If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(USA Today)   How much do the computers hate the Big East? In Jeff Sagarin's computer rankings this week, undefeated Louisville was 24th. That's lower than two 5-4 Pac 12 teams (BCS rankings are the red column)   (usatoday30.usatoday.com) divider line 35
    More: Fail, Jeff Sagarin, Big East, Louisville, NCAA Football, visiting team, Elo, BCS, hate  
•       •       •

1344 clicks; posted to Sports » on 05 Nov 2012 at 7:07 AM (2 years ago)   |  Favorite    |   share:  Share on Twitter share via Email Share on Facebook   more»



35 Comments   (+0 »)
   
View Voting Results: Smartest and Funniest

Archived thread
 
2012-11-05 07:12:41 AM  
Well, it can't be that big a deal, or else they'd force them to modify their computer models to make the rankings less accurate in a manner designed to prevent this from happening again.
 
2012-11-05 07:33:25 AM  
Numbers don't care.
 
2012-11-05 07:46:18 AM  
i bet there are a bunch of undefeated d-3 teams that are also considered less good than some 5-4 fbs teams...
 
2012-11-05 07:58:23 AM  
The rating are too busy over valuing Notre Dame.
 
2012-11-05 08:01:16 AM  
I humbly request that between Oregon, KState, and ND....two of them lose. Plskthx. Preferably to crappy teams since Georgia's loss is to a Top 10 team.
 
2012-11-05 08:03:30 AM  
Well, when your best game is against another Big East school, what do you expect?
 
2012-11-05 08:12:58 AM  
Those same 2 5-4 Pac 12 teams are also rated ahead of 8-1 FSU (8 spots behind Louisville), 8-1 Clemson (6 spots), and 7-2 Mississippi State.

For that matter, they are rated ahead of 6-3 USC and 7-2 UCLA, so it isn't even completely a conference favoritism thing. More a quirk of math this season.

The Big East is bad, but not because of Sagarin's ELO Chess numbers and not nearly as badly as it is being maligned by ESPN. In fact, Sagarin's regular ratings rank the conference ahead of the ACC and all non-major conferences.
 
2012-11-05 08:47:20 AM  
So, by the end of the season, Oregon is gonna be... what... 15th?
 
2012-11-05 09:07:54 AM  
Six Big East schools lost to MAC schools this year.
 
2012-11-05 09:25:55 AM  
But, what's not impressive about beating powerhouses such as Kentucky, Missouri State, North Carolina, Florida International, Southern Miss, Pittsburgh, South Florida, Cincinnati, and Temple? Two of those teams have winning records.
 
2012-11-05 09:48:56 AM  
I bet someone could come up with a really good WWF analogy to explain how the person that scripted this year's CFB season had a problem with the Big East and decided to bury them in the undercard.
 
2012-11-05 10:08:54 AM  

boyofd: Those same 2 5-4 Pac 12 teams are also rated ahead of 8-1 FSU (8 spots behind Louisville), 8-1 Clemson (6 spots), and 7-2 Mississippi State.

For that matter, they are rated ahead of 6-3 USC and 7-2 UCLA, so it isn't even completely a conference favoritism thing. More a quirk of math this season.

The Big East is bad, but not because of Sagarin's ELO Chess numbers and not nearly as badly as it is being maligned by ESPN. In fact, Sagarin's regular ratings rank the conference ahead of the ACC and all non-major conferences.


When West Virginia backed out of their game last-minute with FSU (and was replaced with Savannah State), it really crushed FSU's strength of schedule.
 
2012-11-05 10:12:54 AM  
Louisville should lose ground simply for taking the field against Kentucky. The Wildcats are godawful.

Also, birds don't have teeth, you cretins.
 
2012-11-05 10:18:58 AM  
Jacksonville is ranked 217th. Which is about as well as the other team in Jacksonville is doing.
 
2012-11-05 11:17:04 AM  
Record Versus good team - 0-0

Though, Ohio State's schedule is not much better.
 
2012-11-05 11:33:40 AM  

Guelph35: I bet someone could come up with a really good WWF analogy to explain how the person that scripted this year's CFB season had a problem with the Big East and decided to bury them in the undercard.


t0.gstatic.com

THE BIG EAST BELONGS ON THE IRRELEVANT TAB MARKS!

/go Pitt
//couldn't let WVU take home biggest choke of the week award
 
2012-11-05 11:33:57 AM  

Gonz: Louisville should lose ground simply for taking the field against Kentucky. The Wildcats are godawful.

Also, birds don't have teeth, you cretins.


They have no control over that game. It has to be played every year.

I'm with you on the birds.
 
2012-11-05 11:36:55 AM  

babysealclubber: Well, when your best game is against another Big East school, what do you expect?


Well... technically they beat an SEC team.
 
2012-11-05 11:39:14 AM  

nicoffeine: So, by the end of the season, Oregon is gonna be... what... 15th?


number 2, the bcs is design to reflect the human polls, but pretend it's not just the voters. every year that it doesn't they re-"adjustment" to make the polls more important. the value of the human polls is already at 2/3rds of the bcs total. which is scary when you think of the idiots that have votes in the two polls.

also, it would help the computers if they could take into account things like margin of victory. hell, a lot of the folks that do a bcs computer ranking have two rankings; one for the bcs and their bullshiat rules, and a second one which they say is more accurate, but the bcs won't let them lose.
 
2012-11-05 11:45:57 AM  

EyeballKid: But, what's not impressive about beating powerhouses such as Kentucky, Missouri State, North Carolina, Florida International, Southern Miss, Pittsburgh, South Florida, Cincinnati, and Temple? Two of those teams have winning records.


good point. i'm also currently undefeated, but alas i have played no one, why don't the computers respect my undefeated season? this is an outrage.
 
2012-11-05 12:05:03 PM  

A Fark Handle: also, it would help the computers if they could take into account things like margin of victory. hell, a lot of the folks that do a bcs computer ranking have two rankings; one for the bcs and their bullshiat rules, and a second one which they say is more accurate, but the bcs won't let them lose.


It would also help if they could adjust margin of victory with a cap of, say, 25 points (four possessions). Then you don't get extra credit for super-blowouts, but you still get credit for winning big. Also, if they took the score from when you took your starters out (I know Oregon's starters have been on the bench for about half this season - I think the USC game was the first time they played a full game).
 
2012-11-05 12:23:40 PM  

FLMountainMan: boyofd: Those same 2 5-4 Pac 12 teams are also rated ahead of 8-1 FSU (8 spots behind Louisville), 8-1 Clemson (6 spots), and 7-2 Mississippi State.

For that matter, they are rated ahead of 6-3 USC and 7-2 UCLA, so it isn't even completely a conference favoritism thing. More a quirk of math this season.

The Big East is bad, but not because of Sagarin's ELO Chess numbers and not nearly as badly as it is being maligned by ESPN. In fact, Sagarin's regular ratings rank the conference ahead of the ACC and all non-major conferences.

When West Virginia backed out of their game last-minute with FSU (and was replaced with Savannah State), it really crushed FSU's strength of schedule.


I hope you realize that UofL was supposed to play WVU this year as well (and we beat them last year at WVU). Also, TCU and Georgia were on the schedule but TCU joined the Big 12 (like they should have), and ESPN convinced Georgia to drop the contract with UofL in favor of BSU.
 
2012-11-05 01:09:51 PM  

boyofd: I hope you realize that UofL was supposed to play WVU this year as well (and we beat them last year at WVU). Also, TCU and Georgia were on the schedule but TCU joined the Big 12 (like they should have), and ESPN convinced Georgia to drop the contract with UofL in favor of BSU.


So what you are saying is, they wouldn't be unbeaten right now if not for all the scheduling changes?
 
2012-11-05 01:42:31 PM  

srhp29: So what you are saying is, they wouldn't be unbeaten right now if not for all the scheduling changes?


Also "whoa we could've totally beaten this team that's 2-3 in its conference! That would've bolstered our schedule a ton!"
 
2012-11-05 01:53:35 PM  
srhp29:

So what you are saying is, they wouldn't be unbeaten right now if not for all the scheduling changes?

That's not what I was saying, but I wouldn't argue with the point. UofL is not a great team this year. Should be next year, but almost no one is claiming otherwise this year.
 
2012-11-05 01:59:46 PM  

IAmRight: srhp29: So what you are saying is, they wouldn't be unbeaten right now if not for all the scheduling changes?

Also "whoa we could've totally beaten this team that's 2-3 in its conference! That would've bolstered our schedule a ton!"


Texas Tech and KSU have both used that argument this season, although the bloom may be gone from that rose. LSU received praise for beating them at WVU last season, when UofL did the same thing (by a smaller margin). But it's not like WVU was ranked as high as 5th this season or anything.
 
2012-11-05 02:38:06 PM  

iamskibibitz: babysealclubber: Well, when your best game is against another Big East school, what do you expect?

Well... technically they beat an SEC team.


Technically, yes. But that's the same team that lost to Western Kentucky at home and got beat 40-0 by Vandy, at home.
 
2012-11-05 03:40:48 PM  

A Fark Handle: nicoffeine: So, by the end of the season, Oregon is gonna be... what... 15th?

number 2, the bcs is design to reflect the human polls, but pretend it's not just the voters. every year that it doesn't they re-"adjustment" to make the polls more important. the value of the human polls is already at 2/3rds of the bcs total. which is scary when you think of the idiots that have votes in the two polls.

also, it would help the computers if they could take into account things like margin of victory.


Margin of victory means crap. Every time VA Tech runs up 35 points on Duke the media wets their pants at how awesome possum they are. They they run the table in the weak assed ACC and proceed to get waxed in a bowl game.

Margin of victory might be important against ranked teams. Run up 35 on Alabama and it matters.
 
2012-11-05 04:58:17 PM  

Rent Party: A Fark Handle: nicoffeine: So, by the end of the season, Oregon is gonna be... what... 15th?

number 2, the bcs is design to reflect the human polls, but pretend it's not just the voters. every year that it doesn't they re-"adjustment" to make the polls more important. the value of the human polls is already at 2/3rds of the bcs total. which is scary when you think of the idiots that have votes in the two polls.

also, it would help the computers if they could take into account things like margin of victory.

Margin of victory means crap. Every time VA Tech runs up 35 points on Duke the media wets their pants at how awesome possum they are. They they run the table in the weak assed ACC and proceed to get waxed in a bowl game.

Margin of victory might be important against ranked teams. Run up 35 on Alabama and it matters.


no, margin of victory does not mean crap. it's very important. that's why pretty much everyone who does computer modeling uses margin of victory with a cap on the margin (a lot use 21 points, and a cap still allows for "sportsmanship" concerns). beating a team by 3 touchdowns, no matter the team, is a lot more impressive than beating the same team by 1 point. there's a reason that the pythagorean expectation (essentially point differential) is better than simple wins and losses to predict future success. the margin of victory DOES mater! also, it's margin, not points score, so running up 35 on bama doesn't tell you anything about the margin of victory or defeat.
 
2012-11-05 06:39:44 PM  

A Fark Handle:
no, margin of victory does not mean crap. it's very important. that's why pretty much everyone who does computer modeling uses margin of victory with a cap on the margin (a lot use 21 points, and a cap still allows for "sportsmanship" concerns). beating a team by 3 touchdowns, no matter the team, is a lot more impressive than beating the same team by 1 point. there's a reason that the pythagorean expectation (essentially point differential) is better than simple wins and losses to predict future success. the margin of victory DOES mater! also, it's margin, not points score, so running up 35 on bama doesn't tell you anything about the margin of victory or defeat.


Horseshiat. It is exactly this lame assed (and wrong) thinking that has weak assed ACC teams highly ranked every year, and then get spanked when they have to play out of division foes that aren't St. Mary's School for Blind Girls. Margin of victory over ranked teams should matter. But I'll take ND's tough seven point win over Stanford or whoever wins the next 10-7 LSU/Alabama tilt as way more significant than Florida State's combined 180-3 wins over Murray State, Savannah State, and Wake Forrest.

They *should* win those games by 60 points. They shouldn't get credit for doing it.
 
2012-11-05 07:15:28 PM  

Rent Party: They *should* win those games by 60 points. They shouldn't get credit for doing it.


And what happens if they only pull out narrow wins against those crappy teams? Are you just gonna shrug and say "well, they beat the teams they were supposed to beat"? Margins matter regardless of the matchup. An ideal predictive ranking will take into account point margins and opponent quality, as Sagarin's PREDICTOR and ELO_SCORE (not to be confused with the ELO_CHESS algorithm he uses to comply with BCS rules) presumably do.
 
2012-11-05 10:02:27 PM  

Rent Party: A Fark Handle:
no, margin of victory does not mean crap. it's very important. that's why pretty much everyone who does computer modeling uses margin of victory with a cap on the margin (a lot use 21 points, and a cap still allows for "sportsmanship" concerns). beating a team by 3 touchdowns, no matter the team, is a lot more impressive than beating the same team by 1 point. there's a reason that the pythagorean expectation (essentially point differential) is better than simple wins and losses to predict future success. the margin of victory DOES mater! also, it's margin, not points score, so running up 35 on bama doesn't tell you anything about the margin of victory or defeat.

Horseshiat. It is exactly this lame assed (and wrong) thinking that has weak assed ACC teams highly ranked every year, and then get spanked when they have to play out of division foes that aren't St. Mary's School for Blind Girls. Margin of victory over ranked teams should matter. But I'll take ND's tough seven point win over Stanford or whoever wins the next 10-7 LSU/Alabama tilt as way more significant than Florida State's combined 180-3 wins over Murray State, Savannah State, and Wake Forrest.

They *should* win those games by 60 points. They shouldn't get credit for doing it.


facts do have a "a fark handle" bias. unfortunately, the fact is that margin of victory does have a meaningful predictive ability. sure margin of victory and strength of opponent might need an interactive term in a model, turns out the computer rankers know this. to pretend that beating a team by 1 and beating the SAME FARKING EXACT team by 21 should be equally meaningful is lead paint chewing tarded. i'm not saying margin of victory regardless of opponent is that meaningful, but opponent adjusted margin of victory is informative.
 
2012-11-05 11:26:28 PM  
I'd actually like to suggest, on the topic of computer rankings, that somebody compute a similar statistic that uses the win probability statistic. It would be a more refined version of margin of victory. Something like qualifying wins with the amount of time you spend over a certain threshold for victory. It rewards dominating performances where one team clearly shows it is better than its opponent, while not rewarding running up the score (scoring an extra touchdown when you're up 35-3 doesn't really increase your probability much at all). Using this stat, you could come up with a ranking set that shows, roughly, which teams are truly dominant and which teams are truly cupcakes. If you then used that ranking to adjust the value of the wins that each team had (so that beating up on cupcakes doesn't get you a whole lot of points), you could have a fairly objective strength of schedule stat that shows just how impressive each win is. From there, you could add up the wins/losses and have a decent measure of how impressive a team's body of work is over a season.
 
2012-11-06 01:53:33 AM  
24? What I am reading has Louisville at 44. North Carolina is 34.....Louisville beat them. That makes a lot of sense
 
2012-11-06 05:18:56 AM  
Louisville alum. The team is scrappy. If we get by Rutgers, I'm excited to play the Orange Bowl. Even if Alabama is the only other team to run the table, I am under no delusion that we should play for the championship.

By the time he's a senior, though, Bridgewater will be a Heisman candidate. Excited for the recruiting Charlie is doing.
 
Displayed 35 of 35 comments

View Voting Results: Smartest and Funniest


This thread is archived, and closed to new comments.

Continue Farking
Submit a Link »
On Twitter





In Other Media


  1. Links are submitted by members of the Fark community.

  2. When community members submit a link, they also write a custom headline for the story.

  3. Other Farkers comment on the links. This is the number of comments. Click here to read them.

  4. Click here to submit a link.

Report