Replies: 68
| visibility 1
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
OK, here ya go. 2016 Final Final Prodigal Football Ratings
Jan 11, 2017, 8:55 AM
|
|
It's not the first time it's happened, and I'm not enthused about it, but the national champion Clemson Tigers did not finish in first place in this formula. Look at the numbers below and you'll see why. The records are the same, and the schedule numbers are a little in favor of Clemson. But the scoring margin is the difference. And not just the scoring margin itself, though that is the biggest factor...Alabama's schedule strength numbers when based on scoring margin are also high. In "W-L-only" mode, Clemson is #1, but in "scoring margin-only" mode, they are #4 (behind Bama, OSU, and Michigan). Clemson's SOS based on W-L is #1, but the SOS based on scoring margin has Alabama #1 and Clemson #5.
Overall, it is very close between Alabama and Clemson, such that if Clemson had won by about 18 instead of 4, they would be #1. There is a sizeable jump between Clemson and Ohio St., and a relatively enormous chasm between Ohio St. and #4 Washington. So that's pretty interesting.
Sorry about that folks, but I just report the facts. I ain't gonna change it just because it spit out something I don't like. (If it consistently did something in particular wrong, I would try to change it, and have done so.) Thanks for indulging me...I enjoy doing these ratings very much, even if I get 90% criticism here. I just think it's fun.
Rank | Pvs | Team | W-L | PDiff | SOS |
---|
1 | 1 | Alabama | 14-1 | 2 | 4 | 2 | 3 | Clemson | 14-1 | 6 | 1 | 3 | 2 | Ohio St. | 11-2 | 4 | 2 | 4 | 6 | Washington | 12-2 | 3 | 43 | 5 | 8 | Oklahoma | 11-2 | 10 | 19 | 6 | 7 | Wisconsin | 11-3 | 14 | 6 | 7 | 10 | Florida St. | 10-3 | 30 | 3 | 8 | 4 | Michigan | 10-3 | 1 | 28 | 9 | 11 | USC | 10-3 | 29 | 7 | 10 | 5 | Penn St. | 11-3 | 21 | 14 | 11 | 9 | Western Michigan | 13-1 | 5 | 113 | 12 | 15 | Florida | 9-4 | 42 | 5 | 13 | 13 | Louisville | 9-4 | 8 | 16 | 14 | 16 | Stanford | 10-3 | 45 | 22 | 15 | 12 | Boise St. | 10-3 | 27 | 47 | 16 | 14 | Colorado | 10-4 | 35 | 20 | 17 | 26 | Appalachian St. | 10-3 | 25 | 56 | 18 | 20 | LSU | 8-4 | 16 | 15 | 19 | 27 | Miami | 9-4 | 9 | 32 | 20 | 21 | Western Kentucky | 11-3 | 7 | 95 | 21 | 19 | Tennessee | 9-4 | 38 | 17 | 22 | 32 | Oklahoma St. | 10-3 | 22 | 71 | 23 | 17 | Auburn | 8-5 | 12 | 10 | 24 | 18 | South Florida | 11-2 | 20 | 105 | 25 | 30 | Virginia Tech | 10-4 | 19 | 49 |
Dropped out: #22 Houston, #23 Nebraska, #24 Temple, #25 West Virginia
ACC/SEC teams:
1. Alabama 2. Clemson 7. Florida St. 12. Florida 13. Louisville 18. LSU 19. Miami 21. Tennessee 23. Auburn 25. Virginia Tech 33. Georgia Tech 34. Pittsburgh 37. Texas A&M 47. NC State 49. North Carolina 51. Georgia 54. Arkansas 55. Kentucky 57. Vanderbilt 60. Ole Miss 61. Wake Forest 62. Mississippi St. 64. Boston College 71. South Carolina 84. Syracuse 85. Missouri 88. Duke 116. Virginia
|
|
|
|
All-TigerNet [11860]
TigerPulse: 100%
Posts: 8584
Joined: 2/2/16
|
Thank you for not disappointing us***
Jan 11, 2017, 8:56 AM
|
|
|
|
|
|
|
All-In [40657]
TigerPulse: 100%
Posts: 23591
Joined: 1/29/05
|
I'll be the first to go ahead and say it
Jan 11, 2017, 8:57 AM
|
|
|
|
|
|
|
All-In [40657]
TigerPulse: 100%
Posts: 23591
Joined: 1/29/05
|
easy fix to your formula
Jan 11, 2017, 9:06 AM
|
|
=IF(TeamCWeek18>TeamAWeek18,"1","2")
|
|
|
|
|
Orange Blooded [2053]
TigerPulse: 93%
Posts: 1493
Joined: 10/27/16
|
Re: OK, here ya go. 2016 Final Final Prodigal Football Ratings
Jan 11, 2017, 8:58 AM
|
|
|
|
|
|
|
Rock Defender [54]
TigerPulse: 90%
Posts: 35
Joined: 11/30/98
|
Serious question is serious
Jan 11, 2017, 9:01 AM
|
|
Why even take the time to do this poll? It means less than the coaches poll and is a terrible indicator of how good a team actually is.
|
|
|
|
|
Oculus Spirit [81246]
TigerPulse: 100%
Posts: 56281
Joined: 9/13/04
|
This is actually a good question.
Jan 11, 2017, 9:58 AM
|
|
What Prod's poll has reported is that even though the mathematically two top teams played each other, the winner still isn't #1.
So if this is true, there would be no reason to even play the game, as the top mathematical team could not lose, regardless of the outcome.
His poll states that Clemson's win meant nothing, and invalidates the actual, real result of being National Champion.
Somewhere, the logic isn't following.
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
You're taking this way too seriously.
Jan 11, 2017, 10:00 AM
|
|
This rating means nothing to no one, ever. Has nothing to do with who should be national champions. It's something I do for fun, that I share with y'all.
|
|
|
|
|
Oculus Spirit [81246]
TigerPulse: 100%
Posts: 56281
Joined: 9/13/04
|
I understand all this.
Jan 11, 2017, 10:05 AM
|
|
I get it, its for fun, you do it because you want to etc.
The point still stands that something in your weighted criteria is incorrect; else you are attempting to quantify something that doesn't relate to real world outcomes. Is that your attempt?
If so, then the question of why you do this is very valid.
I mean if I postulate 1+1=3, and can post a convoluted proof to it that mathematically works, does it really matter? I'm still incorrect in stating the answer is 3.
|
|
|
|
|
CU Medallion [66276]
TigerPulse: 100%
Posts: 24781
Joined: 6/12/14
|
OF COURSE BAMA HAD THE TOUGHER SCHEDULE
Jan 11, 2017, 9:03 AM
|
|
THEY PLAYED THE NATIONAL CHAMPIONS AND WE DID NOT
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
Overall, my ratings give Clemson the #1 schedule.***
Jan 11, 2017, 9:04 AM
|
|
|
|
|
|
|
Oculus Spirit [78876]
TigerPulse: 100%
Posts: 78623
Joined: 8/2/03
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
To put it simply...
Jan 11, 2017, 9:04 AM
|
|
Alabama finishes ahead here because they were more dominant scoring-wise than Clemson, and the SEC teams as a whole were generally more dominant scoring-wise in OOC games, particulary the SEC West.
Big stat that likely influences this 1/2 rating in a big way: SEC West teams had a +16.55 per game scoring margin against non-division foes, while ACC Atlantic teams had a +9.96 (which was still the 2nd best of any division.) This is why Bama's schedule numbers are as high as they are.
|
|
|
|
|
Oculus Spirit [81994]
TigerPulse: 100%
Posts: 47128
Joined: 3/18/07
|
So you're compounding scoring margin in both
Jan 11, 2017, 9:12 AM
|
|
individual team formulas and the conference formulas for SOS? Or just providing two different things here?
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
The individual team ratings and the conference
Jan 11, 2017, 9:14 AM
|
|
ratings are done independent of each other. (Though the same formula is used.) The individual ratings have no idea what conference the teams are.
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
To be clear, what I was saying was that...
Jan 11, 2017, 9:26 AM
|
|
The fact that SEC-W teams dominated in that fashion in non-division games, is evidence that shows why Alabama's numbers were so high, since those teams comprised a major portion of their schedule.
|
|
|
|
|
Oculus Spirit [81994]
TigerPulse: 100%
Posts: 47128
Joined: 3/18/07
|
I gotcha, I presumed it was independent and not
Jan 11, 2017, 9:28 AM
|
|
compounded, but wanted to double check because the wording of that post slightly confused me.
For the record, I appreciate your poll you do.
|
|
|
|
|
CU Medallion [55848]
TigerPulse: 100%
Posts: 35354
Joined: 11/30/98
|
ima print this out...
Jan 11, 2017, 9:06 AM
|
|
and wipe my ### with it!
|
|
|
|
|
All-In [38514]
TigerPulse: 100%
Posts: 47162
Joined: 10/28/02
|
I believe you put too much weight on scoring margin.
Jan 11, 2017, 9:07 AM
|
|
Too often it inflates the value of blowing out a mediocre team over a close win over a great team. The fact that you have Clemson as having a stronger SOS. Just appears that your favoring MOV over SOS.
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
The SOS is a bigger factor overall than scoring margin is.
Jan 11, 2017, 9:13 AM
|
|
SOS is about 31% of the rating, scoring margin is 25%. (Those percentages do overlap)
However, Alabama is farther "ahead" in scoring margin than Clemson is in SOS, so that created the small margin they had over Clemson in the overall rating.
I have thought about tamping down scoring margin, but I already did that once a couple years ago. I don't want to change because of one result I don't like...I'd want to change if I consistently though worse teams were being ranked higher because of blowing out lesser teams. I don't know if that's the case.
|
|
|
|
|
CU Medallion [66276]
TigerPulse: 100%
Posts: 24781
Joined: 6/12/14
|
THAT'S FUNNY BECAUSE CLEMSON HAS THE EDGE IN
Jan 11, 2017, 9:15 AM
|
|
SCORING MARGIN 35-31......I MEAN MATHING AIN'T THAT HARD
|
|
|
|
|
All-In [49198]
TigerPulse: 100%
Posts: 38848
Joined: 12/31/97
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
No. At least, I don't know how.***
Jan 11, 2017, 10:02 AM
|
|
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
If I could figure out how, that would be a dream come true.
Jan 11, 2017, 10:06 AM
|
|
But I don't know of any mathematical rating, and obviously many people a million times smarter than me do them, that have a head-to-head component.
|
|
|
|
|
Heisman Winner [138129]
TigerPulse: 100%
Posts: 63868
Joined: 10/22/00
|
|
|
|
|
All-In [38514]
TigerPulse: 100%
Posts: 47162
Joined: 10/28/02
|
Exactly.***
Jan 11, 2017, 9:14 AM
|
|
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
|
|
|
|
Heisman Winner [138129]
TigerPulse: 100%
Posts: 63868
Joined: 10/22/00
|
Do you ever, like, go to post something,
Jan 11, 2017, 9:12 AM
|
|
and then think "y'know, maybe I shouldn't post this"? No? I know the feeling.
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
I post it because I like it.
Jan 11, 2017, 9:13 AM
|
|
I recognize a lot of people don't like it. That's OK.
|
|
|
|
|
Hall of Famer [21888]
TigerPulse: 100%
Posts: 28649
Joined: 8/14/01
|
Makes perfect sense
Jan 11, 2017, 9:14 AM
|
|
In bizarro world. But hey, we're #1 in the real world so. Yea.
|
|
|
|
|
All-In [47835]
TigerPulse: 100%
Posts: 44540
Joined: 9/5/02
|
^^^^folks, a USC education at its finest^^^^***
Jan 11, 2017, 9:20 AM
|
|
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
I said something wrong, that I need to correct.
Jan 11, 2017, 9:22 AM
|
|
I said it's not the first time the national champion was not #1 in my final ratings. It actually is the first time (this is my 5th year doing it). I must have been thinking of my basketball ratings when I said that.
|
|
|
|
|
Hall of Famer [21888]
TigerPulse: 100%
Posts: 28649
Joined: 8/14/01
|
No worries, all knew it was a typo. Of course it is
Jan 11, 2017, 9:27 AM
|
|
the first time. I can think of absolutely no reason this year wouldn't be the first time.
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
No not a typo. I was mistaken.
Jan 11, 2017, 9:28 AM
|
|
I said something without validating what I was saying.
|
|
|
|
|
Oculus Spirit [81246]
TigerPulse: 100%
Posts: 56281
Joined: 9/13/04
|
there's something invalidating your poll***
Jan 11, 2017, 9:52 AM
|
|
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
It's just for fun. I don't have any goal or aspiration
Jan 11, 2017, 9:59 AM
|
|
for it to be called "valid".
|
|
|
|
|
Oculus Spirit [81246]
TigerPulse: 100%
Posts: 56281
Joined: 9/13/04
|
Then how can you determine if its correct or not?
Jan 11, 2017, 10:01 AM
|
|
The idea of constructing a model is to determine real outcomes using different variables.
Your models failed, and invalidate the attempt.
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
That may be some people's purpose. It is not mine.***
Jan 11, 2017, 10:03 AM
|
|
|
|
|
|
|
Oculus Spirit [81246]
TigerPulse: 100%
Posts: 56281
Joined: 9/13/04
|
Then what is the purpose?***
Jan 11, 2017, 10:06 AM
|
|
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
For fun, and to provide incentive for me to keep up
Jan 11, 2017, 10:09 AM
|
|
with college football, especially since my rooting interests have not been particularly strong in several years.
Spreadsheets in general are a hobby of mine. I have many of them, on a vast array of subjects. One time, I made a huge spreadsheet breaking down the phylogeny (classification) of living organisms. I just like doing it.
|
|
|
|
|
Oculus Spirit [81246]
TigerPulse: 100%
Posts: 56281
Joined: 9/13/04
|
DId you determine all living organisms were actually rocks?
Jan 11, 2017, 10:11 AM
|
|
|
|
|
|
|
Legend [17753]
TigerPulse: 100%
Posts: 16306
Joined: 5/13/02
|
Which Finebaum caller are you?***
Jan 11, 2017, 9:24 AM
|
|
|
|
|
|
|
CU Medallion [60244]
TigerPulse: 100%
Posts: 42567
Joined: 11/30/98
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
It is not coincidence that multiple mathematical ratings
Jan 11, 2017, 9:29 AM
|
|
would have Alabama higher. Numbers are numbers. We don't decide national champions by numbers like this, and that's a good thing.
|
|
|
|
|
Heisman Winner [136328]
TigerPulse: 100%
Posts: 41736
Joined: 8/26/07
|
I tried to tell y'all those close games would cost us
Jan 11, 2017, 9:31 AM
|
|
a trophy!
You kept saying "just win baby" "a win is a win" "no such thing as an ugly win"
Now see where it got us? No trophy with a little prodigal head on top for the cabinet.
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
This is the closest #1 and #2 have been in 5 yrs of ratings.
Jan 11, 2017, 9:36 AM
|
|
2nd closest was Alabama edging out Oregon in my first year (2012). Alabama was actually #3 behind ND and Oregon going into the bowls, and jumped to #1 by obliterating the Irish.
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
Clemson's raw rating number, though #2, would have been
Jan 11, 2017, 9:40 AM
|
|
#1 in my ratings in each of the previous 3 years.
These were two really great teams, as measured by my ratings.
|
|
|
|
|
Oculus Spirit [81246]
TigerPulse: 100%
Posts: 56281
Joined: 9/13/04
|
I'm not even going get into the strained arithmetic you
Jan 11, 2017, 9:51 AM
|
|
did to determine your rankings, but just look at the results. Whatever you are doing to come up with the results is invalidating your answer.
The result tells something is incorrect in your calculations.
Is Clemson NOT the National Champion? If that's true (and it is), logically, how could they STILL be ranked < #1?
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
It's math. Math doesn't know "national champions".
Jan 11, 2017, 9:57 AM
|
|
That's not a number that can be put in a formula.
|
|
|
|
|
Oculus Spirit [81246]
TigerPulse: 100%
Posts: 56281
Joined: 9/13/04
|
I understand math.
Jan 11, 2017, 10:00 AM
|
|
I've taken a lot of it in college, too.
But your weighted criteria counter the actual result. This should be telling you your formulas are incorrect.
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
There is no correct or incorrect.
Jan 11, 2017, 10:02 AM
|
|
It says what it says. It doesn't match most people's opinion (it certainly doesn't match up with mine).
|
|
|
|
|
Oculus Spirit [81246]
TigerPulse: 100%
Posts: 56281
Joined: 9/13/04
|
I guess I'm not sure what your measuring then,
Jan 11, 2017, 10:13 AM
|
|
and how you would know if it's correct.
You say it doesn't matter. Whatever.
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
This year, for each team, there were 75 columns
Jan 11, 2017, 9:56 AM
|
|
of data entry for each of the 128 teams, and an addition 82 columns of calculation/formulas. That went into the rating.
That's beside the conference ratings, and the division ratings.
|
|
|
|
|
Orange Blooded [3928]
TigerPulse: 100%
Posts: 4709
Joined: 9/2/03
|
There are lies, there are #### lies, and there are statistic
Jan 11, 2017, 11:22 AM
|
|
s
|
|
|
|
|
Orange Blooded [2013]
TigerPulse: 100%
Posts: 958
Joined: 4/19/16
|
So if Dabo would have left the 1st team in for
Jan 11, 2017, 10:10 AM
|
|
games...let's say against Sakerlina...and scored two more tds, making it 70 to 7, or the OSU game, Cuse game, SC State game, etc., would that had made a difference in your equation?
Assuming the competition didn't score any more points, how many more tds did Clemson need to beat Bama in the Prod rankings?
I thoroughly enjoy your rankings and posts. Thanks for sharing.
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
Yep, absolutely. If Clemson had scored 13 more points
Jan 11, 2017, 10:12 AM
|
|
Or allowed 13 more points, or a combination of the two adding up to 13 points, during the entire season, they would have been #1 in this rating.
So, one more point a game would have done it. Or scoring two more touchdowns during the season, etc.
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
*allowed 13 LESS points****
Jan 11, 2017, 10:15 AM
|
|
|
|
|
|
|
CU Medallion [60244]
TigerPulse: 100%
Posts: 42567
Joined: 11/30/98
|
so Ray Ray dropping the ball at the goal line on his punt
Jan 11, 2017, 10:17 AM
[ in reply to Yep, absolutely. If Clemson had scored 13 more points ] |
|
return, and then Clemson kneeling down on the Gamecocks instead of trying to score that last TD at the end of the game, actually changes your rankings?
That's not too good.
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
Oh, yes. Any impact on scoring impacts the ratings.
Jan 11, 2017, 10:19 AM
|
|
I'd like for meaningless scores not to mean anything in my ratings, but I don't know of a way to classify some scoring as meaningless, and other scoring not meaningless. I either factor in scoring, or I don't.
It's possible to go deeper and go by yards, etc. That's what the big boys, like FPI, do. But I don't have that sort of expertise or time.
|
|
|
|
|
All-In [31938]
TigerPulse: 100%
Posts: 37222
Joined: 11/22/03
|
You can do that by putting a limit to the margin...
Jan 11, 2017, 10:32 AM
|
|
like capping at 21 pts, etc...
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
That's not a bad idea. I'd have to think about
Jan 11, 2017, 10:35 AM
|
|
the implications. Sometimes, isn't it important that a team won by 40 instead of 21? That is a big difference. May be a more appropriate cap would be 35 or something. If you won by 35, you kicked butt thoroughly, for sure.
I think this would be even better in basketball, where really good teams sometimes play really bad teams (even D2) teams, and win by ungodly totals. (I have literally a couple of 100+ pt margins in NCAA basketball this season.)
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
The way I do it, I don't put in scoring margin of each game
Jan 11, 2017, 10:38 AM
|
|
I put in points scored and points allowed each game, and calculate the total margin for the season. So I'd have to change the way that is configured, to be able to put a "cap" on scoring margin.
|
|
|
|
|
All-In [31938]
TigerPulse: 100%
Posts: 37222
Joined: 11/22/03
|
|
|
|
|
All-In [45885]
TigerPulse: 100%
Posts: 23865
Joined: 2/1/99
|
In case you didn't see it on the field
Jan 11, 2017, 10:25 AM
|
|
Clemson best Alabama thus making you wrong. Typical coot.
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
Not me wrong. I think Clemson proved to be
Jan 11, 2017, 10:31 AM
|
|
the best team in the nation.
|
|
|
|
|
CU Medallion [60244]
TigerPulse: 100%
Posts: 42567
Joined: 11/30/98
|
|
|
|
|
All-In [43934]
TigerPulse: 100%
Posts: 18181
Joined: 4/24/15
|
COOT***
Jan 11, 2017, 10:28 AM
|
|
|
|
|
|
|
All-TigerNet [10161]
TigerPulse: 100%
Posts: 13972
Joined: 7/31/04
|
With that logic here is something to ponder...
Jan 11, 2017, 10:55 AM
|
|
Let's look at it this way; If two countries are at war with one another, think Nazi Germany. Germany is defeated and their leader blows his head off as he did, does Nazi Germany still rule the world? Heck No!
|
|
|
|
|
All-TigerNet [10161]
TigerPulse: 100%
Posts: 13972
Joined: 7/31/04
|
By looking at the conference that won more bowl games than
Jan 11, 2017, 11:01 AM
|
|
any other, The ACC Rules! SEC struggled to finish 2nd, but 2nd they were.
|
|
|
|
Replies: 68
| visibility 1
|
|
|