Verdict?

[Arlie Rahn]

A combination of prestige, perceived talent level and some prior performance is what happens in real life. Look at West Virginia in real life. They finished with 9 wins in 2002 (bowl), 8 wins in 2003 (Jan 1 bowl), 8 wins in 2004 (Jan 1 bowl) and yet they still only received one vote in the 2005 AP preseason poll. Now, they ended up finishing with 11 wins and a No. 5 ranking overall. But, do you think a team like Michigan, Tennessee or Iowa would have had only one vote with that track record and incoming team? Prestige matters in BBCF for preseason polls, just like it matters in real life. That's why unranked teams at season's end like Tennessee, Michigan, Iowa and Purdue were ranked #3, 4, 11 and 15 at the start of the season and a team like West Virginia was sitting below Miami (OH), Colorado State and Wyoming with only 1 vote.

[Druez]

[QUOTE=Miral]Druez says this isn't a valid test case, I don't see why not. If a team is stacked with returning talent after a superb year (not necessarily this inflated example), why should it get absolutely no recognition at all for it in the preseason poll, even though the chance of another end of year top 25 is almost guaranteed? The point I am trying to make, is that if at the end of the season a team like Pitt or WV usually wind up in the top 25 2 out of every 3 years, why at the beginning of each new year are they down around 40th? Its kind of like groundhog day. Start the season ranked 40th, climb to 20th, start next season at 40, climb to 20th, etc. I hope you see my point, the preseason poll penalizes smaller schools in this way, while the end of season one does not. I feel the preseason poll doesn't take the end result of a team's performance each of the last few seasons strongly enough. I suppose I should stop complaining before Map takes my fanboy licensce away. :)[/QUOTE] In general Miral, preseason polls penalize smaller schools in real life. If you have seasons like you mentioend above your prestige will soon be in the 50's or above and you will be ranked and not just be a one season wonder.

[SkyDog]

This all brings me back to my longstanding point: when trying to model a system that is clearly flawed in real life, there [U][B]will[/B][/U] be a good number of people who are unhappy with some of the results.

[Miral]

[quote=Druez] preseason polls penalize smaller schools in real life.[/quote] In general yes, but not always. Take Arlie's example of WV the past few seasons. I doubt they will be completely off the radar for the upcoming preseason polls this autumn year considering what they acheived last year with such youth at key positions. The example of Tennessee and Michigan is a different example, both had good teams with lots of talent (no one will argue that), but underperformed (especially the vols) (no one will argue this either). Nobody expectes them to uderperform that bad again so they will be back in the top 25 with the preseason polls. Of course Skydog's assessment is correct, polls are themselves inherently flawed from top to bottom, so how can one program the perfect poll? We can merely hope to tweak it to make it a less imperfect creature.

[alig4321]

[QUOTE=Miral]In general yes, but not always. Take Arlie's example of WV the past few seasons. I doubt they will be completely off the radar for the upcoming preseason polls this autumn year considering what they acheived last year with such youth at key positions. The example of Tennessee and Michigan is a different example, both had good teams with lots of talent (no one will argue that), but underperformed (especially the vols) (no one will argue this either). Nobody expectes them to uderperform that bad again so they will be back in the top 25 with the preseason polls. Of course Skydog's assessment is correct, polls are themselves inherently flawed from top to bottom, so how can one program the perfect poll? We can merely hope to tweak it to make it a less imperfect creature.[/QUOTE] Surely it depends on what the aim is. Is the aim to model the real life flawed system or is it to better this system and create one that works perfectly yet does not reflect the way the real life system works? From what Arlie has said, I believe he is trying to recreate the system that is currently in place, meaning he has done it exactly right.

[John Comey]

For what it's worth, in the pre-pre-preseason Top 25, West Virginia was ranked third in the nation. This was based on what they had returning. Ohio State was #1 I believe, but WVU is #3.