I Created a CFB Ranking System

#76
#76
bamawriter said:
You're completely correct, and I'm surprised no one has brought it up before now.

Here's why I didn't account for that:

It places a subjective value into what is an otherwise objective formula. How should a head-to-head victory be formulated? Are all head-to-head victories created equal? If Team A beats Team B by a field goal and they both finish 11-1, but Team B finishes with a much better SOS, should Team A inevitably be ranked ahead of Team B because of the head-to-head?

These questions are why math systems are so problematic.
Click to expand...

i agree its a mess, and I am not sure how you factor it. but isn't the whole point of your system to go off of results? as far as the numbers are concerned the formula isn't going to recognize that the SOS #52 that beat the SOS #30 is the same team ranked right behind SOS #52.

i guess the only way to account for stuff like that is to have a subjective (human) view of the rankings before final release but after the objective review is done.
 
#77
#77
bamawriter said:
Back in 2011, during the controversy surrounding the Bama-LSU rematch, I started putting together my own version of a BCS computer ranking system. I was going to use it in a future column to detail how stupid the very concept of the computers is when it comes to determining a champion. But, before I had the chance to write that column, the playoff was announced, and it seemed sort of pointless.

Anyway, I was clearing of a hard drive and found my system. Because I'm bored, I decided to update it for the past several seasons just to see what it would spit out for the four team playoff. Thought some on here might find it interesting. Certainly could spur some debate.

Here's the idea:

Bias is completely eliminated. Team names, history, and conference are all ignored. Winning % and strength of schedule are the most important factors. While the BCS eliminated margin of victory, I think that's ridiculous, so it's back in on mine. In order to adjust for home field advantage, I took the standard Vegas rule and subtracted 3 points for a game at home, and added 3 points for a road game, creating an adjusted point differential. I also included average yardage differential to help account for the way a game actually played out on the field. FCS opponents are given a standard .25 winning % and only their games against FBS teams count toward their point differential.

Here's the formula:

Winning % x 100 (too make it a whole number)
+
Opponents' Winning % x 100
+
Average Adjusted Point Differential
+
Opponents' Average Point Differential
+
Average Yardage Differential / 100
=
SCORE


As an example, 2015 Tennessee had the following score (bowl game not included):

(.6667 x 100) + (.5385 x 100) + 12.4167 + .5753 + (52.0833 / 100) = 134.0328

Interestingly, my system from last year produced the same 4 playoff teams, and even the same semifinal matchups, but with different seeding:

1. Alabama - 176.4984
2. Clemson - 175.6221
3. Oklahoma - 174.9711
4. Mich St. - 164.4958

Michigan State got in over Ohio State by .0007.

I went back to 2008 if anyone is interested in more.
Click to expand...

The problem with using straight opponent winning percentage is that you're saying a 3-1 Army team is a bigger win than 2-2 Ole Miss.
 
#78
#78
LouderVol said:
i agree its a mess, and I am not sure how you factor it. but isn't the whole point of your system to go off of results? as far as the numbers are concerned the formula isn't going to recognize that the SOS #52 that beat the SOS #30 is the same team ranked right behind SOS #52.

i guess the only way to account for stuff like that is to have a subjective (human) view of the rankings before final release but after the objective review is done.
Click to expand...

And that's really the point. The system doesn't look at names. It simply looks at whether a team won or lost and how, and whether its opponents' won or lost and how. That's it. It does not have the ability to watch what happens from week to week.
 
Last edited:
#79
#79
VolinArizona said:
The problem with using straight opponent winning percentage is that you're saying a 3-1 Army team is a bigger win than 2-2 Ole Miss.
Click to expand...

Because objectively, as of right now, beating a 3-1 team with a 21.75 point differential is a bigger win than beating a 2-2 team with a 7.75 point differential. You and I both know that Ole Miss is a better team than Army because we can watch the games. But that's a subjective analysis, and that's not what a math system is supposed to accomplish.
 
#80
#80
Average Adjusted Point Differential & Opponents' Average Point Differential...

is there a site or place to view these or do you calculate by hand?
 
#81
#81
LittleCat said:
Average Adjusted Point Differential & Opponents' Average Point Differential...

is there a site or place to view these or do you calculate by hand?
Click to expand...

I explain that number in the OP. It's just the score with home field taken into account.
 
#82
#82
bamawriter said:
I explain that number in the OP. It's just the score with home field taken into account.
Click to expand...

Yeah I had read it and I just now get it. Thanks.

Maybe I'm stupid but not getting the 12.4167 for 2015 UT though. Getting 12.667, but at least I understand it now lol. I'm sure I'm making a silly math error somewhere.
 
#83
#83
LittleCat said:
Yeah I had read it and I just now get it. Thanks.

Maybe I'm stupid but not getting the 12.4167 for 2015 UT though. Getting 12.667, but at least I understand it now lol. I'm sure I'm making a silly math error somewhere.
Click to expand...

I imported it from a stats database to excel. I'm not going to rule out an import or rounding error somewhere. I'm a Bama fan, so don't expect me to math and stuff.
 
#85
#85
LittleCat said:
Yeah I had read it and I just now get it. Thanks.

Maybe I'm stupid but not getting the 12.4167 for 2015 UT though. Getting 12.667, but at least I understand it now lol. I'm sure I'm making a silly math error somewhere.
Click to expand...

I like LC's version better!!!
 
#86
#86
LittleCat said:
Yeah I had read it and I just now get it. Thanks.

Maybe I'm stupid but not getting the 12.4167 for 2015 UT though. Getting 12.667, but at least I understand it now lol. I'm sure I'm making a silly math error somewhere.
Click to expand...

I checked your math.

Is it possible that you counted Bowling Green as an away game? Doing so brings UT's APD to 12.6667 like you said. I didn't make it clear in the OP, but I treat neutral site games as neither home nor away, so there is no adjusted score.
 
#87
#87
bamawriter said:
I checked your math.

Is it possible that you counted Bowling Green as an away game? Doing so brings UT's APD to 12.6667 like you said. I didn't make it clear in the OP, but I treat neutral site games as neither home nor away, so there is no adjusted score.
Click to expand...

I remember seeing that and I'm pretty sure I kept it at 29 assuming it was 0. I'm not sure where I went wrong. I'm terrible with any sort of math so oh well.
 
#88
#88
bamawriter said:
Back in 2011, during the controversy surrounding the Bama-LSU rematch, I started putting together my own version of a BCS computer ranking system.
Click to expand...

Impressive Bamawriter.

But not as impressive as this guy. He threw the first Hale Mary pass, and created the playoff system.
 

Attachments

  • dr-pepper-college-football-meet-larry-large-2.jpg
    dr-pepper-college-football-meet-larry-large-2.jpg
    42 KB · Views: 0
#90
#90
Week 5 Rankings (with AP/Coaches):

1. Michigan (5-0) - 200.9240 (4/4)
2. Ohio St (4-0) - 187.7925 (2/2)
3. Clemson (5-0) - 187.3100 (3/3)
4. Tennessee (5-0) - 183.6780 (9/9)
5. Alabama (5-0) - 175.4700 (1/1)
6. Texas A&M (5-0) - 170.2480 (8/7)
7. Houston (5-0) - 168.1900 (6/5)
8. Western Michigan (5-0) - 166.6440 (UR/25)
9. Washington (5-0) - 163.1680 (5/6)
10. Louisville (4-1) - 162.3320 (7/8)
11. Miami (4-0) - 160.2100 (10/10)
12. Maryland (4-0) - 155.7275 (UR/UR)
13. Boise St (4-0) - 155.2200 (19/19)
14. Wisconsin (4-1) - 153.2680 (11/13)
15. Nebraska (5-0) - 153.0880 (12/12)
16. Troy (4-1) - 152.9600 (UR/UR)
17. Virginia Tech (3-1) - 151.8475 (25/UR)
18. Stanford (3-1) - 150.6675 (15/15)
19. West Virginia (4-0) - 150.6300 (22/20)
20. Air Force (4-0) - 147.7100 (UR/UR)
21. Colorado (4-1) - 147.1080 (21/23)
22. Arkansas (4-1) - 144.3000 (16/17)
23. Oklahoma (2-2) - 143.2450 (20/22)
24. Wake Forest (4-1) - 142.8560 (UR/UR)
25. Florida (4-1) - 141.1680 (18/18)

Looking for a specific team? Let me know.
 
#91
#91
The Rest of the SEC:

26. Ole Miss (3-2) - 141.0000 (14/14)
32. Auburn (3-2) - 132.9720
46. LSU (3-2) - 126.0940
47. Georgia (3-2) - 125.3700
71. Missouri (2-3) - 107.3460
72. Vanderbilt (2-3) - 107.1380
75. Kentucky (2-3) - 105.8840
81. Mississippi St (2-2) - 99.3750
88. South Carolina (2-3) - 91.3500
 
#92
#92
Bottom 10:

119. Rice (0-5) - 62.8400
120. Nevada (2-3) - 62.4740
121. Kent St (1-4) - 61.0920
122. Florida Atlantic (1-4) - 60.8960
123. UTEP (1-4) - 59.0340
124. Florida International (1-4) - 58.4800
125. Louisiana-Monroe (1-3) - 58.0475
126. San Jose St (1-4) - 54.0180
127. Miami (OH) (0-5) - 48.5580
128. Arkansas St (0-4) - 40.0000
 
Last edited:
#94
#94
WEEK 6 Rankings (with AP/Coaches)

1. Michigan (6-0) - 201.8161 (4/4)
2. Clemson (6-0) - 188.9756 (3/3)
3. Ohio St (5-0) - 185.6380 (2/2)
4. Alabama (6-0) - 183.8706 (1/1)
5. Tennessee (5-1) - 176.6544 (9/11)
6. Texas A&M (6-0) - 169.1478 (6/6)
7. Louisville (4-1) - 165.3387 (7/7)
8. Boise St (5-0) - 162.6593 (15/15)
9. Washington (6-0) - 160.8700 (5/5)
10. Wisconsin (4-1) - 159.4280 (8/10)
11. Western Michigan (6-0) - 156.1406 (24/23)
12. West Virginia (4-0) - 155.7008 (20/18)
13. Houston (5-1) - 154.6700 (13/12)
14. Virginia Tech (4-1) - 153.0440 (17/19)
15. Troy (4-1) - 152.8933 (NR/NR)
16. Nebraska (5-0) - 150.0447 (10/9)
17. Navy (4-1) - 148.6433 (25/NR)
18. Florida St (4-2) - 147.1050 (14/16)
19. Penn St (4-2) - 144.5044 (NR/NR)
20. Ole Miss (3-2) - 144.1933 (12/13)
21. Auburn (4-2) - 142.4511 (23/NR)
22. NC State (4-1) - 141.1280 (NR/NR)
23. Wake Forest (5-1) - 140.9700 (NR/NR)
24. Florida (4-1) - 140.4013 (18/14)
25. Maryland (4-1) - 139.8807 (NR/NR)
 
#95
#95
The Rest of the SEC:

29. Arkansas (4-2) - 138.1183 (22/22)
44. LSU (3-2) - 124.6440 (NR/25)
50. Georgia (4-2) - 122.0033
66. Missouri (2-3) - 105.8993
67. Kentucky (3-3) - 105.4839
91. Vanderbilt (2-4) - 90.4178
93. Mississippi St (2-3) - 87.3873
105. South Carolina (2-4) - 80.4717
 
#96
#96
Bottom 10:

119. New Mexico St (2-3) - 65.8067
120. UMass (1-5) - 60.7922
121. Florida International (2-4) - 58.1039
122. Rice (0-5) - 55.1700
123. Louisiana-Monroe (1-4) - 52.5393
124. Florida Atlantic (1-5) - 49.9950
125. Buffalo (1-4) - 47.8900
126. Miami (OH) (0-6) - 47.3566
127. San Jose St (1-5) - 46.2006
128. UTEP (1-5) - 42.3161
 
#97
#97
bamawriter said:
WEEK 6 Rankings (with AP/Coaches)
5. Tennessee (5-1) - 176.6544 (9/11)
6. Texas A&M (6-0) - 169.1478 (6/6)
Click to expand...

this is where I still have a problem with your system. obviously because it has UT higher i don't mind too much.

even though I think we are the better team I don't see how the numbers come out for us to be ranked higher than the team that beat us with them being undefeated.

I go back to my point about having a pair of human eyes look this over again because I don't see how it makes sense.
 
#98
#98
LouderVol said:
this is where I still have a problem with your system. obviously because it has UT higher i don't mind too much.

even though I think we are the better team I don't see how the numbers come out for us to be ranked higher than the team that beat us with them being undefeated.
Click to expand...

According to my system, UT has the best strength of schedule in the country (opposing win % + opposing adjusted point differential). That's what is keeping them over A&M at the moment. Every one of UT's opponents is above .500 and only UGA has a negative point differential at -0.8333. UT's 6 opponents have lost a combined 3 games against teams other than UT.
 
Last edited:
#99
#99
bamawriter said:
According to my system, UT has the best strength of schedule in the country (opposing win % + opposing adjusted point differential). That's what is keeping them over A&M at the moment. Every one of UT's opponents is above .500 and only UGA has a negative point differential at -0.8333. UT's 6 opponents have lost a combined 3 games against teams other than UT.
Click to expand...

If your system is accurate, then last week plus the next two will be very exciting (ut a&m, ut bama, bama a&m).
 
New posts
You must log in or register to reply here.

VN Store



Back
Top