Started By
Message
SOS needs to be an actionable and objective variable
Posted on 12/2/25 at 10:12 am
Posted on 12/2/25 at 10:12 am
My idea is make a baseline of #20 for SOS. For every 10 points in either direction, you either gain or subtract a loss. The metric can be fractional(not sure if this is right word), meaning that being #15 gets you .5 less a loss.
All conferences agree to an objective measure of SOS(preferably based off metrics based power ratings) obviously and all conferences agree to mandatory cross conference OOC scheduling.
All conferences agree to an objective measure of SOS(preferably based off metrics based power ratings) obviously and all conferences agree to mandatory cross conference OOC scheduling.
This post was edited on 12/2/25 at 10:14 am
5
Posted on 12/2/25 at 10:13 am to scottydoesntknow
Posted on 12/2/25 at 10:14 am to scottydoesntknow
quote:
he metric can be fractional(not sure if this is right word), meaning that being #15 gets you .5 less a loss.
So a team that's #15 in your rankings would be penalized less for a loss than #20? Is that how I'm understanding this?
Posted on 12/2/25 at 10:17 am to BluegrassBelle
quote:
quote:
he metric can be fractional(not sure if this is right word), meaning that being #15 gets you .5 less a loss.
So a team that's #15 in your rankings would be penalized less for a loss than #20? Is that how I'm understanding this?
Yes
Posted on 12/2/25 at 10:18 am to Jon Ham
Posted on 12/2/25 at 10:20 am to scottydoesntknow
This may be an unpopular opinion, but SOS is the worst metric to use for ranking teams. In order to establish a strength of schedule, you first have to rank the teams to determine relative strength.
So you have to rank the teams to create a metric to rank the teams. It’s circular. SOS is just a smoke screen to hide whatever formula they used to establish their relative strength.
So you have to rank the teams to create a metric to rank the teams. It’s circular. SOS is just a smoke screen to hide whatever formula they used to establish their relative strength.
This post was edited on 12/2/25 at 10:21 am
Posted on 12/2/25 at 10:25 am to scottydoesntknow
quote:
Yes
So even if you lose to a lower ranked team, if you're ranked higher your loss is penalized less? But then if you're a lower ranked team that loses to a higher ranked team (rightfully) you're penalized more?
That doesn't make any sense.
Posted on 12/2/25 at 11:33 am to BluegrassBelle
quote:
So even if you lose to a lower ranked team, if you're ranked higher your loss is penalized less? But then if you're a lower ranked team that loses to a higher ranked team (rightfully) you're penalized more? That doesn't make any sense.
Thats because your logic makes no sense. What you are ranked means nothing. Im not sure why you even mention that.
If your schedule is a 9 game SEC schedule, with the Eagles, Giants, Cowboys as your OOC teams and you finish 9-3, it should absolutely be accounted for when measuring that against an 11-1 team with one ranked win against number #22 and 10 creampuffs.
You cant just say "11-1 team is more deserving because they blew out 10 creampuffs and have 2 less losses" Right now the 11-1 team is given better seed 95% of the time(depending on conference).
We cant mutually agree to a playoff where one conference is always handicapped by having more better teams that must play each other. Either the other conferences are gonna have to start playing better schedules or the SEC is gonna say "hell no" once the next negotiation comes around
This post was edited on 12/2/25 at 11:34 am
Posted on 12/2/25 at 11:52 am to Jon Ham
Thanks for that link. Mizzou is #20 after hearing all year our SOS was horrible.
Posted on 12/2/25 at 11:57 am to Jon Ham
This link is crazy different from other SOS rankings, which is the entire problem.
If SoS is a thing, it should have one formula that everyone can apply correctly.
I certainly don't believe OSU has played the 8th hardest schedule, and there's no way Oregon has played the hardest schedule
So that link is interesting and very different from all the other SoS rankings I've seen
If SoS is a thing, it should have one formula that everyone can apply correctly.
I certainly don't believe OSU has played the 8th hardest schedule, and there's no way Oregon has played the hardest schedule
So that link is interesting and very different from all the other SoS rankings I've seen
Posted on 12/2/25 at 12:15 pm to scottydoesntknow
SOS would not take into account rivalries.
My thought is that there should be a scoring system based on an algorithm that takes into account the opponents last games and next games to adjust for outlier games, common opponents, points for wins, minus points for losses, and points adjusted for point spread of wins, point spread of losses, home vs away wins and losses, etc.
If your team plays a gimme game that has a low power ranking, like Southeastern, you get maybe 1 point for the win, whereas as win against a mid tier is worth 5 points and a top 10 team is worth 10 points. The points can adjust retroactively as your opponents power ranking changes throughout the season.
A point system that takes away emotions and is based on a standard for every team and game would be ideal.
My thought is that there should be a scoring system based on an algorithm that takes into account the opponents last games and next games to adjust for outlier games, common opponents, points for wins, minus points for losses, and points adjusted for point spread of wins, point spread of losses, home vs away wins and losses, etc.
If your team plays a gimme game that has a low power ranking, like Southeastern, you get maybe 1 point for the win, whereas as win against a mid tier is worth 5 points and a top 10 team is worth 10 points. The points can adjust retroactively as your opponents power ranking changes throughout the season.
A point system that takes away emotions and is based on a standard for every team and game would be ideal.
This post was edited on 12/2/25 at 12:27 pm
Posted on 12/2/25 at 12:21 pm to scottydoesntknow
quote:
Thats because your logic makes no sense. What you are ranked means nothing. Im not sure why you even mention that.
Because you said:
quote:
The metric can be fractional(not sure if this is right word), meaning that being #15 gets you .5 less a loss.
And then you confirmed that your ranking would affect the fraction in which the loss would be counted:
quote:
So a team that's #15 in your rankings would be penalized less for a loss than #20? Is that how I'm understanding this?
quote:
Yes
Posted on 12/2/25 at 12:40 pm to BluegrassBelle
quote:
quote:The metric can be fractional(not sure if this is right word), meaning that being #15 gets you .5 less a loss. And then you confirmed that your ranking would affect the fraction in which the loss would be counted:
No I didnt. I confirmed that your SOS would be applied in an objective way when comparing other teams for the playoff. What you are ranked is meaningless to your SOS. Why is this such a difficult concept for you?
Your SOS affects the way your loss is counted, not your ranking. How you came to that conclusion, I have no idea
*ok I see the confusion. You used the word "rankings" for SOS rating...which is weird but whatever. Yes I am saying that a team with a SOS rating of 15 would be granted .5 less TOTAL losses as a direct comparison metric only to other similar teams. That wouldnt be the ONLY metric but it would be one so that the Committee couldnt use losses as justification for an 11-1 team with a 40 SOS getting in over a 10-2 team with 10 SOS.
This post was edited on 12/2/25 at 12:46 pm
Posted on 12/2/25 at 12:45 pm to scottydoesntknow
quote:
metric can be fractional(not sure if this is right word)
Divided. The word you’re looking for is divided.
Posted on 12/2/25 at 12:47 pm to Buckeye06
quote:
If SoS is a thing, it should have one formula that everyone can apply correctly.
We should just reinstall the BCS but with a 12-team playoff. I like the idea of just making it a math problem.
Posted on 12/2/25 at 12:47 pm to scottydoesntknow
It absolutely should be. Honestly what they need to do is go back to the BCS computer algo. Moving away from that was always a big mistake. You’re going to trust a bunch of media dipshits and politicians to pick the teams in the playoffs? Give me a break.
Posted on 12/2/25 at 12:49 pm to scottydoesntknow
quote:
If your schedule is a 9 game SEC schedule, with the Eagles, Giants, Cowboys as your OOC teams and you finish 9-3, it should absolutely be accounted for when measuring that against an 11-1 team with one ranked win against number #22 and 10 creampuffs.
You cant just say "11-1 team is more deserving because they blew out 10 creampuffs and have 2 less losses" Right now the 11-1 team is given better seed 95% of the time(depending on conference).
quote:
SOS needs to be an actionable and objective variable
IOW they should have a system like the BCS to determine rankings. What a novel concept.
Posted on 12/2/25 at 12:51 pm to T1gerNate
quote:
BCS computer algo
Im fine with that too but no longer hiding the formula. We dont need to guess about it(which is what simulted BCS rankings are now). Didnt the BCS also use the coaches poll as an input...definitely stop that
This post was edited on 12/2/25 at 12:52 pm
Page 1 of 1
Popular
Back to top
Follow SECRant for SEC Football News