|
|
#71
11-20-2007, 12:19 PM
|
|||
|
|||
Re: NCAAF Rank\' Em (approaching Week 13)
[ QUOTE ]
I don't worship the predictor like most, because I tracked every single game last year using the Sagarin predictor, and ATS it was worse than 50%. [/ QUOTE ] I think this should be totally expected. Do have its performance straight up by any chance? 
|
|
#72
11-20-2007, 01:50 PM
|
|||
|
|||
|
Re: NCAAF Rank\' Em (approaching Week 13)
[ QUOTE ]
Oh, my SOS numbers are Sagarin's SOS. [/ QUOTE ] Well you'd better make up your mind on Sagarin because he says Kansas is a lot better. Also, the Sagarin SoS rank doesn't tell the whole story. That's why undefeated Kansas was sometimes ahead of other undefeated teams with a better schedule rank. M2TR's blog explains it well, but essentially if you played 2 #40 teams or the #1 team and a #80 team it gives about the same SoS number but one schedule is much much harder.
|
|
#73
11-20-2007, 01:57 PM
|
|||
|
|||
|
Re: NCAAF Rank\' Em (approaching Week 13)
[ QUOTE ]
[ QUOTE ] Oh, my SOS numbers are Sagarin's SOS. [/ QUOTE ] Well you'd better make up your mind on Sagarin because he says Kansas is a lot better. Also, the Sagarin SoS rank doesn't tell the whole story. That's why undefeated Kansas was sometimes ahead of other undefeated teams with a better schedule rank. M2TR's blog explains it well, but essentially if you played 2 #40 teams or the #1 team and a #80 team it gives about the same SoS number but one schedule is much much harder. [/ QUOTE ] And which schedule would you say Kansas has played with their 0-0 record vs. the Top 35? Your point's completely irrelevant.
|
|
#74
11-20-2007, 02:56 PM
|
|||
|
|||
|
Re: NCAAF Rank\' Em (approaching Week 13)
Iggy, I've pointed this out before but many of our road games are equivalent (according to Sagarin) to playing a top 30 team at home.
If we were 3-0 vs the top 30 but they were all at home I suspect your criticism would be much less. My point was simply that our schedule is actually pretty decent despite having some very bad teams on it. I think the home/road factor is the #1 thing that gets overlooked. For instance, I'd gladly take Texas in Lawrence over playing at KSU, but we all know which one we'd get more credit for.
|
|
#75
11-20-2007, 03:39 PM
|
|||
|
|||
|
Re: NCAAF Rank\' Em (approaching Week 13)
[ QUOTE ]
Well you'd better make up your mind on Sagarin because he says Kansas is a lot better. [/ QUOTE ] How do I need to make up my mind? I said Sagarin was a good tool, but not the end all. His SOS #s are better than most. Kansas's SOS from Sagarin is #101!
|
|
#76
11-20-2007, 04:05 PM
|
|||
|
|||
|
Re: NCAAF Rank\' Em (approaching Week 13)
[ QUOTE ]
[ QUOTE ] Oh, my SOS numbers are Sagarin's SOS. [/ QUOTE ] Well you'd better make up your mind on Sagarin because he says Kansas is a lot better. Also, the Sagarin SoS rank doesn't tell the whole story. That's why undefeated Kansas was sometimes ahead of other undefeated teams with a better schedule rank. M2TR's blog explains it well, but essentially if you played 2 #40 teams or the #1 team and a #80 team it gives about the same SoS number but one schedule is much much harder. [/ QUOTE ] Alright, just read MT2R's blog about SOS, and although I respect him, his thoughts have flaws, too. I don't think he'll take this as me being a terd (we've had numerous conversations about this foooozball we love). His ideas about SOS have merit, for sure. However, it falls into the same trap as saying "Tennessee beat #6 Georgia" in 2006. Georgia went on to be a very mediocre team in 2006, and the victory by Tennessee was obviously not as impressive. That's the way the lines work, too. The lines aren't made in hindsight, obviously. Lines are made when we know what we know. Making a SOS based on how MT2R was alluding to would be worse, IMO, than the current system (and the current system stinks too). Take this year, for example. Kentucky was a dog at top 10 Louisville, and won the game. In MT2R's approach, that win by Kentucky would be wayyyyyyy overvalued. We all know Louisville is average or below. However, MT2R's SOS would see "ZOMG Kentucky won on the road as a 6 point dog" even though NOW in November, Kentucky would probably be favored in that matchup. Does this make sense? In the end, no system for SOS has it nailed. Hopefully one day ...
|
|
#77
11-20-2007, 04:19 PM
|
|||
|
|||
|
Re: NCAAF Rank\' Em (approaching Week 13)
MT2R uses the current Sagarin rankings to retroactively set hypothetical lines for previous games. If a team turns out to suck, that will be adjusted for, it's not just based on what the spread was when the game was played.
|
|
#78
11-20-2007, 04:27 PM
|
|||
|
|||
|
Re: NCAAF Rank\' Em (approaching Week 13)
[ QUOTE ]
MT2R uses the current Sagarin rankings to retroactively set hypothetical lines for previous games. If a team turns out to suck, that will be adjusted for, it's not just based on what the spread was when the game was played. [/ QUOTE ] Ahha - somehow I missed that and it makes much more sense. Still minor flaws, because Sagarin is flawed, but good stuff all around.
|
|
#79
11-20-2007, 07:27 PM
|
|||
|
|||
|
Re: NCAAF Rank\' Em (approaching Week 13)
[ QUOTE ]
[ QUOTE ] Well you'd better make up your mind on Sagarin because he says Kansas is a lot better. [/ QUOTE ] How do I need to make up my mind? I said Sagarin was a good tool, but not the end all. His SOS #s are better than most. Kansas's SOS from Sagarin is #101! [/ QUOTE ] Someone mentioned that Kansas got criticized for its schedule, but Oklahoma didn't. Ftr, even after a Missouri game and an Oklahoma title game (which may not happen since OSU may beat them), the SOS for Kansas will still not even be up to Oklahoma's. Otoh, Missouri's SOS will most likely be in the top 20 or even top 15 after same two games.
|
|
#80
11-20-2007, 07:30 PM
|
|||
|
|||
|
Re: NCAAF Rank\' Em (approaching Week 13)
[ QUOTE ]
[ QUOTE ] [ QUOTE ] Missouri's resume, IMO, is better than Kansas's. Look at common opponents. Look at SOS. Look at talent. In my mind, if Kansas beats Missouri this weekend, Kansas will have proven they won 1 good game. Beat OU in the B12 title game, then I'll feel fine about Kansas. Kansas STILL has the worst SOS of any BCS conference team (#101 now). [/ QUOTE ] Pirateboy, put aside the bias for a minute, come on. If it were true that you've performed better given victory margins and opponents you wouldn't be 12th in Sagarin's predictor. Also, your understanding of SoS is very poor. M2TR's blog can help you out. [/ QUOTE ] And it is true that Missouri has beaten common opponents better than Kansas has. Colorado - Kansas by 5, Missouri by 45 Kansas State - Kansas by 6, Missouri by 17 TAMU - Kansas by 8, Missouri by 14 Nebraska - Kansas by 37, Missouri by 35 Iowa State - Kansas by 38, Missouri by 14 So, overall, Missouri has a 31 point edge over Kansas in their 5 common games. That's a 6 point edge. [/ QUOTE ] The scores don't even tell the whole story. Kansas beat Nebraska 76-39 and Missouri beat Nebraska 41-6. Kansas won by 2 more, but which score is more impressive? If I were putting down any money, I'd take Missouri. My guess is Missouri by 7.
|
 |
|
|
Linear Mode
