Keeper's Top 30 Power Ratings for 2006

[Keeper]

KCFR pre-season power ratings
(beginning of season rating only, not representing predicted outcome)

1 ILLINOIS ST.......... 67.75
2 MASSACHUSETTS........ 66.91
3 APPALACHIAN ST....... 65.92
4 FURMAN............... 62.61
5 RICHMOND............. 62.41
6 EASTERN KENTUCKY..... 62.11
7 NEW HAMPSHIRE........ 60.84
8 JAMES MADISON........ 60.67
9 UC-DAVIS............. 60.59
10 CAL POLY............. 60.36
11 EASTERN ILLINOIS..... 59.73
12 NORTH DAKOTA ST...... 58.38
13 HOFSTRA.............. 58.18
14 MCNEESE ST........... 57.26
15 GEORGIA SOUTHERN..... 56.04
16 TOWSON............... 55.38
17 MONTANA.............. 55.21
18 MONTANA ST........... 55.13
19 WESTERN KENTUCKY..... 55.06
20 YOUNGSTOWN ST........ 54.83
21 NICHOLLS ST.......... 54.80
22 HAMPTON.............. 54.33
23 TEXAS ST............. 54.08
24 COASTAL CAROLINA..... 53.89
25 ALABAMA A&M.......... 53.50
26 PORTLAND ST.......... 53.46
27 SOUTH CAROLINA ST.... 52.08
28 SAMFORD.............. 52.01
29 IDAHO ST............. 51.72
30 WESTERN ILLINOIS..... 51.68

Any bets where comes the first response?
Good luck to your team. More later.

[SunCoastBlueHen]

Finally, a ranking that makes sense! : retard :

[Tribe4SF]

KCFR pre-season power ratings
(beginning of season rating only, not representing predicted outcome)

1 ILLINOIS ST.......... 67.75
2 MASSACHUSETTS........ 66.91
3 APPALACHIAN ST....... 65.92
4 FURMAN............... 62.61
5 RICHMOND............. 62.41
6 EASTERN KENTUCKY..... 62.11
7 NEW HAMPSHIRE........ 60.84
8 JAMES MADISON........ 60.67
9 UC-DAVIS............. 60.59
10 CAL POLY............. 60.36
11 EASTERN ILLINOIS..... 59.73
12 NORTH DAKOTA ST...... 58.38
13 HOFSTRA.............. 58.18
14 MCNEESE ST........... 57.26
15 GEORGIA SOUTHERN..... 56.04
16 TOWSON............... 55.38
17 MONTANA.............. 55.21
18 MONTANA ST........... 55.13
19 WESTERN KENTUCKY..... 55.06
20 YOUNGSTOWN ST........ 54.83
21 NICHOLLS ST.......... 54.80
22 HAMPTON.............. 54.33
23 TEXAS ST............. 54.08
24 COASTAL CAROLINA..... 53.89
25 ALABAMA A&M.......... 53.50
26 PORTLAND ST.......... 53.46
27 SOUTH CAROLINA ST.... 52.08
28 SAMFORD.............. 52.01
29 IDAHO ST............. 51.72
30 WESTERN ILLINOIS..... 51.68

Any bets where comes the first response?
Good luck to your team. More later.

You don't have a brother named Pete, do you?:eek:

[bluehenbillk]

WTF?

[Cocky]

You don't have a brother named Pete, do you?:eek:

Pete's poll makes more sense than this one.

[TexasTerror]

UMass at 2nd? Towson at 16, ahead of the Montana institutions?

Where's Pete?!?

[HENJOHN]

Looks like a Poll done with a dart board to me.......

[cosmo here]

would you like to explain your methodology? it looks like a pretty complex formula went into producing your results, those numbers are meaningless to everybody on this board but you unless you give some background.

[FlyYtown]

YSU at 20?????
C-MON....

[walliver]

How can you calculate a power poll when

Every I-AA team is tied for best record at 0-0 (That right, ASU and Savannah State are tied for best record)
Nobody has scored a point
Nobody has given up a point
Nobody has gained a yard on offense
Nodody has given up a yard on defense
Nobody has had a home game?

[OL FU]

How can you calculate a power poll when

Every I-AA team is tied for best record at 0-0 (That right, ASU and Savannah State are tied for best record)
Nobody has scored a point
Nobody has given up a point
Nobody has gained a yard on offense
Nodody has given up a yard on defense
Nobody has had a home game?

I agree. May actually take balls to bring out a power poll pre-season.

[Fordham]

Relax fellas. This guy took the time to put in numbers down to the hundredths!! It certainly therefore looks legit to me, although not as much as if he had gone down to the thousandths or beyond.

Although I have to say that the personal poll that I just did in my head had Montana at 55.31, so unless that's a typo in his poll this thing must be bs.

[bluehenbillk]

Further proof that computer ratings have NO place in college football.

[Pauly LB]

Laughing out loud...

[OL FU]

Relax fellas. This guy took the time to put in numbers down to the hundredths!! It certainly therefore looks legit to me, although not as much as if he had gone down to the thousandths or beyond.

Although I have to say that the personal poll that I just did in my head had Montana at 55.31, so unless that's a typo in his poll this thing must be bs.

I won't be happy until Furman is a solid 69:nod:

[Black Saturday]

KCFR pre-season power ratings
(beginning of season rating only, not representing predicted outcome)

1 ILLINOIS ST.......... 67.75
2 MASSACHUSETTS........ 66.91
3 APPALACHIAN ST....... 65.92
4 FURMAN............... 62.61
5 RICHMOND............. 62.41
6 EASTERN KENTUCKY..... 62.11
7 NEW HAMPSHIRE........ 60.84
8 JAMES MADISON........ 60.67
9 UC-DAVIS............. 60.59
10 CAL POLY............. 60.36
11 EASTERN ILLINOIS..... 59.73
12 NORTH DAKOTA ST...... 58.38
13 HOFSTRA.............. 58.18
14 MCNEESE ST........... 57.26
15 GEORGIA SOUTHERN..... 56.04
16 TOWSON............... 55.38
17 MONTANA.............. 55.21
18 MONTANA ST........... 55.13
19 WESTERN KENTUCKY..... 55.06
20 YOUNGSTOWN ST........ 54.83
21 NICHOLLS ST.......... 54.80
22 HAMPTON.............. 54.33
23 TEXAS ST............. 54.08
24 COASTAL CAROLINA..... 53.89
25 ALABAMA A&M.......... 53.50
26 PORTLAND ST.......... 53.46
27 SOUTH CAROLINA ST.... 52.08
28 SAMFORD.............. 52.01
29 IDAHO ST............. 51.72
30 WESTERN ILLINOIS..... 51.68

Any bets where comes the first response?
Good luck to your team. More later.

Where'd UMASS come from? I haven't seen anyone pick them to win their division or even the A-10. James Madison should be a little higher. Montana should probably should be about 10+ spots higher kinda stand out to me. But, what do I know? Nice to see your poll.:smiley_wi

[lucchesicourt]

The GWFC won't complain- 3of 5 in the top 12.

[birdsflyhigh]

Great to see the ISU Redbirds on top of a poll, but.....man, this poll is strange to say the least. Two of the better teams in the Gateway (UNI and SIU) don't rate, while WIU comes in at #30?

Hey, I enjoyed seeing the Birds on top, but would have to agree with some of the other posters here about the validity of the poll....hmmmm???

[Maroons]

My interest is piqued... what the hell has he done?

[89Hen]

How can you calculate a power poll when
Pete always had an intriguing explanation. Part of his rankings were that at the start of the season no conference could have more than one in the top 10, no matter how good two teams were. He said that it worked itself out, but there is no way it can. If a team is inflated just because their conference has to be ranked in the top 10, every team that plays them will have an inflated ranking going forward. The problems can lessen over the course of the season, but with only 11 game seasons, there is no way you can eliminate it. Computer models cannot avoid this problem.

[Killtoppers90]

I liked the results of this poll. Not certain how he came up with the numbers but it works in WKU's favor, so I like it.

[McNeese75]

WTF?


:lmao: :lmao: :lmao: :lmao: :lmao: :lmao: :lmao:

[Paladin1aa]

I get it.

ITS A JOKE !!

[smallcollegefbfan]

I have to agree with everyone else. I would love to know what he is smoking. It should be ban if has not been already... :)

[GannonFan]

Any truth to the rumor that the GPI is looking to incorporate this power ranking into its methodology??? ;)

[Spider]

I agree this poll doesn't look to accurate but Keepers does a good job of predicting the weekly point spreads and projected winners during the season.

[Tribe4SF]

KCFR pre-season power ratings
(beginning of season rating only, not representing predicted outcome)

1 ILLINOIS ST.......... 67.75
2 MASSACHUSETTS........ 66.91
3 APPALACHIAN ST....... 65.92
4 FURMAN............... 62.61
5 RICHMOND............. 62.41
6 EASTERN KENTUCKY..... 62.11
7 NEW HAMPSHIRE........ 60.84
8 JAMES MADISON........ 60.67
9 UC-DAVIS............. 60.59
10 CAL POLY............. 60.36
11 EASTERN ILLINOIS..... 59.73
12 NORTH DAKOTA ST...... 58.38
13 HOFSTRA.............. 58.18
14 MCNEESE ST........... 57.26
15 GEORGIA SOUTHERN..... 56.04
16 TOWSON............... 55.38
17 MONTANA.............. 55.21
18 MONTANA ST........... 55.13
19 WESTERN KENTUCKY..... 55.06
20 YOUNGSTOWN ST........ 54.83
21 NICHOLLS ST.......... 54.80
22 HAMPTON.............. 54.33
23 TEXAS ST............. 54.08
24 COASTAL CAROLINA..... 53.89
25 ALABAMA A&M.......... 53.50
26 PORTLAND ST.......... 53.46
27 SOUTH CAROLINA ST.... 52.08
28 SAMFORD.............. 52.01
29 IDAHO ST............. 51.72
30 WESTERN ILLINOIS..... 51.68

Any bets where comes the first response?
Good luck to your team. More later.

Tribe plays at Towson this year, and I'll take that bet!:cool:

[GeauxColonels]

:confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused:

: retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard : : retard :

:doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh: :doh:

:dizzy:

[RabidRabbit]

The GWFC won't complain- 3of 5 in the top 12.

:bow: Yeah, But we're about there with the AGS preseason :bow:

[LeopardFan04]

Relax fellas. This guy took the time to put in numbers down to the hundredths!! It certainly therefore looks legit to me, although not as much as if he had gone down to the thousandths or beyond.

Although I have to say that the personal poll that I just did in my head had Montana at 55.31, so unless that's a typo in his poll this thing must be bs.


xlolx :rotateh: :rotateh: xlolx

[eaglesrthe1]

Any truth to the rumor that the GPI is looking to incorporate this power ranking into its methodology??? ;)


xlolx xlolx :nod: :nod:

Bet your bottom dollar. If by any fluke it turns out to be anywhere close to reality...even by sheer chance....:nod: :nod: Look out for 2007:rolleyes:


I would like to see the method. How about it Keeper?

[McNeese75]

I would like to see the method. How about it Keeper?


http://www.****throwingmonkeys.com/stm-com.jpg

[OL FU]

I agree this poll doesn't look to accurate but Keepers does a good job of predicting the weekly point spreads and projected winners during the season.

AGS Ranking to the right

1 ILLINOIS ST.......... 9
2 MASSACHUSETTS 14
3 APPALACHIAN ST 1
4 FURMAN............... 5
5 RICHMOND...........25
6 EASTERN KENTUCKY..24
7 NEW HAMPSHIRE........ 2
8 JAMES MADISON........8
9 UC-DAVIS............. 20
10 CAL POLY...........6
11 EASTERN ILLINOIS.. 15
12 NORTH DAKOTA ST. 12
13 HOFSTRA.............. 37
14 MCNEESE ST..........11
15 GEORGIA SOUTHERN...13
16 TOWSON............... 49
17 MONTANA............4
18 MONTANA ST......19
19 WESTERN KENTUCKY..... 27
20 YOUNGSTOWN ST........ 7
21 NICHOLLS ST.......... 22
22 HAMPTON.............. 16
23 TEXAS ST............. 21
24 COASTAL CAROLINA..... 18
25 ALABAMA A&M..........
26 PORTLAND ST..........
27 SOUTH CAROLINA ST....32
28 SAMFORD..............
29 IDAHO ST............. 31
30 WESTERN ILLINOIS.....


UNI 3
Delaware 10
SIU 17
EWU 23

I'll be the contrarian. There are some obvious problems, but taken as a whole and considering how difficult it is to do a pre-seaons poll. I don't really think it looks that bad. :cool:

I did not put IlSU #1, but it is not without possibility.
Richmond and EKU look very high. But once you get outside of the top 5 we are fairly close.

I still think Furman should not be in the top 5:eyebrow:

[Maverick]

http://home.att.net/~keepers.football/wsb/html/view.cgi-html2.html--SiteID-797229.html

Here is the website that Keeper explains what he is doing. We'll see where it goes from here.

[griz&beer]

LOL.

[Tod]

LOL.

Dude, change your sig. :bawling: :bawling: :bawling: :bawling: :bawling: :bawling: :bawling: :bawling: :bawling:

[Tod]

Personally, I think Keeper should just go for it. He is thinking and doing his best to come up with a composite way to rank teams. I will not blame him for that. In fact, I congatulate him for it!

Keep up the good work, Keeper. Maybe you'll end up better than other predictions and you can do this: :p :p :p :p :p

[Frosty The Snowbuff]

Frostfucious has gone Confusious :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused: :confused:


With that said....is he's using last years stats to come up with these computations??

[Maverick]

Everybody will have a chance to "critique" his methods once the regular season ends. Then his playoff predictions would be next on the menu!

One thing does seem sure about him is that he is not afraid of criticisms or nay sayers, but after doing this kind of thing for as long as he has I'm sure he will have heard most of the arguments he will encounter.

I applaud anybody willing to take the time to do this and then be public about it, it takes more courage than a lot of those who will firing away at him will exhibit. Everybody has an opinion which is in reality usually an uproven solution to the stated "problem".

Still it will stimulate "discussion".

[OL FU]

Everybody will have a chance to "critique" his methods once the regular season ends. Then his playoff predictions would be next on the menu!

One thing does seem sure about him is that he is not afraid of criticisms or nay sayers, but after doing this kind of thing for as long as he has I'm sure he will have heard most of the arguments he will encounter.

I applaud anybody willing to take the time to do this and then be public about it, it takes more courage than a lot of those who will firing away at him will exhibit. Everybody has an opinion which is in reality usually an uproven solution to the stated "problem".

Still it will stimulate "discussion".

I like his pre-season ranking better than Sagarin:smiley_wi

I agree. I am not sure how computer rankings work at the start of the season ( Hell I am not sure how they work during the season). But if the purpose is to negate subjectivity then about the only thing the pre-season poll can take into account is returning players and historical statistics.


Computer rankings (Ralph, you have taught me well) mean very little until a few weeks into the season

[madcow]

I agree it doesnt seem right, i like it though, having GWFC teams at 9,10, and 12, which is way more respect than other polls has indicated....GO UC DAVIS... #9 in this poll.

[Maverick]

As far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality.

Albert Einstein
Sidelights on Relativity


Maybe this is what sums it up for many people regarding stastical analysis. As for me, why do you think they play the games? The human factor is not a constant, but more likely the source of an anomaly. Possibility vs. Probability does not cover the random margin of error that humanity can interject. What is it the old ABC Wide World of Sports opening said, "The human drama of athletic competition." In this drama, some people forget to read the script and the improvised results is never forgotten by some members of the audience.

[Ronbo]

We're too high.

[*****]

I would like to see the method. How about it Keeper?Keeper is not new or a secret. Check out the website:
http://home.att.net/~keepers.football/wsb/html/view.cgi-home.html-.html

[crunifan]

Western Illinois at #30. xlolx

[ngineer]

xlolx xlolx xlolx xlolx xlolx xlolx xidiotx xidiotx xidiotx xidiotx :eyebrow: :eyebrow: :eyebrow: :eyebrow: :confused: :confused: :confused: :confused: :confused: :confused: :confused:

[blukeys]

[QUOTE=Maverick]Everybody will have a chance to "critique" his methods once the regular season ends. Then his playoff predictions would be next on the menu!

[QUOTE]


I agree that he is to be commended for his honesty in posting his preseason rankings upfront. I have gone to the website and have seen what he prioritizes but I don't see how his formuma brings this about. In short their is no detail on the website about how a score of 55.14 is arrived at for any team.

I also question some of his key indicators of quality for a team. For instance he states that the total yardage difference between the two teams is a key factor in his rankings (he does not indicate how much). Most Professional defensive coaches consider Turnover margin a much better indication of defensive efficiency then yardage given up. As pointed out here on numerous occasions a team can give up tons of yardage when they are ahead significantly and/or their backup defense is in the game. In addition a team can force 3 successive turnovers and have a 21-0 lead and only have 100yds total offense. (These were the stats of UD v SIU in the 2003 playoff game)

The website does not indicate if any return yardage either for kickoffs or punt returns are considered. I did not see that they were and this basically penalizes teams that have very good special teams.

I did not get enough information from the website to analyze the system. I saw a broad outline and that was it.

The lack of detail shows why so many here are critical. If this is a mathematical formula, why not post the details????????

I have no problem with any rating honestly arrived at but at least share the details of the formula. (Something Sagarin and Dunkel always did)

[Keeper]

Wow! Great conversation guys. Passion and Thoughtfulness abound.

Yes, at this point it is TRULY meaningless, and of course never
completely accurate pending the final outcome, but what is?
Rating college football is a very enjoyable and time-consuming hobby,
but I have a full-time paying job and this is not it. If it was, then I
would be able to incorporate all variables and statistical aspects into
the spreadsheets relative to college football. Such as it is, this is my
attempt to rate teams more accurately beyond the use of scores,
winning pct, and number of returning starters.

I agree that turnovers are significant to the outcome of most games
and often reverse the results of dominating yardage. However, the
statistical use of say '+1.5 turnover margin per game' has little
predictive quality. It is only after the fact, or retrodictive, that
turnovers have a bearing on a team's rating. Likewise, special teams
stats are not included (however fair or unfair) because of their game
frequency and use of non-starters. I agree that a potentially fair
and accurate formula would include ALL variables. What about
coaching? Weather? Turf? Scholarship numbers and recruits?

My system attempts to rate game performance using yards gained at
the line of scrimmage, the true POWER aspect of football. I have
found this to be a good predictive source, with many upsets, rather
than the use of final scores. The ability of one team to run over or
pass through another can be consistently measured and confidently
displayed.

I prefer not to divulge the exact formula of my system, but readily
admit it is not advanced mathematics (like Sagarin) but is simple
algebra. For rating a game performance, a team's Net Yards Percentage
(usually between 65% to 35%) is converted into a point value ranging
from 50.00 for best to 0.00 for worst, the typical difference between
best and worst in conference, as well as being close to average of 50
points scored per game. The NYP value is added to the previous
rating of the team's opponent, and becomes the Power Performance
Rating for that game.

Example: If Harvard outgains Yale 500 yards to 300, Harvard has 62.5%
of the game's net yards (very dominant). This converts to a yardage
rating of 45.83 pts. Then this is added to Yale's pregame power rating
(say 35.60) to arrive at Harvard's Power Performance Rating of 81.43.
So basically, a team is rated by its performance vs the relative strength
of its opponent. A 10 point win over another powerhouse obviously
would be rated higher than a blowout over a bad team.

During the season, separate ratings are kept for each individual game
using the current existing rating for each opponent played. Each most
recent game rating in the schedule is given more consideration in the
formula to better measure up-to-date progress or lack-of. Early
season game ratings gradually replace the pre-season rating, which
disappears after the teams fifth game played.

Yes, the Pre-Season Ratings! They are based on last season's final
stats, and are adjusted according to the yardage performance lost
relative to graduated players, or anyone suspended, injured, etc.
Right down to a single yard rushed for or caught. Adjustments are also
made for offensive line losses based on number of games played or
started, and defensive performance based on number of tackles made.
Of course, these are calculated the same across the board for all teams
regardless of their program size or ability to reload, and are based on
years of research and study of those trends.

My ratings are without bias and honestly just a big numbers crunch.
I am frequently dissapointed with some of the rankings, but I refuse
to change or alter any of the final output, especially predictions.
No opinions are involved nor the use of any scouting or insider info.

At the very least, I am hopeful that these ratings provide entertainment
for many college football fans who enjoy the sport as much as I do.
There are millions of fantasy football players, but only about a hundred
computer ratings providers. I would be pleased to be considered as
being among the best of them. I have enjoyed all of your responses
and happy to be a member of AGS. Hopefully, this little blog will
answer many of your burning questions. Best Regards, Keeper.