Is the Fourpeat bad for I-AA?

[Ibanez]

Stupid facts.....

OK.

I was agreeing with your comments and the fact that Horny doesn't pay attention to details.

[89Hen]

OK.

I was agreeing with your comments and the fact that Horny doesn't pay attention to details.

Thought you were calling my facts stupid.

[Ibanez]

I was agreeing with your comments and the fact that Horny doesn't pay attention to details.

Thought you were calling my facts stupid.

Nope.

[Gil Dobie]

Meh. Computer rankings of I-AA teams compared to I-A teams have NEVER been accurate. And Massey...

Final Massey rankings:
4. South Dakota State
9. Indiana State
14. Western Illinois
15. Southern Illinois
20. Missouri State

Yet he claims computers don't self-implode on conferences.

SDSU might be the 4th best team in the country, losing to NDSU twice, ISUR, Missouri and YSU. They probably win the CAA or any other conference outside the MVFC except the Big Sky, EWU would have been tough to beat when healthy. With a healthy QB, SDSU may have won the YSU game.

The other schools listed would be in the upper half of any conference outside the MVFC. Massey did a good job with these teams.

You have not been following FCS very close this year if you don't see how good the Valley was. What was it 32-2 out of conference record.

[89Hen]

Meh. Computer rankings of I-AA teams compared to I-A teams have NEVER been accurate. And Massey...

Final Massey rankings:
4. South Dakota State
9. Indiana State
14. Western Illinois
15. Southern Illinois
20. Missouri State

Yet he claims computers don't self-implode on conferences.

SDSU might be the 4th best team in the country, losing to NDSU twice, ISUR, Missouri and YSU. They probably win the CAA or any other conference outside the MVFC except the Big Sky, EWU would have been tough to beat when healthy. With a healthy QB, SDSU may have won the YSU game.

The other schools listed would be in the upper half of any conference outside the MVFC. Massey did a good job with these teams.

You have not been following FCS very close this year if you don't see how good the Valley was. What was it 32-2 out of conference record.

We call that speculation Gil. The Valley was the best conference this year, but if you can say that a 6-6, 5-7 and 4-8 team are in the Top 20 with a straight face, you're either a really good actor, a complete homer, or just a fool.

[Gil Dobie]

SDSU might be the 4th best team in the country, losing to NDSU twice, ISUR, Missouri and YSU. They probably win the CAA or any other conference outside the MVFC except the Big Sky, EWU would have been tough to beat when healthy. With a healthy QB, SDSU may have won the YSU game.

The other schools listed would be in the upper half of any conference outside the MVFC. Massey did a good job with these teams.

You have not been following FCS very close this year if you don't see how good the Valley was. What was it 32-2 out of conference record.

We call that speculation Gil. The Valley was the best conference this year, but if you can say that a 6-6, 5-7 and 4-8 team are in the Top 20 with a straight face, you're either a really good actor, a complete homer, or just a fool.

The is speculation on how they would compare to the other teams in FCS. They should beat anyone below them in the rankings. The playoff committee uses GPI, see below. Very similar to Massey.

The GPI Top 25 - 11/25/2014
1. N Dakota St (1.43)
2. Illinois St (3.71)
3. Jacksonville St (4.00)
4. Northern Iowa (4.86)
5. New Hampshire (5.00)
6. E Washington (6.71)
7. S Dakota St (7.29)
8. Villanova (8.29)
9. Coastal Carolina (8.57)
10. Chattanooga (9.29)
11. Harvard (13.43)
12. Indiana St (13.86)
13. SE Louisiana (14.29)
14. Montana (15.86)
15. Youngstown St (17.00)
16. James Madison (17.43)
17. Fordham (17.86)
18. Liberty (18.29)
19. Sam Houston St (18.71)
20. S Illinois (19.29)
21. Idaho St (19.86)
22. Richmond (20.71)
23. W Illinois (21.57)
24. E Kentucky (23.29)
25. Montana St (24.86)
26. Missouri St (26.00)

GPI Link

[89Hen]

The is speculation on how they would compare to the other teams in FCS. They should beat anyone below them in the rankings. The playoff committee uses GPI, see below. Very similar to Massey.

Holy circular reasoning Batman. You want to use the GPI and Massey to defend Massey?

[Gil Dobie]

The is speculation on how they would compare to the other teams in FCS. They should beat anyone below them in the rankings. The playoff committee uses GPI, see below. Very similar to Massey.

Holy circular reasoning Batman. You want to use the GPI and Massey to defend Massey?

Well, head-to-head during the season and playoffs confirmed he MVFC rankings.

[89Hen]

Holy circular reasoning Batman. You want to use the GPI and Massey to defend Massey?

Well, head-to-head during the season and playoffs confirmed he MVFC rankings.

MVFC vs CAA head to head...

Illinois State 21 - UNH 18

If that doesn't tell us that South Dakota State would have won the CAA...

[Gil Dobie]

Overall, there are ups and downs, but FCS has done better since the CAA teams last won a championship.

Year - Attend - teams
2013 - 8,112, 124
2012 - 8,575, 122
2011 - 9,042, 121
2010 - 8,722, 117
2009 - 8,025, 118
2008 - 8,100, 118
2007 - 8,346, 116
2006 - 8,039, 116
2005 - 7,999, 116
2004 - 7,666, 119
2003 - 7,739, 121

Link

[Gil Dobie]

Well, head-to-head during the season and playoffs confirmed he MVFC rankings.

MVFC vs CAA head to head...

Illinois State 21 - UNH 18

If that doesn't tell us that South Dakota State would have won the CAA...

Pretty good for the 2nd place MFVC team to beat UNH at home. Nova had a tough time against a Liberty team that was blown out by Indiana St, almost as bad as SDSU blew out Indiana St. UNH also lost bad to Toledo, at team that Iowa St beat.

[89Hen]

MVFC vs CAA head to head...

Illinois State 21 - UNH 18

If that doesn't tell us that South Dakota State would have won the CAA...

Pretty good for the 2nd place MFVC team to beat UNH at home. Nova had a tough time against a Liberty team that was blown out by Indiana St, almost as bad as SDSU blew out Indiana St. UNH also lost bad to Toledo, at team that Iowa St beat.

I think we've elimitated acting. That leaves homer or fool. I'll give you the benefit of the doubt and go with homer.

Liberty > App State > Louisiana Monroe > Wake Forest > Virginia Tech > Ohio State

So really Indiana State > Ohio State

[Gil Dobie]

Pretty good for the 2nd place MFVC team to beat UNH at home. Nova had a tough time against a Liberty team that was blown out by Indiana St, almost as bad as SDSU blew out Indiana St. UNH also lost bad to Toledo, at team that Iowa St beat.

I think we've elimitated acting. That leaves homer or fool. I'll give you the benefit of the doubt and go with homer.

Liberty > App State > Louisiana Monroe > Wake Forest > Virginia Tech > Ohio State

So really Indiana State > Ohio State

You concession speech is accepted.

Have a great weekend.

[kalm]

Overall, there are ups and downs, but FCS has done better since the CAA teams last won a championship.

Year - Attend - teams
2013 - 8,112, 124
2012 - 8,575, 122
2011 - 9,042, 121
2010 - 8,722, 117
2009 - 8,025, 118
2008 - 8,100, 118
2007 - 8,346, 116
2006 - 8,039, 116
2005 - 7,999, 116
2004 - 7,666, 119
2003 - 7,739, 121

Link

In your face, 89!

[Grizo406]

I'd love for the Griz to have another 1 peat...

[Brock Landers]

I'd love for the Griz to have another 1 peat...

That's like my mantra every season

Dear, god, give me one more championship. Ever. And I'll die a happy man.

I use to say we're like Ohio State... now that went out the window.

[JohnStOnge]

Computer rankings of I-AA teams compared to I-A teams have NEVER been accurate.

On what basis do you say that? How do you assess accuracy?

You know how I'd do it. I'd look at a large number of games for which predictions are made, look at the percent correct, look at the bias (average of actual margin minus predicted margin), and mean squared error (or absolute error which is the average of the absolute values of the actual margins minus the predicted margins).

That's an objective way of doing it. Saying, "I don't think FCS team A is really better than FBS team B" is not. That's just an unsubstantiated opinion.

[89Hen]

Computer rankings of I-AA teams compared to I-A teams have NEVER been accurate.

On what basis do you say that? How do you assess accuracy?

You know how I'd do it. I'd look at a large number of games for which predictions are made, look at the percent correct, look at the bias (average of actual margin minus predicted margin), and mean squared error (or absolute error which is the average of the absolute values of the actual margins minus the predicted margins).

That's an objective way of doing it. Saying, "I don't think FCS team A is really better than FBS team B" is not. That's just an unsubstantiated opinion.

For going on 5 or 6 years I've asked you to participate in any of the pick-em pools where humans pick the games each week with you using the computer models. You've never stepped up to the plate. Couldn't be any more simple. You scared bro?

[89Hen]

Overall, there are ups and downs, but FCS has done better since the CAA teams last won a championship.

Year - Attend - teams
2013 - 8,112, 124
2012 - 8,575, 122
2011 - 9,042, 121
2010 - 8,722, 117
2009 - 8,025, 118
2008 - 8,100, 118
2007 - 8,346, 116
2006 - 8,039, 116
2005 - 7,999, 116
2004 - 7,666, 119
2003 - 7,739, 121

Link

In your face, 89!

Why, because Gil cherry picks years or you ignore the ones I posted? I-AA is not what it was thanks to some extremely well supported teams moving to I-A and others way down in support. 2012 to 2013 deltas for the important conferneces...

Southern -1,179 per game
Missouri Valley -342
Colonial -1,760
Big Sky -903
Southland -533

[Gil Dobie]

In your face, 89!

Why, because Gil cherry picks years or you ignore the ones I posted? I-AA is not what it was thanks to some extremely well supported teams moving to I-A and others way down in support. 2012 to 2013 deltas for the important conferneces...

Southern -1,179 per game
Missouri Valley -342
Colonial -1,760
Big Sky -903
Southland -533

So I cherry pick by showing all years.

Someone is cherry picking by selecting 2012 and 2013. GSU and App St attendance dropped by 2-4000 in 2013, during a transition from FCS to FBS. Basically their 2013 season was for pride and attendance dropped and ODU was not included in 2013 because of timing of their move to FBS. During that same period NDSU attendance went up by over 100 fans per game and they dropped from #6 in 2012 to #7 in 2013. Comparing average conference attendance is difficult, because it changes due to membership and who the best teams are at the time. Delaware is down because they are not a playoff team as are several other school. When and if those schools start hitting the playoffs, attendance will be up. The does not mean FCS is up. I'm curious to see the 2014 stats.

[JohnStOnge]

For going on 5 or 6 years I've asked you to participate in any of the pick-em pools where humans pick the games each week with you using the computer models. You've never stepped up to the plate. Couldn't be any more simple. You scared bro?

I don't recall you ever asking me that. I'm not saying you never did. Could be I didn't see those particular posts. I don't mind doing that but anybody can. I don't have to participate. All you have to do is look at who was picked by, say, the Sagarin system. Look both at the percentage of time that winners were picked correctly as well as the other measures of accuracy (bias and absolute error or mean squared error).

One thing, though, is that to get a good measure of accuracy you can't just pick the games people want to look at. You either have to pick ALL the games or a "scientific sample" of them. You can't say one thing or the other is more accurate in picking games unless you have that.

That's the problem with a lot of pick 'em contest selections. They don't involve a "scientific sample" of games. Doing something like picking the games involving top 25 teams doesn't do it.

[89Hen]

For going on 5 or 6 years I've asked you to participate in any of the pick-em pools where humans pick the games each week with you using the computer models. You've never stepped up to the plate. Couldn't be any more simple. You scared bro?

I don't recall you ever asking me that. I'm not saying you never did. Could be I didn't see those particular posts. I don't mind doing that but anybody can. I don't have to participate.

You saw them.

[89Hen]

So I cherry pick by showing all years.

Someone is cherry picking by selecting 2012 and 2013.

Those were the last two years available (not sure why 2014 isn't out, or so hard to find).

[JohnStOnge]

You saw them.

Well I sure don't recall seeing them. I've written a number of times about doing something where somebody picks against the system.

Another thing about it though. What you need to do for a test like that is pick one power rating system to test and one person to test. Pick a person you think ahead of time is good and compare the performance of that one person to that one power rating system.

If you compare one power rating system to a bunch of people you have the problem of increasing the likelihood that one or a few people just get lucky by chance and beat the system. So if you were to say you want to compare the effectiveness of something like the Sagarin system to the performance of someone who is good at picking games you would need to pick the one person in your pool or whatever who you consider to be the best at picking games ahead of time then that would be the test. No throwing multiple people in then picking the one who did best for the comparison after the fact.

[89Hen]

You saw them.

Well I sure don't recall seeing them. I've written a number of times about doing something where somebody picks against the system.

Another thing about it though. What you need to do for a test like that is pick one power rating system to test and one person to test. Pick a person you think ahead of time is good and compare the performance of that one person to that one power rating system.

If you compare one power rating system to a bunch of people you have the problem of increasing the likelihood that one or a few people just get lucky by chance and beat the system. So if you were to say you want to compare the effectiveness of something like the Sagarin system to the performance of someone who is good at picking games you would need to pick the one person in your pool or whatever who you consider to be the best at picking games ahead of time then that would be the test. No throwing multiple people in then picking the one who did best for the comparison after the fact.

I've asked you many times to participate in one of the pick threads/games and you can pick your rating system and play with the rest of the field. I would think your computer should be able to beat anyone over the course of a year. BTW, not hard to find...

One of these years I'm going to remember to start at the beginning and have you pick randomly selected games from among all the games and see how you do against one or more of the better models in predicting outcomes. Maybe I'll pick Massey. And you will almost certainly lose the contest.

Bullshit. I asked you 5 straight years to do this and you never take me up on it. Who gives a shit if the computer does better on Valpo vs Campbell games? For years, the gohens pool has completely out performed the compter models.

From 2013. So this back and forth has been going on a long time. You choose to not play the game.

http://www.championshipsubdivision.com/ ... ck#p824628