Posted by the0nlyis on 11/3/2015 11:21:00 AM (view original):

Posted by all3 on 11/3/2015 10:56:00 AM (view original):
I'm guess the formula is already giving teams a ton of credit for beating rotten opponents, so why give them even more credit for compiling great stats against that same rotten opponent? Maybe lessening one of those factors would help regulate teams like Iowa, Penn St., App. St., and the like.

everything is sos weighted it's just that teams like app st are just destroying their schedule by so much they are still liked even though they have a weak sos, adding anymore value to sos and teams like USC, Texas Tech, and West Virginia start to shoot up the rankings.

which probably means you have something wrong with your methodology.  There is something in your system that isn't working right.

When colonels first put his system out he didn't have tiers in his schedule component.  I think when he added in tiers it helped with his full methodology.  What I mean by tiers is there is no real difference between the 100th ranked team and the 120th ranked team so a win over them should be treated the same.  His system originally treated a win over #1 vs. #21 the same as a win over #100 vs. #120, which is just silly. 

Posted by the0nlyis on 11/3/2015 11:21:00 AM (view original):

Posted by all3 on 11/3/2015 10:56:00 AM (view original):
I'm guess the formula is already giving teams a ton of credit for beating rotten opponents, so why give them even more credit for compiling great stats against that same rotten opponent? Maybe lessening one of those factors would help regulate teams like Iowa, Penn St., App. St., and the like.

everything is sos weighted it's just that teams like app st are just destroying their schedule by so much they are still liked even though they have a weak sos, adding anymore value to sos and teams like USC, Texas Tech, and West Virginia start to shoot up the rankings.

I was kind of hinting that stats appear to need deemphasized a bit.

He's only doing it using stats.  

I don't even get what you're disputing.   I say "I don't care that Ole Miss beat Bama.  Bama is the better team".    You have a fit(with moranis throwing a semi-fit in the background).    This guy produces a formula that negates my statement, only using results, and you say he needs to do it differently because the results aren't telling the full story.     Is your distaste for anything I post that bad?   That you just have to argue against it even if you agree?

I'm not sure what the problem here is.

Are you saying my method is bad because they don't align with your specific opinions, or they aren't aligning with popular opinion.

And the whole think its being based on stats every single input is a statistic, chaging that takes away from the "computer" poll statistic.  If I wanted to do something not based on statistics I'd just do it based on my opinion, but I decided to create a computer poll fon stats alone

I think the problem moranis/all3 is having is that your numbers don't agree with their opinion.     They don't agree with mine either but I know numbers-only based formulas will create outliers.    I've done dozens upon dozens of them for my HBD teams and every last one has had an outlier and/or a result I didn't buy.

Although, to be fair, moranis and I have agreed on one thing about formulas.    The difference between team 80 and 120 is much smaller than the difference between 1 and 8.   There have to be levels instead of just flat out Team 2 and Team 27. 

App St is a problem in this ranking.  The only good team it played crushed it and it barely beat a 2 win team in its last game needing 3OT's to win by 3. 

If you look at the composite poll from the BCS guys I posted last week, Wolfe has App St at 27 the others are 34, 42, 52, and 55.  Again leading one to the conclusion that this ranking has to be pretty similar to Wolfe's methodology. 

If you look at a huge composite http://masseyratings.com/cf/compare.htm.  App St comes in at 40.  And yes it is 7-1 with 39 ppg and 16 papg, but its schedule is awful.  I meant absolutely atrocious.  It beat 6-2 Georgia Southern, but Troy at 2-6 is its next best win and that includes a 1 win FCS school.  A schedule rank (according to Massey of 122) can't be much worse and frankly its remaining schedule isn't much better.

Given that it appears schedule strength is not being given enough weight in the formula being used. 

Even when Mike has no idea what people are saying, (which is most of the time) he tries to explain it to others. Many times he "credits" people with comments / thoughts that they never made. Don't bother even reading his interpretation of other people's posts. If you question them, ask them directly.
To be as short and clear as possible, I applaud anyone who takes the time and effort to create something like this, and respect that they would ask for input. I don't give a rat's azz if they agree with my thoughts or not. I just think it appears that the yards per game stats might need devalued a bit, as games against "bad" teams can easily greatly distort them.

No one has an idea what you're saying.    You make about zero sense.    Hence, this post:


"I don't even get what you're disputing.   I say "I don't care that Ole Miss beat Bama.  Bama is the better team".    You have a fit(with moranis throwing a semi-fit in the background).    This guy produces a formula that negates my statement, only using results, and you say he needs to do it differently because the results aren't telling the full story.     Is your distaste for anything I post that bad?   That you just have to argue against it even if you agree?"

i never said I agree with App state they are a clear outlier, but I mentioned adding anymore sos weight to the formula and I see more than 1 outlier as USC jumps to the top 10, and depending how much more sos is weighed they've gotten as high as #3 with some formula's. West Virginia and Texas Tech start to make top 25.  You'd love Iowa still being #2 with sos heavily valued.

Rather have 1 outlier in App St then multiple more outliers that come in when SOS is weighted too heavy.

I think the problem may be how SOS schedule is usually calculated.  I have some questions for the0nlyis. If a team wins a game against a 8-2 team and a 2-8 team, will they get the same "credit" (all other things equal) as beating two 5-5 teams?  Also, do you have a margin of victory cap for each game?

Posted by MikeT23 on 11/3/2015 6:02:00 PM (view original):
No one has an idea what you're saying.    You make about zero sense.    Hence, this post:


"I don't even get what you're disputing.   I say "I don't care that Ole Miss beat Bama.  Bama is the better team".    You have a fit(with moranis throwing a semi-fit in the background).    This guy produces a formula that negates my statement, only using results, and you say he needs to do it differently because the results aren't telling the full story.     Is your distaste for anything I post that bad?   That you just have to argue against it even if you agree?"

Do I think you're a huge azz? Yes! Do you "misremember" things? Yes. Do I care what you think or misremember? No.
Everyone can see it is you who feels the need to criticize my comments to the0nlyis, even though there is zero need for you to comment at all, so why don't you just stop, while you're way behind, as usual?
The 41,000+ posts make it clear to everyone who posts just to argue.
Sorry for the sideshow the0nlyis.

You're a moron.   I've decided the best course of action is to block you so your stupidity will not affect me.   Have good, yet surprisingly stupid, life.

Posted by MikeT23 on 11/4/2015 11:12:00 AM (view original):
You're a moron.   I've decided the best course of action is to block you so your stupidity will not affect me.   Have good, yet surprisingly stupid, life.

I guess I'm supposed to feel hurt, but instead I feel incredibly lucky.