- My Forums
- Tiger Rant
- LSU Recruiting
- SEC Rant
- Saints Talk
- Pelicans Talk
- More Sports Board
- Winter Olympics
- Fantasy Sports
- Golf Board
- Soccer Board
- O-T Lounge
- Tech Board
- Home/Garden Board
- Outdoor Board
- Health/Fitness Board
- Movie/TV Board
- Book Board
- Music Board
- Political Talk
- Money Talk
- Fark Board
- Gaming Board
- Travel Board
- Food/Drink Board
- Ticket Exchange
- TD Help Board
Customize My Forums- View All Forums
- Show Left Links
- Topic Sort Options
- Trending Topics
- Recent Topics
- Active Topics
Started By
Message
re: lsumatt and Colley Future Polls Released
Posted on 10/28/10 at 10:50 am to LSUtoOmaha
Posted on 10/28/10 at 10:50 am to LSUtoOmaha
quote:
That Colley ranking does not look good for us. Looks like we'd have to have Boise and TCU lose to jump them.
I don't have us in the SECCG in that poll, but we would get a huge bump from that. Also, that is sort of a best case scenario in my opinion for Boise/TCU it assumes Oregon State, Va Tech, and Utah don't get upset. If LSU opponents get upset it will be by other top LSU opponents.
Posted on 10/28/10 at 10:51 am to The Mick
quote:
A 2-loss OU is ranked higher than undefeated Oregon? Wow.
I don't think it will actually happen that way. They are close..in fact about 5 of those teams are very close. Keep in mind that I have both of OU's losses to the #1 team (Missouri) so it doesn't hurt them that much.
Posted on 10/28/10 at 12:08 pm to lsumatt
You could probably research the home team winning percentage instead of jut assuming 10%. I'm curious to know exactly the quantity of advantage a home team has. I've always thought home-field advantage was overrated, so I would assume only a 2%-4% increase for the home team.
Posted on 10/28/10 at 12:12 pm to LSUJuice
You know, I could actually do that pretty quickly using all the game data I have.
...to be continued
...to be continued
Posted on 10/28/10 at 2:21 pm to lsumatt
sorry if this question has come up before, Matt, but what are the odds that if every other undefeated BCS team doesn't win out that LSU could face an undefeated (or one loss) AUBURN in the national championship?
I can't get excited about sneaking into the NC without winning our division. But I could stomach it if we got a rematch with AU on a neutral field.
And yes, I understand this is all pie in the sky. I just want something to root for.
Also, it seems like a one loss BAMA v. one loss AU rematch could be entirely possible, like when the press was urging for MICH v. OSU in 06 when FLA won.
I can't get excited about sneaking into the NC without winning our division. But I could stomach it if we got a rematch with AU on a neutral field.
And yes, I understand this is all pie in the sky. I just want something to root for.
Also, it seems like a one loss BAMA v. one loss AU rematch could be entirely possible, like when the press was urging for MICH v. OSU in 06 when FLA won.
Posted on 10/28/10 at 2:29 pm to lsumatt
Think you could do a little work on the theory of relativity while youre at it?
Cant be nearly as complex as the BCS...
Nice work!
Cant be nearly as complex as the BCS...
Nice work!
This post was edited on 10/28/10 at 2:30 pm
Posted on 10/28/10 at 2:36 pm to BobBoucher
same formula i use - but I got an answer of 2.
Posted on 10/28/10 at 4:46 pm to jeffsdad
Lsumatt you should start a website man. I think college fans around the country would be interested in your breakdown for their teams.
Good job man as always!
Good job man as always!
Posted on 10/28/10 at 5:08 pm to hobo with a rolex
FWIW, I got that the home team wins 58% of the time (away team 42%). So that is a big difference.
To do that in the model you would have to scale it. If team A has a 99% chance of winning on a neutral sit, it obviously isn't 99 + 8% at home.
To do that in the model you would have to scale it. If team A has a 99% chance of winning on a neutral sit, it obviously isn't 99 + 8% at home.
Posted on 10/28/10 at 7:24 pm to lsumatt
that would be okay except they lose to nebraska this weekend in lincoln. They will NEVER beat OU in the big 12 CG. Can't beat them twice in 1 year. It isa always about "when" you lose. Why Bama is in the NC game assuming they run the table.
In lincoln the record since 1980 is
Neb 16 Mizzou 1
In lincoln the record since 1980 is
Neb 16 Mizzou 1
This post was edited on 10/28/10 at 7:30 pm
Posted on 10/28/10 at 8:13 pm to lsumatt
I see you screwed over LSU, Dude. 
Posted on 10/28/10 at 11:16 pm to lsumatt
Not ranked in the top ten?
Posted on 10/29/10 at 2:19 am to Dalosaqy
In the Colley poll, LSU is ranked ahead of Alabama. The "Futures poll" assumes that the highest ranked teams win out. Thus, Alabama would have three losses (LSU and then Auburn) in this model.
Posted on 10/29/10 at 10:30 am to LSUtoOmaha
This makes for good conversation but not much else. Too many assumptions. LSU will not win out from here...they will lose at least one more, but likely two more games.
Posted on 10/29/10 at 12:19 pm to lsumatt
quote:
FWIW, I got that the home team wins 58% of the time (away team 42%). So that is a big difference.
does that take into account rent-a-wins?
Posted on 10/29/10 at 12:34 pm to lsumatt
lsumatt, what algorithm(s) do you use to solve the nonlinear systems of equations in (2)? It seems to me that if you have 120 equations for each team, then it is a computationally intensive task to solve for the ratings you want.
Some algorithms may be faster than others for this particular problem, so from an efficiency standpoint, it may be worth a look. The Newton-Raphson algorithm may be slower than others.
By the way, good work.
Some algorithms may be faster than others for this particular problem, so from an efficiency standpoint, it may be worth a look. The Newton-Raphson algorithm may be slower than others.
By the way, good work.
This post was edited on 10/29/10 at 12:46 pm
Posted on 10/29/10 at 1:44 pm to LordSaintly
Newton is about as fast as it gets. There are other things I could do to improve speed, but it isn't much of an issue.
Where/what are you getting your PhD?
Where/what are you getting your PhD?
Popular
Back to top


0





