Well, another perspective would be to look where you're ranked and who you'd rather play.    If I'm ranked 60th, give me the 54/56 games.    I might lose both but I'd have a better chance at winning both.   At 10/100, I'm 1-1 in all likelihood with little chance of changing the outcomes.

Posted by the0nlyis on 11/16/2015 2:53:00 PM (view original):

Rank Name F Value

1 Clemson 272.9930587

2 Alabama 268.5984972

3 Ohio State 248.0676603

4 Iowa 246.0954116

5 Oklahoma 242.011896

6 Notre Dame 241.2933581

7 Navy 236.9352634

8 Florida 232.7363437

9 North Carolina 230.4889584

10 Michigan 229.6540185

11 Oklahoma State 229.3133961

12 Houston 227.2142976

13 Baylor 222.0726151

14 Mississippi 219.9529392

15 USC 218.8777541

16 Stanford 218.4123785

17 Michigan State 217.8912753

18 Memphis 217.5080855

19 LSU 216.7265517

20 Florida State 214.2713176

21 Mississippi State 214.2290543

22 Utah 210.1556701

23 Wisconsin 208.3960826

24 TCU 207.9107658

25 Toledo 206.1602989

 

RK	TEAM
1	Clemson
2	Alabama
3	Ohio State
4	Notre Dame
5	Iowa
6	Oklahoma State
7	Oklahoma
8	Florida
9	Michigan State
10	Baylor
11	Stanford
12	Michigan
13	Utah
14	Florida State
15	LSU
16	Navy
17	North Carolina
18	TCU
19	Houston
20	Northwestern
21	Memphis
22	Ole Miss
23	Oregon
24	USC
25	Wisconsin

That is the current playoff rankings.  You have several teams (not just Navy) that are ranked quite differently than their rankings.

I have

Navy at +9
USC +9
UNC at +8
Mississippi +8
Houston +7

Okie State -5
Michigan State -8

So Navy and USC are very highly rated in my model.  Navy has the 21st Offense and 15th defense with a 8-1 record.  USC has 11th offense and 21 defense with a 7-3 record.

Struggling teams like Okie State and MSU have 12th offense and 65th defense.  MSU has 44th and 32nd.

Makes sense.  MSU really doesn't seem that great at all with injuries, lucky wins, and playing down to their competition.  A team like USC is very talented and has had coaching troubles.

Okie State will jump with their back-loaded schedule

You are rating offense and defense rankings far too much.

uh this is a computer poll hence why good staistical teams are jumping up

depends on the stats you plug in and how you rate those stats.  You are rating offense and defense rankings too highly.

Posted by moranis on 11/20/2015 1:30:00 PM (view original):
depends on the stats you plug in and how you rate those stats.  You are rating offense and defense rankings too highly.

What else do you think should be incorporated

this week definitely ****** up the rankings

Rank	Name	F Value
1	Alabama	272.39
2	Clemson	265.19
3	Oklahoma	249.71
4	Navy	241
5	Ohio State	239.74
6	Iowa	238.01
7	Baylor	235.6
8	Notre Dame	235.34
9	Michigan	235.02
10	Michigan State	229.72
11	North Carolina	226.19
12	Florida	226.14
13	Mississippi	225.33
14	Florida State	220.23
15	Oklahoma State	219.99
16	Stanford	219.6
17	Mississippi State	216.58
18	Toledo	212.02
19	Houston	211.37
20	Northwestern	211.31
21	TCU	206.72
22	LSU	206.72
23	Tennessee	205.95
24	Pittsburgh	205.11
25	Texas A&M	204.68

Probably going to need to figure a way out to adjust this, the SOS adjustment I've been using I don't really like.

take out the offense and defense and I bet you it looks a lot more realistic

"I think it would be better to place a value on each game result individually (then add them up)."  I said this earlier in the thread.........

Posted by moranis on 11/22/2015 12:32:00 PM (view original):
take out the offense and defense and I bet you it looks a lot more realistic

take out offense and defense and i have like nothign left

then maybe that should tell you something.

Posted by the0nlyis on 11/22/2015 4:13:00 PM (view original):

Posted by moranis on 11/22/2015 12:32:00 PM (view original):
take out the offense and defense and I bet you it looks a lot more realistic

take out offense and defense and i have like nothign left

http://www.mvgsports.com/2015ncaaf   what do you think about colonels19's computerized rankings?

his pass the eye test a lot more, but it took him like 10 years of tinkering to get to that point.