Pre-Season Polls?

[Milktruck74]

I was reading a thread on another board about how worthless the pre-season polls are and how the voters are usually uninformed homers.....thus, regionally biased poles... blah, blah, blah. I thought it would be interesting to try and quantify the relevance of pre-season poles by looking at the percentage of teams ranked in the pre season top 25 and seeing how many were there at the end of the season....I also thought there needs to be some type of way to weight the teams that are there (ie. #1 at preseason that is #4 post should be worth more than #1 that slips to #23). I feel the AGS pollsters take more time and have more overall knowledge than any other poll out there (including the bigs)...that is why I no longer vote in it. I did not feel like I had the necessary knowledge outside of my region to keep up with the rest of you. Hopefully someone who loves to analyze statistics will take this idea and run with it!

[MR. CHICKEN]

19379.........WE DO GOOD.....'CAUSE WE LISTEN....TA EACH OTHERAH...........WE DEPEND ON DUH LADS......FROM ALL CONFERENCES.....WHO GIVE HONEST ASSESSMENTS......VERAH LITTLE HOMERISM.......YA CAN OWN-LAH LEARN BAH....LISTENIN'...........WE TRUST....EACH OTHERAH................xhighfivex.......xnodx..... ...xhandshakex........BRAWK/OBAMA/BIDEN!!

[superman7515]

I was reading a thread on another board about how worthless the pre-season polls are and how the voters are usually uninformed homers.....thus, regionally biased poles... blah, blah, blah. I thought it would be interesting to try and quantify the relevance of pre-season poles by looking at the percentage of teams ranked in the pre season top 25 and seeing how many were there at the end of the season....I also thought there needs to be some type of way to weight the teams that are there (ie. #1 at preseason that is #4 post should be worth more than #1 that slips to #23). I feel the AGS pollsters take more time and have more overall knowledge than any other poll out there (including the bigs)...that is why I no longer vote in it. I did not feel like I had the necessary knowledge outside of my region to keep up with the rest of you. Hopefully someone who loves to analyze statistics will take this idea and run with it!


I've actually been working on this a little bit with my History of the AGS Top 25 spreadsheets. It will be a little bit before I am finished, but one little snippet that may be of interest:

Only one team that was not ranked in the preseason Top 25 has ever won the national championship: 2004 James Madison. Maybe because it was the first year of doing the poll? Or maybe because it was a fluky win... Only two teams have ever started the season ranked #1 and finished unranked... 2011 Eastern Washington and 2005 James Madison (coming off the aforementioned unexpected national championship win).

The point being that the chances are exceedingly good that whoever will win this years national championship is already ranked.

[ursus arctos horribilis]

I was reading a thread on another board about how worthless the pre-season polls are and how the voters are usually uninformed homers.....thus, regionally biased poles... blah, blah, blah. I thought it would be interesting to try and quantify the relevance of pre-season poles by looking at the percentage of teams ranked in the pre season top 25 and seeing how many were there at the end of the season....I also thought there needs to be some type of way to weight the teams that are there (ie. #1 at preseason that is #4 post should be worth more than #1 that slips to #23). I feel the AGS pollsters take more time and have more overall knowledge than any other poll out there (including the bigs)...that is why I no longer vote in it. I did not feel like I had the necessary knowledge outside of my region to keep up with the rest of you. Hopefully someone who loves to analyze statistics will take this idea and run with it!

I've seen your votes for years now and making that statement is incorrect so get off yer ass and get back into the voting/discussing mix and learning from others as MR. CHICKEN just pointed out.

You're a good voter truck and as you know the SoCon peoops need to fill in where their former mates have left off or there will be way too low a turnout from one of the top conferences.

[clenz]

I'll do some working on something later tonight... Might not be exactly what is asked for in the op, but it will be something

[chattanoogamocs]

I've seen your votes for years now and making that statement is incorrect so get off yer ass and get back into the voting/discussing mix and learning from others as MR. CHICKEN just pointed out.

You're a good voter truck and as you know the SoCon peoops need to fill in where their former mates have left off or there will be way too low a turnout from one of the top conferences.

I have often felt the same way...because I respect the job AGS voters do in knowing all of FCS, I have never felt comfortable with my overall knowledge outside of the SoCon and maybe OVC/Big South.

I assume I am eligible due to my longevity and post count, but honestly, I am not sure where to even begin if I wanted to vote.

[superman7515]

I was reading a thread on another board about how worthless the pre-season polls are and how the voters are usually uninformed homers.....thus, regionally biased poles... blah, blah, blah. I thought it would be interesting to try and quantify the relevance of pre-season poles by looking at the percentage of teams ranked in the pre season top 25 and seeing how many were there at the end of the season....I also thought there needs to be some type of way to weight the teams that are there (ie. #1 at preseason that is #4 post should be worth more than #1 that slips to #23). I feel the AGS pollsters take more time and have more overall knowledge than any other poll out there (including the bigs)...that is why I no longer vote in it. I did not feel like I had the necessary knowledge outside of my region to keep up with the rest of you. Hopefully someone who loves to analyze statistics will take this idea and run with it!

Just a start but...



In 2004, 11 of 25 (44%) teams from the preseason poll finished in the final Top 25. All five of the preseason Top 5 (100%) and eight of the preseason Top 10 (80%) finished ranked.


In 2005, 13 of 25 (52%) teams from the preseason poll finished in the final Top 25. Four of the preseason Top 5 (80%) and six of the preseason Top 10 (60%) finished ranked.


In 2006, 17 of 25 (68%) teams from the preseason poll finished in the final Top 25. All five of the preseason Top 5 (100%) and nine of the preseason Top 10 (90%) finished ranked.


In 2007, 16 of 25 (64%) teams from the preseason poll finished in the final Top 25. All five of the preseason Top 5 (100%) and nine of the preseason Top 10 (90%) finished ranked.


In 2008, 15 of 25 (60%) teams from the preseason poll finished in the final Top 25. Four of the preseason Top 5 (80%) and seven of the preseason Top 10 (70%) finished ranked.


In 2009, 17 of 25 (68%) teams from the preseason poll finished in the final Top 25. All five of the preseason Top 5 (100%) and all ten of the preseason Top 10 (100%) finished ranked.


In 2010, 15 of 25 (60%) teams from the preseason poll finished in the final Top 25. Four of the preseason Top 5 (80%) and six of the preseason Top 10 (60%) finished ranked.


In 2011, 13 of 25 (52%) teams from the preseason poll finished in the final Top 25. Three of the preseason Top 5 (60%) and eight of the preseason Top 10 (80%) finished ranked.


In 2012, 16 of 25 (64%) teams from the preseason poll finished in the final Top 25. All five of the preseason Top 5 (100%) and eight of the preseason Top 10 (80%) finished ranked.


In 2013, 12 of 25 (48%) teams from the preseason poll finished in the final Top 25. Three of the preseason Top 5 (60%) and five of the preseason Top 10 (50%) finished ranked.




I would say that just from the basic look, 2009 was the best year for the AGS preseason poll and 2013 was the worst year (2004 had fewer Top 25 teams correct, but the Top 10 was much better).

[clenz]

I have often felt the same way...because I respect the job AGS voters do in knowing all of FCS, I have never felt comfortable with my overall knowledge outside of the SoCon and maybe OVC/Big South.

I assume I am eligible due to my longevity and post count, but honestly, I am not sure where to even begin if I wanted to vote.

Very very few are "comfortable" outside of their home conference, especially after the top 3 or 4 teams.

The preseason poll is tough to do when you're just starting voting, but what it really does it make you pay attention to the FCS more. The first season is tough, but after that it gets easier, and really by the end of season 1 you have a real good grasp of things.

[clenz]

Just a start but...



In 2004, 11 of 25 (44%) teams from the preseason poll finished in the final Top 25. All five of the preseason Top 5 (100%) and eight of the preseason Top 10 (80%) finished ranked.


In 2005, 13 of 25 (52%) teams from the preseason poll finished in the final Top 25. Four of the preseason Top 5 (80%) and six of the preseason Top 10 (60%) finished ranked.


In 2006, 17 of 25 (68%) teams from the preseason poll finished in the final Top 25. All five of the preseason Top 5 (100%) and nine of the preseason Top 10 (90%) finished ranked.


In 2007, 16 of 25 (64%) teams from the preseason poll finished in the final Top 25. All five of the preseason Top 5 (100%) and nine of the preseason Top 10 (90%) finished ranked.


In 2008, 15 of 25 (60%) teams from the preseason poll finished in the final Top 25. Four of the preseason Top 5 (80%) and seven of the preseason Top 10 (70%) finished ranked.


In 2009, 17 of 25 (68%) teams from the preseason poll finished in the final Top 25. All five of the preseason Top 5 (100%) and all ten of the preseason Top 10 (100%) finished ranked.


In 2010, 15 of 25 (60%) teams from the preseason poll finished in the final Top 25. Four of the preseason Top 5 (80%) and six of the preseason Top 10 (60%) finished ranked.


In 2011, 13 of 25 (52%) teams from the preseason poll finished in the final Top 25. Three of the preseason Top 5 (60%) and eight of the preseason Top 10 (80%) finished ranked.


In 2012, 16 of 25 (64%) teams from the preseason poll finished in the final Top 25. All five of the preseason Top 5 (100%) and eight of the preseason Top 10 (80%) finished ranked.


In 2013, 12 of 25 (48%) teams from the preseason poll finished in the final Top 25. Three of the preseason Top 5 (60%) and five of the preseason Top 10 (50%) finished ranked.




I would say that just from the basic look, 2009 was the best year for the AGS preseason poll and 2013 was the worst year (2004 had fewer Top 25 teams correct, but the Top 10 was much better).I'll cancel what I was doing, because this is better and more concise.

[bonarae]

Because I am not that comfortable with how different the final polls are from the preseason ones (since I have limited working knowledge of FCS football outside of the Northeast), I do not vote in the preseason poll and have not been for the past few seasons now.

Yes, it was a good thing to move the preseason polls to July. May was way too early for such prognostications.

[superman7515]

The number 1 team in the preseason has finished in the Top 25 in 8 of 10 seasons (80%).



The number 2 team in the preseason has finished in the Top 25 in 10 of 10 seasons (100%).



The number 3 team in the preseason has finished in the Top 25 in 8 of 10 seasons (80%).



The number 4 team in the preseason has finished in the Top 25 in 9 of 10 seasons (90%).



The number 5 team in the preseason has finished in the Top 25 in 8 of 10 seasons (80%).



The number 6 team in the preseason has finished in the Top 25 in 6 of 10 seasons (60%).



The number 7 team in the preseason has finished in the Top 25 in 7 of 10 seasons (70%).



The number 8 team in the preseason has finished in the Top 25 in 7 of 10 seasons (70%).



The number 9 team in the preseason has finished in the Top 25 in 6 of 10 seasons (60%).



The number 10 team in the preseason has finished in the Top 25 in 8 of 10 seasons (80%).



The number 11 team in the preseason has finished in the Top 25 in 4 of 10 seasons (40%).



The number 12 team in the preseason has finished in the Top 25 in 6 of 10 seasons (60%).



The number 13 team in the preseason has finished in the Top 25 in 4 of 10 seasons (40%).



The number 14 team in the preseason has finished in the Top 25 in 4 of 10 seasons (40%).



The number 15 team in the preseason has finished in the Top 25 in 7 of 10 seasons (70%).



The number 16 team in the preseason has finished in the Top 25 in 5 of 10 seasons (50%).



The number 17 team in the preseason has finished in the Top 25 in 7 of 10 seasons (70%).



The number 18 team in the preseason has finished in the Top 25 in 6 of 10 seasons (60%).



The number 19 team in the preseason has finished in the Top 25 in 7 of 10 seasons (70%).



The number 20 team in the preseason has finished in the Top 25 in 1 of 10 seasons (10%).



The number 21 team in the preseason has finished in the Top 25 in 4 of 10 seasons (40%).



The number 22 team in the preseason has finished in the Top 25 in 4 of 10 seasons (40%).



The number 23 team in the preseason has finished in the Top 25 in 4 of 10 seasons (40%).



The number 24 team in the preseason has finished in the Top 25 in 3 of 10 seasons (30%).



The number 25 team in the preseason has finished in the Top 25 in 4 of 10 seasons (40%).





So the number 2 team in the preseason poll (North Dakota State) is a lead pipe lock to finish in the Top 25 and the number 20 team (Montana State) does not have history on their side.

[clenz]

I wonder what it is about the 20 spot that is so difficult to predict?
Funny that 15-19 have better odds than 11-14 even though most will say "the first 15 was pretty damn easy, it the rest that was hard".

[superman7515]

Funny that 15-19 have better odds than 11-14 even though most will say "the first 15 was pretty damn easy, it the rest that was hard".

Just hypothesizing, maybe because people already have their minds made up a bit about their top 15 and actually have to look into it a bit to find the rest.

[clenz]

Just hypothesizing, maybe because people already have their minds made up a bit about their top 15 and actually have to look into it a bit to find the rest.

I suppose that makes sense.

[chattownmocs]

The data shows that the poll is terrible. Sorry ursus

[clenz]

The data shows that the poll is terrible. Sorry ursus
On the contrary, I'd be willing to bet that the data shows the AGS poll is better than any other preseason poll there is.

[Mattymc727]

The data shows that the poll is terrible. Sorry ursus

I think the data shows that Chattanooga sucks and wont make the playoffs again, regardless of preseason rank and your effort to make us all think they are the best team in the land.

[Professor Chaos]

The data shows that the poll is terrible. Sorry ursus
....says the guy who's #2 team (apparently barely #2 and almost #1) in his 2013 preseason poll ballot (http://www.anygivensaturday.com/showthread.php?135388-AGS-Poll-Results-Preseason-2013&p=1982886&viewfull=1#post1982886) failed to make the final AGS top 25 (https://docs.google.com/spreadsheet/ccc?key=0AgtKAORz8z6vdHlWRmhnSWFTNjR3eWhXNUFDSjJsa 3c&usp=sharing#gid=9).



1: North Dakota State Bison
2: Chattanooga Mocs
3: Montana State Bobcats
4: Eastern Washington Eagles
5: Central Arkansas Bears
6: Sam Houston State Bearkats
7: Stony Brook Seawolves
8: Towson Tigers
9: Montana Grizzlies
10: Northern Iowa Panthers
11: New Hampshire Wildcats
12: Villanova Wildcats
13: South Dakota State Jackrabbits
14: Youngstown State Penguins
15: Richmond Spiders
16: Samford Bulldogs
17: Cal Poly Mustangs
18: Coastal Carolina Chanticleers
19: Illinois State Redbirds
20: Eastern Illinois Panthers
21: Indiana State Sycamores
22: Wofford Terriers
23: Tennessee State Tigers
24: Jacksonville State Gamecocks
25: Tennessee Tech Golden Eagles
Speaking of that none of the teams bolded made the final AGS top 25 which was 12 of the 25 teams on your ballot. It's a wonder the AGS poll is as good as it is with ballots like yours in it.

Although, if it's any solace, having Chatty at #2 wasn't your worst gaffe in that ballot. Having an Indiana St team that finished 0-11 vs D1 competition at #21 takes that prize.

[chattownmocs]

....says the guy who's #2 team (apparently barely #2 and almost #1) in his 2013 preseason poll ballot (http://www.anygivensaturday.com/showthread.php?135388-AGS-Poll-Results-Preseason-2013&p=1982886&viewfull=1#post1982886) failed to make the final AGS top 25 (https://docs.google.com/spreadsheet/ccc?key=0AgtKAORz8z6vdHlWRmhnSWFTNjR3eWhXNUFDSjJsa 3c&usp=sharing#gid=9).


Speaking of that none of the teams bolded made the final AGS top 25 which was 12 of the 25 teams on your ballot. It's a wonder the AGS poll is as good as it is with ballots like yours in it.

Although, if it's any solace, having Chatty at #2 wasn't your worst gaffe in that ballot. Having an Indiana St team that finished 0-11 vs D1 competition at #21 takes that prize.

And yet I had as many teams in the final poll as the preseason AGS poll did apparently.

[superman7515]

I missed four teams somehow, so I will go back and double check each conference later on, but 4 out of 250 will still give you a fairly decent idea of how it breaks out. For some reason I feel I may have ignored Old Dominion unintentionally, which would explain two of them, but I'm not positive. Anyway, I'll go back and when I figure out where I missed them, I'll post those conferences that were updated.

The Atlantic 10 Conference had 8 of 17 (47%) teams that started in the preseason Top 25 finish in the final Top 25.
The Big Sky Conference had 20 of 31 (65%) teams that started in the preseason Top 25 finish in the final Top 25.
The Big South Conference had 3 of 5 (60%) teams that started in the preseason Top 25 finish in the final Top 25.
The Colonial Athletic Association had 27 of 43 (63%) teams that started in the preseason Top 25 finish in the final Top 25.
The Great West Conference had 11 of 16 (69%) teams that started in the preseason Top 25 finish in the final Top 25.
The Ivy League had 2 of 4 (50%) teams that started in the preseason Top 25 finish in the final Top 25.
The Mid-Eastern Athletic Conference had 4 of 7 (57%) teams that started in the preseason Top 25 finish in the final Top 25.
The Missouri Valley Football Conference had 28 of 42 (67%) teams that started in the preseason Top 25 finish in the final Top 25.
The Northeast Conference had 0 of 1 (0%) teams that started in the preseason Top 25 finish in the final Top 25.
The Ohio Valley Conference had 5 of 10 (50%) teams that started in the preseason Top 25 finish in the final Top 25.
The Patriot League had 3 of 10 (30%) teams that started in the preseason Top 25 finish in the final Top 25.
The Pioneer Football League has never had a team ranked in the preseason Top 25.
The Southern Conference had 21 of 35 (60%) teams that started in the preseason Top 25 finish in the final Top 25.
The Southland Conference had 11 of 23 (48%) teams that started in the preseason Top 25 finish in the final Top 25.
The Southwestern Athletic Conference had 0 of 2 (0%) teams that started in the preseason Top 25 finish in the final Top 25.


And just an idea of how one consistent team can alter how well the pollsters do for a conference, if you take Montana's numbers out of the Big Sky, they fall to an 11-10 (52%) success rate. AppState for instance does a similar thing for the SoCon.

[clenz]

A10 and CAA are the same conference, FWIW.xcoffeex

Or, you need to break the MVFC and Gateway apart...

[superman7515]

A10 and CAA are the same conference, FWIW.xcoffeex

Or, you need to break the MVFC and Gateway apart...

Haha. I actually did that just for you, but it does show how the ECB (intentional or otherwise) is getting better as people become more familiar with the teams.

[ursus arctos horribilis]

The data shows that the poll is terrible. Sorry ursus

It's a prediction of the future based on little, if any, physical evidence so to think it's going to be extremely accurae would be a foolish venture.

The real beauty of this poll is not it's preseason (comparitively very good though) but the fact that teams who do very well early are recognized, and ranked earlier in this poll than in any other I've noticed.

The AGS Poll adjust very quickly to surprise teams like SHSU a couple years ago and Maine last year and a host of others.

[pike51]

It's funny to me that this thread, downplaying the validity of a pre-season poll, was started by a Chatty fan. Why? Because the FCS coaches poll is released and not a single SoCon team made the top 10. The happens when Chatty is supposed to be the top dog finally. Only thing that would have been better is if Chattown had started this thread.

On a side note, when was the last time a SoCon team wasn't ranked top 10 in the pre-season?

[PaladinFan]

It's funny to me that this thread, downplaying the validity of a pre-season poll, was started by a Chatty fan. Why? Because the FCS coaches poll is released and not a single SoCon team made the top 10. The happens when Chatty is supposed to be the top dog finally. Only thing that would have been better is if Chattown had started this thread.

On a side note, when was the last time a SoCon team wasn't ranked top 10 in the pre-season?

It's a cross section of perception against reality. The SoCon is not as dominant as it once was. I don't know whether the talent has been lacking or the other conferences have really closed the gap nationally. Even the top of the conference was not strong nationally. That is a good deal difference from the last 10 or 15 years where the SoCon was in the national title game virtually every season.

2014 is without question a transition year for the conference. I think with Furman and UTC in particular, there is a chance to start pushing the SoCon flag back into the national conversation. Both of those teams have tough schedules with a number of ranked opponents and tough OOC games. Lot of opportunity to show the polls that the SoCon isn't dead yet.

Preseaon polls are what they are. I do think that preseason polls are important to a degree. In a division where teams are selected for the post season, and where teams are usually punished less for losing than praised for winning, starting in a good position is a helpful leg up.

[walliver]

The "accuracy" of the ASG poll is due to a great extent, to the fact that it is not really a poll. Homer votes are discouraged and in more blatant cases, not counted. In many ways, it is a "moderated" poll, and somewhat insulated from some of the regional biases of other polls. On the other hand, much of its "accuracy" may reflect the degree to which the AGS poll mirrors the general thoughts of the AGS membership.

I probably should try to participate to help better represent the SoCon (I personally think Chatty and FU are ranked too low), but, to be honest, I don't have a good feel for most of FCS football outside of the SoCon and the Big South. When I tried voting previously (10 years ago +/-), I found it very difficult to fill out a complete schedule and gave up.

[superman7515]

On a side note, when was the last time a SoCon team wasn't ranked top 10 in the pre-season?

2004: #4 Furman, #6 Wofford, #9 Georgia Southern
2005: #2 Furman, #5 Georgia Southern
2006: #1 Appalachian State, #5 Furman
2007: #1 Appalachian State, #9 Furman
2008: #1 Appalachian State, #10 Wofford
2009: #2 Appalachian State, #10 Elon
2010: #3 Appalachian State, #7 Elon
2011: #2 Georgia Southern, #4 Appalachian State, #7 Wofford
2012: #3 Georgia Southern, #5 Appalachian State
2013: #8 Wofford

First time ever for the AGS poll.

[walliver]

2004: #4 Furman, #6 Wofford, #9 Georgia Southern
2005: #2 Furman, #5 Georgia Southern
2006: #1 Appalachian State, #5 Furman
2007: #1 Appalachian State, #9 Furman
2008: #1 Appalachian State, #10 Wofford
2009: #2 Appalachian State, #10 Elon
2010: #3 Appalachian State, #7 Elon
2011: #2 Georgia Southern, #4 Appalachian State, #7 Wofford
2012: #3 Georgia Southern, #5 Appalachian State
2013: #8 Wofford

First time ever for the AGS poll.

Not a single SoCon team put together a complete season last year. Furman started poorly, Wofford finished poorly, Chatty still couldn't win the games that really matter, and Samford looked beat down the stretch. It is hard to get Top 10 votes that way. Replacing ASU and GSU with VMI and Mercer won't help for this year either.

That said, I suspect there will be at least one Top 10 by the end of the year.

[OL FU]

Not a single SoCon team put together a complete season last year. Furman started poorly, Wofford finished poorly, Chatty still couldn't win the games that really matter, and Samford looked beat down the stretch. It is hard to get Top 10 votes that way. Replacing ASU and GSU with VMI and Mercer won't help for this year either.

That said, I suspect there will be at least one Top 10 by the end of the year.

Very good point. Most Preseason polls (some more than others) are reflective of the previous year with some adjustments. With that consideration, even if GSU and ASU were in this year, it might be tough to get them top ten votes also.

[kdinva]

I've never liked preseason polls........for football, let everyone play 4 or 5 games before voting. For basketball; play at least 9-10 games before the first poll.

[ursus arctos horribilis]

Very good point. Most Preseason polls (some more than others) are reflective of the previous year with some adjustments. With that consideration, even if GSU and ASU were in this year, it might be tough to get them top ten votes also.

I really can't see how either of them would have been top 10.

[PaladinFan]

I really can't see how either of them would have been top 10.

I think he was being kind. Neither GSU nor App State would be top 25 teams in any preseason poll.

[Milktruck74]

It's funny to me that this thread, downplaying the validity of a pre-season poll, was started by a Chatty fan. Why? Because the FCS coaches poll is released and not a single SoCon team made the top 10. The happens when Chatty is supposed to be the top dog finally. Only thing that would have been better is if Chattown had started this thread.

On a side note, when was the last time a SoCon team wasn't ranked top 10 in the pre-season?


Go back and re-read my OP...I was not "downplaying the validity of a pre-season poll." I was making a statement that the AGS voters are far more informed than the average pollster and felt it would be reflected in the percentage of top 25 teams that were still in the top 25 at the end of the season. I was actually think about the accuracy of the AGS poll in FCS vs some of the BCS polls out there.

And if you go back and look, I've never said my Mocs are there yet...they still have to prove it and I think they will soon......but will wait to brag until that time!!!!