It's funny having Florida and LSU on there, considering during the BCS era both are 2 of the top 4 programs (USC and Alabama are also in there).

It was a decent article, but as others have touched on, the outside the top 25 thing is really flawed, and you're also a lot more penalized for having a great program who is consistently in the mix. I mean, Florida has made 7 BCS games in 15 years, going 5-2 and winning 2 titles. They have also played in 7 SEC Championship games in that same 15 year span. I guess another clear flaw is it really penalizes you for having a down season, and REALLY being down.

That's basically as dominant as you can get in a conference, especially one that has been as strong as the SEC. Yet they finish 4th on the list. I guess the five years during this 15 year span, they also had 5+ losses, including a 6-loss season. It's hard to say they were really overrated in most of those years, especially after Muschamp took over.

The "list" is just about getting people to talk. The criteria is arbitrary, but it accomplishes its purpose....and puts ESPN's name out there, which is what they want. There is nothing wrong with that either, as it fills the void till practice starts.

Agree with Azubuike about Florida. They were probably ranked high in Zook's 3 years that were all 5 loss seasons. Meyer went in and won a BCS trophy with Zook's recruits, that's how good Meyer is. Muschamp was 7-6 his first year at Florida, not surprising because Meyer told him 'the program was broken'. 11-1 his 2nd year was much better than expected until the bowl loss to Louisville.

It only makes sense that the more successful programs would be in this list. When you are in the top 10 year after year, you typically only have one way to go, and that is down. It actually proves just how good OSU was during that time period. They were ranked high quite often and didn't fall far if at all.

Azubuike24;1448149 wrote:It's funny having Florida and LSU on there, considering during the BCS era both are 2 of the top 4 programs (USC and Alabama are also in there).

Replace USC with OSU and I'll agree with you.

OSU has the most appearances, most wins, has absolutely dominated their conference, and has a national title that they didnt vacate.

I'd add Boise St to the list...only because I hate them and if they played in a major conference, they'd be on par with Indiana.

Pick6;1448193 wrote:Replace USC with OSU and I'll agree with you.

OSU has the most appearances, most wins, has absolutely dominated their conference, and has a national title that they didnt vacate.

We can call it top 5...but agreed. Not to mention last year, where OSU would have had another shot at a Rose Bowl...

I'd like to see a most underrated list as well.

If you reversed the criteria for underrated, you would probably just see a bunch of teams who made a random BCS appearance or 2, when nobody expected them to. Louisville would probably be there. Auburn as well, given they had 2 unbeaten seasons in the BCS era, both winning BCS games and both times being ranked outside of the top 15 to start the season.

GoPens;1448205 wrote:I'd add Boise St to the list...only because I hate them and if they played in a major conference, they'd be on par with Indiana.

whoa, I haven't seen Indiana winning any BCS games recently

Boise is probably near the top of the most underrated. If the criteria is AP poll for sure. Many, many years they climbed in the rankings significantly. Even other years where they dropped a game or 2, they usually end up in the top 10-15 anyway.

Iliketurtles;1447815 wrote:You're right it logical to say that just because a team didn't finish in the top 25 that they automatically finished 26th. That is pretty much the main and most obvious flaw.

That is one major flaw since it creates an artificial floor effect. I can think of a few more related flaws.

1.This doesn't account for regression to the mean; the schools that are consistently ranked very high are more likely to fall back to a lower number. Not only will this occur between schools in a given season, but this will happen within a school between seasons (e.g. Alabama in 2010).

2. There is a ceiling effect since a ranking cannot discern between say the 2001 Miami team vs. a subpar #1 team. Therefore, the assumption is that a ranking in a given year is equal to that same ranking every other year.

3. Related to regression to the mean and the ceiling effect, highly ranked teams can only regress. So statistically, most top teams are bound to regress back anyways; however, if a team actually improves, the best it can do is stay in the position.

4. Rankings are ordinal but this makes the assumption that this is on an interval scale and there is equal spacing between rankings. Some years it is a toss-up between 1 and 10 while other years there is only 2 dominant teams. Basically, the difference between 1 and 10 in one year may be less than the difference in 2 and 3 the next.

5. Then you have to take into considerations the subjective nature of polling and the biases of pollsters. Pollsters often feel the need to arbitrarily drop a team so many rankings, especially if they were a top team. Furthermore, a team that is 9-3 that is perceived as an underachiever is often looked at more harshly by pollsters than a 9-3 that is perceived as an overachiever.

Overall, this is a quick and interesting way to look at the schools, but such a simple analysis will always be biased against schools that are traditionally ranked at the top in the preseason.

Another easy analysis that can account for some of the glaring flaws would be to look at the the preseason top 25 and the gains or losses in rankings by each initial ranking to identify an expected postseason ranking. For a more statistically sound analysis, regress preseason ranking on postseason ranking to get an expected value for the ranking.

Basically if preseason #1 generally loses 2 rankings by postseason, then a preseason #1 that stays #1 would actually have a +2 for a given year rather than a 0.

You have to be ranked preseason top 10 in order to fall 15 spots or more in a season. Notre Dame has only been preseason top 15 once in the last 10 years I think.