BCS computer rankings and the Big 12
I like that computers are part of the BCS rankings. They're objective. That doesn't mean they're smart, though, since any computer ranking is only as smart as the people writing the ranking algorithm.
I'll pick on Jeff Sagarin, since his rankings are the best known and he's the most extreme with the Big 12. Here are the Big 12 teams in his BCS rankings:
2. Oklahoma State (10-0; AP#2)
6. Oklahoma (8-1; AP#5)
7. Kansas State (8-2; AP#16)
9. Baylor (6-3; AP#25)
12. Texas A&M (5-5; no AP votes)
13. Texas (6-3; AP#31)
17. Missouri (5-5; no AP votes)
26. Texas Tech (5-5; no AP votes)
28. Iowa State (5-4; no AP votes)
66. Kansas (2-8; no AP votes)
For reference, he has Michigan at #27 and his top ranked Big Ten team, Michigan State, is #22. In other words, he believes that 70% of the Big 12 has had a more impressive season than every single Big Ten team. Every Big 12 team but Kansas has been better than Wisconsin.
It seems like these rankings are far too kind to teams struggling in conferences that perform well in a few nonconference games. The Big 12 performed relatively well, granted, but here are all of their games against AQ conference teams:
Wins (6): Arizona (OSU), Florida State (OU), Miami (K-State), UCLA (Texas), Iowa (ISU), UConn (ISU) [note: they also had a few solid wins against TCU, Tulsa, etc.]
Losses (3): Arkansas (TAM), Arizona State (Missouri), Georgia Tech (Kansas)
This is a trivial thing to get worked up about, but these rankings make a big difference in determining who plays in BCS games and the championship game. Problems here can have major consequences.
November 13th, 2011 at 10:41 PM ^
Seems a bit odd..Where is Jeff Sagarin from? Anyone know background on the guy?
November 13th, 2011 at 10:48 PM ^
why thank you sir
November 13th, 2011 at 10:47 PM ^
Calling them "computer" rankings is total bullshit. Comparing teams across different conferences is completely subjective and humans are the ones making the real judgments before/after data is imputed.
November 13th, 2011 at 11:25 PM ^
Explain yourself. Data goes in, rankings come out. Where is the bias?
November 14th, 2011 at 12:49 AM ^
love it! lol
November 14th, 2011 at 9:37 AM ^
I have tried to watch those shows a few times but every time they cut to this dude, it is sooo funny that I must change channels, he is not even close to being serious.
November 13th, 2011 at 10:54 PM ^
Traditionally the Pac 10 got this same boost.
November 13th, 2011 at 11:03 PM ^
In fairness to Sagarin, the BCS elects to use the politically correct ELO-CHESS version of Sagarin's ratings. In Sagarin's "pure" ratings, Oklahoma is 3 and OK State is 4, but Wisconsin is 7 and Michigan is 9. Missouri, TAMU, and UT are next at 10, 11, and 14, but then we get Nebraska at 17 and Sparty at 22. K State is 23 and Penn State is 25. Then we get Baylor at 29, but Ohio (33), Illinois (40), Iowa (45), and Northwestern (53) are all ahead of iowa State at 62. Purdue (66) is ahead of Texas Tech (68). Minnesota (101) is ahead of Kansas (106).
November 13th, 2011 at 11:19 PM ^
True, but that's not what's used in the BCS rankings. I'm not saying that this is all Sagarin's fault - just that the computer formulas in use seem wildly flawed right now.
November 14th, 2011 at 12:17 AM ^
That's because the BCS puts ridiculous stupulations on the computer algorithms, like not being able to use the margin of victory.
November 14th, 2011 at 12:40 AM ^
I agree that this is a big reason why these rankings are so terrible. If it were up to me, I'd call all wins by 20/25+ points equivalent to keep teams from having incentives to hang 100 on their small school neighbors, but there's no reason to throw away all margin of victory data.
Note, however, that even with margin of victory in there, the Sagarin rankings produce some ridiculous rankings. Missouri at #10 and Texas A&M at #11 with five losses each?
November 14th, 2011 at 6:08 AM ^
This is 100% the BCS' fault and not Sagarin's fault.
Sagarin is saying the B1G is very good, the BCS jumps in and says Sagarin can't use margin of victory or other stats...so Sagarin's other formular just craps out wrong for us this year. Every year one conference looks like its ranked too high in his BCS poll (ELO Chess), while Sagarin's real poll (Predictor) looks more accurate.
November 13th, 2011 at 11:05 PM ^
does this seriously have tx am as the #12 team in the nation ? there 5-5.......am I reading this wrong ?
November 13th, 2011 at 11:16 PM ^
That's exactly right. Among others, Sagarin has them ahead of Georgia, Virginia Tech, Houston, Boise, and every Big Ten team.
November 13th, 2011 at 11:25 PM ^
The fuck did Texas do?
November 14th, 2011 at 9:44 AM ^
Please continue to stay in character for all of your posts.
November 13th, 2011 at 11:32 PM ^
First, there is no proper way to rank 120 teams with such few data points that a college football season provides.
But these rankings make absolute sense, especially when there is no human judgment involved. The Big 12 plays nine conference games, meaning they have one less cupcake on average than your typical SEC, ACC, Big Ten, and Big East school. That's a big boost in the computers. Second, this is the collective list of their non-conference losses: Arkansas, Arizona State, and Georgia Tech. That's it! There's no New Mexico, FCS team, Rice, Army, and North Texas dragging the conference down. Kansas is the only tire fire, and even they managed to beat Northern Illinois. Add in their non-conference victories, which include: Miami, Florida State, Tulsa, Arizona, TCU, BYU, UCLA, Iowa, and UConn, and it's easy to see why the conference as a whole is highly rated. That's not a list of world beaters, but its a respectable slate of non-conference victims.
I believe the computers are over-rating the conference somewhat, but I do think from top to bottom they're the best and deepest conference in football this year.
November 13th, 2011 at 11:51 PM ^
I'm not arguing that the Big 12 is a weak conference this year. Find me anything in the coaches or Harris rankings, though, that's even close to this absurd:
Texas A&M is ranked 12th with these results...
beat Southern Methodist (home), Idaho (home), Texas Tech (road), Baylor (home), Iowa State (home)
lost to Oklahoma State (home), Arkansas (home), Missouri (home), Oklahoma (road), Kansas State (road)
This gives far too much credit for losing to highly rated teams.
[EDIT: Some of those losses were close, but remember that the BCS rankings do not incorporate margin of victory. Sagarin's rankings would be identical if all of those wins were close and losses were blowouts.]
November 14th, 2011 at 10:00 AM ^
Yes, it is a bit absurd. No question about it. But, according to Sagarin, the Big 12 has 8 out of the 10 toughest schedules. That leads to highly ranked teams.
Now, do I believe they really have 8 of the 10 toughest schedules? No, but it's close. It's all about that extra league game and a great non-conference performance. It also helps that the rest of the BCS leagues are garbage this year. Again, the Big 12 has 3 non-conference loses. The ACC has 13, the Big Ten has 13, the SEC has 5 (with four more match-ups with the ACC, of which they'll be lucky to win 2), the Pac 12 has 13 and the Big East has 13.
Lets be realistic, an average of 10 BCS opponents + not losing to anyone bad out of conference makes them all look good.
November 14th, 2011 at 9:09 AM ^
I guess an important question is - how many teams do you think would have a better record against that schedule? I'm not sure teams outside the top 20 should expect to break .500 there, or teams outside the top 10 to do better than 6-4. Yes, my gut instinct agrees with you that they're overrated at 12, but not by that much.
November 14th, 2011 at 1:11 AM ^
November 14th, 2011 at 1:28 AM ^
If Big 12 computer bias can prevent a LSU-Bama rematch in the MNC game, I am all for it.
November 14th, 2011 at 3:06 AM ^
how Baylor made it in last week when they had the same record as (ohio) and yet (ohio) had just won back to back games against ranked Illinois and Wisconsin. Not that I thought they deserved any credit, just that i was mad that beating ranked B1G was worth less to the BCS than beating unranked Mizzou and having Kansas take you to overtime.
November 14th, 2011 at 8:21 AM ^
I don't know why people are complaining about Sagarin when they should just be complaining about the BCS. The computers can only use games lost and games won as data points. MOV used to be included but then Miami made it into the title game with a weak schedule and giant MOVs. This is obviously an issue except when you consider they won the game that year. Algorithms are only as good as the data provided. Instead of providing more and more data each year, the BCS is limiting the data to a bare, minimum. That's why you see such shitty calculations being used for the BCS and comparable calculations with much more data look pretty accurate.
November 14th, 2011 at 11:50 AM ^
I agree with everything you said here, but there's still something crappy with Sagarin's ratings in particular. His preferred model (with POV) has 5-loss Missouri at #10 and 5-loss Texas A&M at #11, among other oddities. That puts them well ahead of #18 Arkansas (9-1), #21 South Carolina (8-2), #22 Michigan State (8-2), #24 Houston (10-0), etc.
November 14th, 2011 at 10:46 AM ^
Fucking Android App...
November 14th, 2011 at 8:53 AM ^
Those rankings seem way off, since Sagarin's actual rankings have more fewer Big 12 teams in the top 10 (I think a couple of Big 1G teams are in the mix as well, including UM with Sparty just on the outside).
The BCS is a system designed to generate money - I know, shocking observation on my part - and so common-sense rules do not weigh in. The fact that they impose a system wherein freaking Iowa State is higher ranked than anybody in the Big 1G outside of MSU is shocking.
November 14th, 2011 at 10:01 AM ^
First off, Sagarin gets little to no money for participating in the BCS, so I'm failing to see what he's gaining by not making the rankings to the best of his ability. Secondly, Iowa St is behind four Big Ten teams, not one. But basically you are right, the flaw that we see with Iowa St is not on Sagarin's behalf, its on the BCS'. Four of Iowa St's five wins have come by four points or fewer. Their four losses are by an average of 24 points. This is the exact reason we need to use margin of victory, and we can see by Sagarin's "Predictor" ranking that using such a formula puts Iowa St in a much more reasonable place.
November 14th, 2011 at 9:35 AM ^
We shouldn't throw stones. The B1G is terrible.
November 14th, 2011 at 10:27 AM ^
but using a poll system to determine anything is stupid. It's all based on perception and past games. Let's have a playoff, settle it on the field, and we won't have to worry about who's computer algorithm best determines how much better LSU and Alabama are than everyone else