their own. Those who actually vote, coaches or surrogates, are less informed (and generally more biased) than the writers/broadcasters who make up the AP poll roster.
5. The polls matter because of what they lead to. But it’s a good idea to remember what they were for in the first place: to generate interest in the sport, not to determine which schools get millions from BCS or MNC game selections.

no, you can’t do anything about it to change it. the only way things will change is if the bcs moneyhats find a new way to make more money.

which will only happen when a business case is made for it, I think the following changes are needed:

1. take into account the opponent’s ranking at the time of the game, not where they are at the end of the season. Abilities and player availability changes, thus their effectiveness changes from game to game. I never understood how you can play a top 5 team one week but let it count as a lower SOS if that same team happens to lose a later game. Why is that a penalty on SOS, when you got yourself up for a top 5 team?

2. A loss, or a win for that matter, should be weighted as the season goes on. Just as an early loss might drop you 5 or 10 spots, the same loss should be weighted to allow only a drop of 1 or 2. I’m saying the same loss. An early loss by 10 points, in a close game, etc. is not the same as a complete breakdown in another game losing by 30 etc.

I realize this doesn’t fix everything (or comes close), and that poll bias and lack of voter understanding will still have an effect, but we need something to help cancel out voters’ poor memories.

Now, if you want to discuss an overhaul… I like the system used by the English Premiership. Yeah, it’s soccer and there are much fewer teams, but setting up a tiered system and allowing tier winners and losers to move up and down between seasons, seems like a much better way of letting up-and-comers (Boise State, for example) take the place of what-have-you-done-for-me-lately’s (ahem, aTm). But that’s another ball of wax…