Writers of Pro Football Prospectus 2008

Most Recent FO Features


» 2018 Free Agency Cost-Benefit Analysis

Is Kirk Cousins the best free-agent quarterback in recent memory? Should Trumaine Johnson or Malcolm Butler have gotten the larger contract? And what makes a free-agent contract good or bad, anyway?

13 Nov 2006

Confessions of a Football Junkie: Upstream Red Team

by Russell Levine

It took a magical night in the birthplace of college football to get Rutgers into the national championship debate.

Barely 48 hours later -- not even long enough for the tidal wave of Rutgers calls to New York's sports-talk radio stations to subside -- a series of upsets had the potential to turn Rutgers' nice little moment in the spotlight into something much greater.

Even the most euphoric of Rutgers fans following their school's stunning, come-from-behind win over Louisville Thursday night knew that even finishing the regular season undefeated probably wouldn't be enough to get the Scarlet Knights into the national championship game. But that was before a string of upsets decimated the ranks of the once-beaten Saturday. By day's end, Rutgers's chances had improved from nearly impossible to merely improbable.

With Auburn, Cal, and Texas all picking up a second loss, the number of teams standing between Rutgers and the coveted second spot in the Bowl Championship Series standings has been thinned. And it no longer takes an abacus and vivid imagination to cook up a scenario that could see Rutgers rise that high. Sunday's BCS standings release showed Rutgers at No. 6, within striking distance of the top.

What's more, several of the teams in the top 10 have yet to play one another, beginning with next Saturday's epic 1 vs. 2 showdown between Ohio State and Michigan. How, exactly, could the Scarlet Knights get to the January 8 title game in Glendale, Arizona? Call the following the Six Degrees of Rutgers:

1. Ohio State beats Michigan
2. Cal beats USC
3. LSU beats Arkansas
4. Arkansas beats Florida in the SEC championship
5. USC beats Notre Dame
6. Rutgers wins out

None of the first five steps will matter if Rutgers doesn't complete the final one, beginning with a classic trap game this weekend at Cincinnati and concluding with a road trip to West Virginia on December 2. But if the Scarlet Knights bring the same defensive intensity to their final three games that they showed in completely dominating one of the most explosive offenses in the nation for the final three quarters Thursday, they stand an excellent chance to finish the task.

While Rutgers certainly benefited from Saturday's results, it wasn't the only school smiling at the sight of the out-of-town scoreboard. The day's biggest winner was USC, which jumped to no. 3 in the BCS standings, putting the Trojans in position to play the Ohio State-Michigan winner in the championship game. USC easily dispatched Oregon Saturday, and with home contests against Cal and Notre Dame the next two weeks, its strength of schedule -- a key component in the computer rankings -- will remain high. And USC not only got a boost from other teams losing, it also got a lift from Arkansas, which routed Tennessee Saturday night.

Every Arkansas win is good news for USC, since the Trojans are the proud owners of a 50-14 thumping of the Razorbacks on the season's opening Saturday. That win, which took place on Arkansas's home field, grows in importance each week that the Razorbacks remain unscathed in the SEC. Arkansas may turn out to be the best team in the nation's toughest conference, but as long as it and USC have the same record, that September result should tilt the scales in favor of the Trojans.

All this debate is over the second spot in the BCS title game. The first spot will be claimed by Saturday's winner in Columbus after Michigan and Ohio State each sidestepped the final roadblocks to their first-ever meeting as the top-ranked teams and their first meeting as unbeatens since 1973. Just as Rutgers's chances of reaching the championship match got a boost, so to did the possibility that the Wolverines and Buckeyes could play a rematch in Glendale.

A rematch can probably only happen if Ohio State wins a close game, and everyone else loses. If Michigan wins, or if Ohio State wins by a wide margin, the loser is likely to fall far enough in the human polls to negate any possibility of a return date. But if visiting Michigan loses a tight game as the underdog, it's entirely possible the Wolverines could hold the second spot in the BCS standings next Sunday. But even that won't guarantee a rematch. The poll voters would still have two weekends of football to effectively vote to nullify such a scenario by jumping another one-loss team over Michigan.

No matter what happens to Rutgers the rest of the way, the fact that this once-woeful program is even being discussed in such lofty circles is a sure sign of its arrival. In fact, Rutgers had validated its rebuilding project even before kicker Jeremy Ito nailed the game-winner against Louisville, simply by virtue of its comeback against the Cardinals.

Few gave Rutgers a chance in that game, fewer still would have offered the Scarlet Knights and their ground-based offense any hope of rallying from an 18-point deficit caused in part by three killer mistakes committed by the home side. That Rutgers could overcome a kick return for a touchdown, a successful Louisville fake punt, and a critical roughing-the-kicker penalty that extended another Louisville drive was the most stunning development in Thursday's game. Not only did Rutgers win, they did it in the most improbable fashion.

To this point, this college season has had something for everyone. For traditionalists, there is the prospect of two of the most storied rivals in all of sport -- Michigan and Ohio State -- meeting with more at stake than in any of their 102 previous renewals. College football's popularity is built on tradition, something that is ever more challenged in the era of BCS standings, talent-distributing scholarship limits, and weekday night TV games. But the flipside of changing traditions are stories like those taking place at Rutgers and at Wake Forest, where the 9-1 Demon Deacons are poised to reach the ACC title game. In any other year, Wake would likely be the story of this college season, but the Deacons have had to take a back seat in the Cinderella department to Rutgers.

With three weekends of play remaining in the regular season, the year's storylines are not yet complete. And for Rutgers and its fans, a year that has already included a signature moment to last a lifetime stands to get even better.

John L. Smith Trophy

The JLS Trophy was born to reward wacky coaching decisions, but somewhere along the line, probably around the time JLS conducted that epic halftime interview during last year's Ohio State meltdown, it became an award for the worst coaching decision of the week.

I'm trying to get back to award's origins this week by giving it to Kansas State coach Ron Prince, even though his team just pulled off a stunning upset of Texas. So what could Prince have done to merit JLS consideration, you ask?

Check the play-by-play logs. Leading, 45-42, Prince's team faced a third-and-6 at the Texas 38, with the Longhorns out of timeouts. Simple call, right? You run up the middle, wait for the septuagenarian ref to take 15 seconds to spot the ball, and then begin celebrating, right?

Not Prince. He called a pass play, which quarterback Josh Freeman completed for the first down. Let's say that pass had fallen incomplete, leaving 48 seconds on the clock. Kansas State would have been punting back to Texas and most likely leaving the Longhorns something like 35-40 seconds with which to work, needing only a field goal to tie.

Congratulations, Ron Prince on the big win. And congrats, too, on your JLS Trophy. (HT: Brian Cook)

BlogPoll Ballot

This season, I'll again be voting in the BlogPoll, hosted by MGoBlog. I'll post my ballot in Junkie each week. Feel free to comment -- my rankings may change based upon your suggestions.

Further explanation required for the following:

  • With Florida looking mortal and Texas, Louisville, Cal and Auburn all out of the championship picture, USC gets a big boost to the third spot. A bid in the title game is the Trojans' to lose at this point, and they deserve it. That's what they get for a non-conference schedule of at Arkansas, Nebraska, and Notre Dame. You think Florida's going to regret scheduling that Western Carolina game this week?
  • I'm now firmly convinced that Arkansas is the best team in an SEC that is deep, but lacks any dominant teams. Ask yourself, would Arkansas, Florida, Tennessee, or Auburn go undefeated in the Pac-10, Big-10, or Big-12? I don't think so. I think you'd have a bunch of one- or two-loss teams, just as you do now.
  • I feel better about Rutgers at No. 6 than I do about Wake Forest at No. 8, mostly because I think the Big East is better than the ACC this season.
  • The rest, quite frankly, is a crapshoot. I'm open to suggestions.
Rank Team Delta
1 Ohio State --
2 Michigan --
3 Southern Cal 5
4 Arkansas 3
5 Florida 1
6 Rutgers 8
7 Notre Dame 4
8 Wake Forest 7
9 Louisville 5
10 West Virginia --
11 LSU 1
12 Texas 9
13 Wisconsin 4
14 Auburn 9
15 Boston College 3
16 Oklahoma 4
17 Boise State 1
18 Tennessee 5
19 California 10
20 Maryland 3
21 Georgia Tech --
22 Virginia Tech --
23 Hawaii 2
24 Brigham Young 2
25 Nebraska 1

Dropped Out: Oregon (#19), Texas A&M (#24).

Portions of this article appeared in Monday's New York Sun.

Posted by: Russell Levine on 13 Nov 2006

61 comments, Last at 16 Nov 2006, 4:19pm by MRH


by Brad (not verified) :: Mon, 11/13/2006 - 7:29pm

I'd have LSU somewhere in my top 10 simply because I think they'd have a great chance of beating any team on a neutral field, depite their 2 losses.

I didn't think SOS was included in the BCS rankings anymore?

by Tim Gerheim :: Mon, 11/13/2006 - 7:30pm

It kills me, absolutely kills me, that USC bumps ahead of Florida. I think the only reason that happens is because they're jumping all the other teams between them that lost, and then let's go one more just for fun.

(This is not peculiar to Russell's ballot, I'm mostly annoyed with the BCS voters. Also in the interest of full disclosure, USC, and the Pac-10 in general, annoys me and I think they've been overrated for as long as I've paid much attention to college football.)

If Texas and Auburn win, what happens? Does USC jump all them because of their first impressive win since September? Does Florida fall behind all three for BEATING South Carolina, but not by much? If not, then it's totally inappropriate for USC to jump Florida in the current scenario. Even leaving in the Auburn loss since it affects Florida's schedule, Florida shouldn't drop behind Texas and USC shouldn't jump ahead of Texas, so the absence of Texas in between shouldn't change anything.

Non-conference schedule is nice, and it may be legitimate for USC to be ahead of Florida at the end of the season if they beat Cal and Notre Dame and Florida just beats Directional Carolina and Florida State, but not now. Florida's wins over serious SEC football teams are immensely more impressive than USC's wins (which by and large are closer) over the glorified WAC.

The only thing that makes me sadder than seeing both of ESPN's bowl projectors put USC in the championship game is seeing them both put Texas against Boise State in the Fiesta Bowl.

by Fourth (not verified) :: Mon, 11/13/2006 - 7:33pm

Tim is my hero.

by Fourth (not verified) :: Mon, 11/13/2006 - 7:41pm

Florida is virtually tied with USC in the human polls, but took a tumble this week in the computers. In 3rd a couple weeks ago, they're now 6th in comp ranks....BEHIND Notre Dame. What do the computers like so much about service academies?

by Larry (not verified) :: Mon, 11/13/2006 - 7:42pm

Can't argue too much with your KSU JLS award, but I'd like to nominate Spurrier for letting too much time run off the clock before getting into

by Larry (not verified) :: Mon, 11/13/2006 - 7:44pm

STUPID, STUPID less-than sign!!

Can't argue too much with your KSU JLS award, but I'd like to nominate Spurrier for letting too much time run off the clock before getting into less than 40 yard range for his kicker. And if you want truly bizarre, he took a TO with 8 seconds left for the kick on 4th down. That is, he'd have had to Kick Off if he made it (though he could have Bielema'd it, i guess) and he gained nothing by leaving 8 seconds since it was 4th down anyway.

That said, time to talk about Rutgers chances to make the title game if they win out. I don't think it is as improbable as Russel says. Since one of USC/ND, ARK/UF, OSU/UM all will lose, and WVU as well if RU wins out, I figure they gain 2.5 - 3.5 spots in the polls (they're inexplicably behind WVU in the Coaches poll right now). That's 2/3*(0.1 to 0.14) BCS points. Rutgers is going to maintain #2 in the computers almost for sure (USC is the only possibility to move ahead I'd think, but I doubt it). They trail USC by 0.08 or so right now. That makes it really, really close with no particular results needed. That is, unless the voters leave RU behind the (2-loss) losers above. Even stuck behind the OSU/UM loser in the polls, I think RU would be very, very close. Anybody else see this differently? The only thing I see I might be getting wrong is is USC's schedule down the stretch is tough enough for them to move ahead of RU in (some of) the computer models.

Speaking of that, can we stop calling them computer "polls"? They're models, or ratings, maybe, but not "polls". Well, except for Billingsly, which basically tries to be a poll it seems.

by Mnatr (not verified) :: Mon, 11/13/2006 - 8:08pm

If it makes Tim feel any better, lots of teams should have been ahead of Florida, and so they never should have been "jumped."

by Mike (not verified) :: Mon, 11/13/2006 - 8:11pm

I think Rutgers can drop in the comp polls if OSU wins, Larry, I could see OSU moving up from 3rd without Michigan dropping behind Rutgers pretty easily in such a scenario.

by Mnatr (not verified) :: Mon, 11/13/2006 - 8:15pm

And yes, stop calling them computer polls. I'd prefer it if they stopped using the words computer and poll. Yes, computers compute the ratings much more quickly than us humans, but it's not like it's Big Blue or WHOPPER sitting in a room designing its own ratings.

Of course, the only other way to put it is "Objective Ratings," but that doesn't really tickle my fancy.

by Matu (not verified) :: Mon, 11/13/2006 - 8:25pm

Sorry Tim, but the Pac-10 isn't as weak as you think. As Russell pointed out, the Pac-10 along with the Big Ten and some other conferences have been on par with the SEC. I can't speak for voters in the human polls but, to me, Florida dropping behind U.S.C. is justified not just because of the South Carolina game but because of the way Florida has been playing over the last month. They have been very unimpressive and actually look worse now than when U.S.C. was struggling. I also feel that there schedule has been a little weaker to this point then U.S.C. and U.S.C.'s best win (Arkansas) looks better to this point then Florida's best (L.S.U). I don't think Florida has anything on U.S.C. right now and won't if U.S.C. wins out.

by Larry (not verified) :: Mon, 11/13/2006 - 8:27pm

Good point, Mike. That is certainly possible. Of course it is also possible (though pretty unlikely) Rutgers could end up #1 in the computers in such a scenario (and end the computers' participation the BCS forever?) If ND loses to USC, then that will diminish the meaning of UMs win in South Bend, which is likely to be the major thing pushing the UM/OSU pair above everyone else. I also wonder how much the next 2 games for RU against the bottom of the Big East will hurt their position in the computer models.

Perhaps it is too early for such speculation, but the number of scenarios among the top teams is dwindling to a countable number, and who doesn't love trying to imagine what it will take to break the BCS standings.

by Tarrant (not verified) :: Mon, 11/13/2006 - 8:38pm

Serious SEC football teams.

Would that be the serious Georgia, last year's champion, that lost in the Sugar Bowl to overrated and most certainly not serious West Virginia, and was beaten by serious Kentucky recently?

Or the serious Auburn, that was just blown out of the water by serious Georgia, that lost to serious Kentucky?

Or the serious Arkansas, who, well, we can't use them as an example - they obviously suck and should have no place in the championship picture, since highly overrated USC beat them (and yes, Arkansas then is not Arkansas now, but that doesn't change the score).

Or perhaps the serious South Carolina (maybe that's the USC you're thinking is actually overrated?), home of the Ol' Ball Coach, one of the most well-known coaches in SEC history, who lost last weekend over serious Florida because they had three kicks blocked.

Or serious LSU, a school that admittedly does have the most recent SEC national title (a split title with overrated and decidedly non-serious USC in 2003). They managed all of three points against serious Auburn, which managed to give up 35 points to serious Georgia, and 27 points to serious (but out of the picture because they suck) Arkansas.

You're absolutely right that had Texas, serious Auburn, and Louisville (I note you left them out) not lost, then USC wouldn't be #3 in the BCS Standings. But those teams did lose, and it's kind of hypocritical to say "Does Florida move down for BEATING South Carolina, albeit not by much" and at the same time earlier in the post say that USC should be lower because, pre-Oregon State, they weren't winning by much either.

No one seemed to complain earlier in the year as USC was passed up over and over again in the polls for winning "ugly" by other teams that won "big".

Both teams have a loss. The polls make it a virtual tie at the moment. The computers give USC a slight edge, which at the moment is likely to increase. And both teams have multiple games to go - and only after those games are played, and we see whether overrated USC, serious Florida, or both, have won out, is it time for bitching by either party.

by Larry (not verified) :: Mon, 11/13/2006 - 8:53pm

Experimenting with Colley's "add games" feature, it seems that, no, the OSU/Mich loser is not going to stay above Rutgers. USC, however, if they win out, looks very good to move ahead of Rutgers (and an undefeated OSU). Hard to be certain, you can only "add" 5 results, so I didn't get to look at whether what Arkansas does makes a difference. USC is going to end up with one monster schedule and every other title contender should probably feel shame just thinking about it.

by Tim Gerheim :: Mon, 11/13/2006 - 8:55pm

I left Louisville out because they were ahead of Florida in the BCS in the first place, so I wouldn't have wanted anybody to jump them if they won, so they weren't really relevant to what I was talking about.

The rest is absolutely valid criticism, and while I think my argument stands up, I don't want to get into a pissing contest over this. I mostly just wanted to get that off my chest from the moment I saw the BCS rankings this morning.

by Will Allen (not verified) :: Mon, 11/13/2006 - 9:25pm

To me, the nadir of Pac 10 football came seven years ago, when a soft, soft, soft, UCLA team made Ron Dayne look like Jim Brown in the Rose Bowl. Since then, I think the conference has made some big strides at the top the conference, obviously greatly aided by the re-emergence of USC, which I think is spurring the other programs on as well. It no longer appears to be the glorified flag football league among BCS conferences.

I hope the Big Ten isn't reverting back to small elite atop an otherwise entirely mediocre (or worse) balance of the league. I think the new coach in Wisconsin is the real deal, however, and Ferentz is going to have more success than he had this year. If the folks in Happy Valley can manage the inevitable succession well, then Tiller in Purdue and the occasional decent year by Minnesota should keep things in good shape. Heck, the Zooker may be able to make Illinois respectable again.

Ideally, as a Big Ten fan, I'd love to see a playoff system force Notre Dame into a conference, and hopefully they would go with their traditonal rivalries, resulting in the conference name being off by two.

by blacksuit (not verified) :: Mon, 11/13/2006 - 10:06pm

Arkansas can beat anybody right now. USC ran up the score in that first game, and they were playing without Darren McFadden and a decent quarterback. Yeah, the game still counts, but teams change over the course of a year. If I had to guess, we'll end the season with the hogs as a one loss team, and USC as a more than one loss team.

by BillWallace (not verified) :: Mon, 11/13/2006 - 10:13pm

I really don't follow college football, but why on earth would either Ohio State or Michigan not stay at #2 unless they lost badly?

What argument would put another one loss team that didn't lose to the #1 ahead of one that did, other than 'because it was like 2 months ago so we can't remember it very well'.

by Will Allen (not verified) :: Mon, 11/13/2006 - 10:24pm

I agree, Bill, and it is the outcome I'm hoping for, given I'm a Big Ten fan who hates the BCS system. Ohio State coming from behind to win on a last play touchdown would be ideal, in that it would nearly gurantee a rematch, while more than a couple one loss teams (or even an undefeated team) are shut out of the championship. The BCS flaws would be even more dramatically highlighted.

by Bill (not verified) :: Mon, 11/13/2006 - 11:06pm

One of the interesting thing is that the human polls allow for flexibility of a spot - if the #4 team is only two votes ahead of the #5 team, it's reflected in the BCS bowl percentages. However, in the objective/computer/statistical ratings, a team could have twice the rating of the team below it, and it counts the same in the BCS standings as a team just one one-thousandth of a point lower.

by Zac (not verified) :: Mon, 11/13/2006 - 11:16pm

Sadly, BillWallace, that's basically it. Teams that lose early in the season have an advantage over teams that lose late.

by Jesse (not verified) :: Mon, 11/13/2006 - 11:16pm

Imagine if OU hadn't gotten screwed against Oregon. They'd be 9-1, and genuinely in the hunt for a spot in the National Title game. Instead they're moving up 1, maybe 2 spots a week, despite winning everything but their Texas game. I'm just sayin

Really nice to see you keep Florida where they are, instead of skyrocketing them like everyone else did, even though they were outplayed.

Wake Forest in the top ten is absurd. They've played two quality opponents, and lost to one of them.

Tennessee is kinda low...

by Zac (not verified) :: Mon, 11/13/2006 - 11:40pm

Sorry for the double, Bill's last post wasn't there when I wrote.

I believe you're saying that the rankings make the difference between each spot seem the same, when really the difference could be much larger or smaller depending on circumstances.

The problem is that each computer rankings has its own rating system.

The ELO CHESS version of Sagarin's ratings (which the BCS uses) has Michigan in 1st place with 106.18, and Florida International last (among I-A teams) with a rating of 37.54.

The Anderson ratings have Michigan in first place with a rating of .833, and Florida International last with a rating of .178.

The Billingsley ratings have Ohio State in first place with a rating of 318.964, and Duke in last place with a rating of 150.288.

Colley's ratings have Michigan in first place with a rating of 0.980314, and Memphis in last with a rating of 0.041638.

Massey's ratings have Michigan in first place with a rating of 2.675, and has Florida International last (among I-A teams) with a rating of .319.

The Wolfe ratings have Michigan in first place with a rating of 8.776, and Duke in last with a rating of .632.

So you see the problem with the numbers. With the human polls, they divide by the highest score possible, and turn it into a percentage. What is the highest score possible in these computer ratings? I don't know. Is it even possible to figure out? Plus, they like to throw out the high and the low ranking. If each computer rating had a maximum value, then they could still do % of maximum like they do the human polls, but it would be much more difficult.

by Bill (not verified) :: Tue, 11/14/2006 - 12:17am

22 - True. I suppose you could scale the #1 team as 1.000, and run every other team as a percentage of that. Of course, that may not change the ratings at all.

by Larry (not verified) :: Tue, 11/14/2006 - 1:37am

Re: 22/23
The ratings aren't necessarily linear. So, it isn't obvious that (rating)/(#1 rating) even makes any sense. Clearly there ought to be some way to capture finer gradations in the computers, but until the computers tell you what those numbers actually mean there isn't much you can say.

This is also a nice place to complain about the computer models being proprietary. There's no way that should be allowed. The formulas all ought to be public domain. I was listening to Artie Spanier on Sporting News Radio tonight ask Jerry Palm some of the dumbest questions ever, like "OSU is #6 in Massey, do you think he made a mistake?" and "No, well then he must be putting some bias in there against OSU." Arrrrgh... If the computers formulae were public, then at least you could figure out why they give the results they do.

by Peter Libero (not verified) :: Tue, 11/14/2006 - 1:46am

Is that REALLY the worst coaching decision of the week? I find it almost impossible that running an aggressive play that worked and won the game was the best thing out there.

Anyway, I think if Rutgers wins out they should be in the title game. There will be no other undefeated teams, and they will have faced two #3 teams and defeated both. Their schedule won't be that bad; I have no idea, besides human prejudice, what would make them inferior to any one-loss team. There's a reason the computers love them, and they'll love them even more if they were to beat WVU. I think the poll voters would be eager to shoot them up the ranks as well.

by kibbles (not verified) :: Tue, 11/14/2006 - 5:51am

Really nice to see you keep Florida where they are, instead of skyrocketing them like everyone else did, even though they were outplayed.
I wouldn't say Florida was outplayed. I strongly suspect that if we had VOA for college football, Florida would have posted a significantly higher ranking than South Carolina in that game. Discounting the kneel-down at the end, Florida had 7 drives. Those drives went for 77, 53, 79, 54, 1, 74, and 80 yards. Those drives took 12, 9, 7, 9, 5, 11, and 11 plays. The reason they didn't blow SC out is because they missed a 30-yard FG, they had a holding penalty on 1st-and-10 from the 11, had a 16-yard loss on 3rd-and-3 from the USC 38, had a 15-yard intentional grounding penalty on a QB-throwback (bad call, since Leak was in the area), and a horrible throw on 3rd-and-2 from the 6- generally, those events are pretty non-predictive.

I'm not saying that South Carolina didn't deserve to win that game- they absolutely did deserve to win that game, and Florida was really lucky to walk out with a victory there. Still, I was at the game, and it was clear that Florida was pushing USC around all day, and then just repeatedly shooting itself in the foot in the red zone.

Also, I have to say it... I have *NEVER* heard any stadium as loud as when USC was driving at the end after UF took the lead- and I've been to a lot of SEC stadiums, so I'm familiar with rowdy. Of the 90,000 people in attendence, I suspect at least 70,000 were still standing on their seats screaming their heads off 30 minutes after the game ended. Seriously, a full half hour after the block, I couldn't see a single seat in the entire stadium through the sea of people (well, except in the South Carolina section, of course). I have a Chinese friend who had never seen a football game in his life before Saturday. What a game for an initiation to American Football.

by mactbone (not verified) :: Tue, 11/14/2006 - 10:13am

Re SC:
I still can't get over the passion those guys have for their team. South Carolina isn't a big place and I can't remember them having a good year, so it's surprising to me that they pack that place and root so hard. Talk about a rivalry too... people talk about a lot of rivalries, but USC-Clemson is nasty. I've heard some stories that make the Big Ten rivalries very tame.

by jebmak (not verified) :: Tue, 11/14/2006 - 10:25am

I figured that Rutgers has a better shot if Mich beat OSU, because the game is at OSU, and they would be looked upon more poorly if they lost.

by Pat (not verified) :: Tue, 11/14/2006 - 10:25am

What is the highest score possible in these computer ratings? I don’t know. Is it even possible to figure out?

Most don't - they're unbounded up (and down). Elo rankings specifically aren't. One downside to trying to combine the statistical rankings is that you'd have to figure out a way to combine the scores, which isn't exactly obvious. To be honest, this is probably why they do that (they do ask statisticians about stuff like this).

You can't just do something like "divide by the highest score" to normalize, because then the other teams shift around by the shifting of the quality of the top team. It's a moving target. You'd need to normalize to "score given to an average team", something fixed and independent.

However, that still doesn't make rankings comparable: Colley's "average team" would be an average Division IA team. Sagarin's "average team" would be the average team in the combined Division IA/IAA. It's not an easy problem.

by Not saying (not verified) :: Tue, 11/14/2006 - 10:28am

Re: 25 "Is that REALLY the worst coaching decision of the week?"

No. "The JLS Trophy was born to reward wacky coaching decisions, but somewhere along the line . . . it became an award for the worst coaching decision of the week.
"I’m trying to get back to award’s origins this week by giving it to Kansas State coach Ron Prince."

It was specifically mentioned as not the worst decision, just a wacky one. Which it was, since just running it would have won the game as well.

by Pat (not verified) :: Tue, 11/14/2006 - 10:39am

This is also a nice place to complain about the computer models being proprietary. There’s no way that should be allowed. The formulas all ought to be public domain. I was listening to Artie Spanier on Sporting News Radio tonight ask Jerry Palm some of the dumbest questions ever, like “OSU is #6 in Massey, do you think he made a mistake?� and “No, well then he must be putting some bias in there against OSU.� Arrrrgh… If the computers formulae were public, then at least you could figure out why they give the results they do.

Massey's (BCS) rating isn't *that* proprietary. There's enough description on the Theory page to reconstruct it if you felt like it. It'd just take a lot of effort and time. The reason OSU is #6, for instance, is because Massey time-weights the games (which is stupid, you've only got 12 data points, throwing some of them away is absolutely not going to help you) and it's been forever since Ohio State played even a reasonably non-sucky team.

Billingsley and Anderson-Hester's are the really "you suck, document your system" ones. Colley's and Wolfe's are explicitly explained, and Massey's is only a little proprietary. Sagarin's probably could be explained more, but it is just an Elo-type rating.

by bowman (not verified) :: Tue, 11/14/2006 - 10:39am

10. Florida was better in their loss (Auburn) than USC was in their loss (Oregon). As losses are more important in NCAA than wins, I would think that this plays an effect.

As a (biased) football fan, I'd rather see the team that lost to another good team on the road be rewarded over the team that lost to an average team on the road.

Of course, this assumes that USC can beet Cal and ND...

by snik75 (not verified) :: Tue, 11/14/2006 - 11:07am

SO for my two cents...the current system does NOT reliably allow us to determine a "national champion" in any sort of meaningful way. The only teams that realistically have a shot to compete for the pseudo-national championship are "name" schools that play and beat other "name" schools. If Rutgers goes undefeated and isn't allowed into the "title game" then the team and it's fans will feel like they got screwed, after the best season in school history!!! All they should be doing at that point is celebrating. Lets forget about crowning a national championship in division 1 football and just enjoy the season, hope our team goes to a good bowl and doesn't embarrass themselves.

Or, if you MUST have a champion... divide up the division into smaller subsets, so that most competing teams can play each other or at least several common opponents. But why? We don't fret about crowning a high school football "national champion".

by hrudey (not verified) :: Tue, 11/14/2006 - 11:55am

Shoot, I'm as orange and blue homer as you can be, but I don't think Florida should be #3 right now. Of course, I don't think USC should either. USC has one win over a top-20 BCS team right now, that opener against Arkansas, and one loss to a team outside the top 25. Florida has one win over a top-20 BCS team, and one loss to a top-20 BCS team. Freaking Rutgers has one win over a top-20 BCS team and zero losses. Of course, if they lose to WV, all bets are off, but right now, they deserve the #3 rating more than any of the other teams. It's not like Boise State, who haven't beaten anyone even remotely decent -- Rutgers has a quality win and zero losses. If they win out, they deserve the right to be overmatched by the tOSU-Michigan winner.

by Dervin (not verified) :: Tue, 11/14/2006 - 12:08pm


the reason why sh*tty southern football teams can sell out 80,000 seat stadiums is because there is nothing else to do there. Other than hunting
and Gun Shows

by Tarrant (not verified) :: Tue, 11/14/2006 - 12:12pm

If Ohio State loses, the odds of a rematch are practically zero. The computers already have Ohio State #3, behind Rutgers - and as was pointed out earlier in the thread, should USC and OSU win out, they'd actually overtake OSU in some computers despite USC's loss. OSU cannot recover from a home loss, regardless of the margin - the polls may not punish them much, but the computers will kill their chances if they don't beat Michigan.

Michigan, however, is a (after dropping the high/low) unanimous #1 in the computers. That may drop a bit if USC beats Notre Dame, as their marquee win (Notre Dame) will drop - the primary reason Ohio State dropped in the computers last week is that their marquee win, Texas, lost a second time. A close Michigan loss leaves the rematch possibility in play, as they'll still be pretty well regarded by the computers.

Rutgers being so high in the computers is pretty remarkable actually, due to their nonconference schedule - some of the computers weight nonconference schedule much more heavily than the conference schedule (note - this weighting often hurts SEC teams in the computers).

Should be an interesting rest-of-season.

by Travis (not verified) :: Tue, 11/14/2006 - 12:16pm

I can't see putting an ACC team higher than #15, myself, barring some non-conference wins next week or Wake Forest winning out. The league just isn't that good.

Ohio State's previous opponents (most notably Texas) went 2-8 last week (with the wins over Michigan State, who Ohio State had previously played, and Temple). That, plus playing 3-8 Nortwestern (who earlier in the season had lost to I-AA New Hampshire), explains part of their drop in the computer rankings.

by young curmudgeon (not verified) :: Tue, 11/14/2006 - 12:27pm

Tarrant, I hope you're right. To duplicate my post in the Garden Party thread, which is also pertinent here: It’s going to be hard to take seriously next week’s “Game of the Century� if the only thing at stake is which team gets to wear the home uniform in the national championship game. If, as we hear ad infinitum, the merit of the college system with no playoff is “the regular season really matters,� the loser of this game should be dropped from consideration. I don’t care if they are, “in the minds of the voters� and in various GIGO computer rankings “the second best team.� A rematch just doesn’t seem like a satisfactory national championship.

by MCS (not verified) :: Tue, 11/14/2006 - 1:12pm

and still no respect for the Badgers.
One loss
On the road
Against the #2 team in the country

by BB (not verified) :: Tue, 11/14/2006 - 1:12pm

Michigan needs not only Notre Dame but also Wisconsin to finish out with one loss as well for there to be any shot of a rematch (should they lose to tOSU, obviously) -- that game is really propping up their computer rankings as well, having given two different teams their only loss of the year. Of course, Wisco has only Buffalo remaining, so one would think they're pretty certain to finish with the lone loss. Vandy winning over Tennessee again this year would help them as well.

I don't think it's going to happen, but if Michigan loses on a bizarre play late or in OT, I would not be shocked at all to see them stay #2 -- a loss like that at tOSU would easily be the 'best' loss of any of the one-loss teams. And put me in the camp that says Rutgers should get the shot if they finish undefeated. The shutting down of a high-powered Louisville offense coupled with a win in Morgantown over a good WVa team is enough 'quality' wins for me if they've run the table.

by MCS (not verified) :: Tue, 11/14/2006 - 1:24pm

Michigan needs not only Notre Dame but also Wisconsin to finish out with one loss as well for there to be any shot of a rematch

I don't understand it. Isn't the fact that UM handed two teams their only loss of the season make UM that much stronger?

by Tarrant (not verified) :: Tue, 11/14/2006 - 1:24pm

I agree that should Rutgers win out, they will have proven themselves and should get a shot.

That said, should USC win out, I think it's moot right now - USC will get the second spot in the title game unless they win out "ugly". Even then, the computers will love them.

Re: #39. I feel bad for Wisconsin. One loss, to Michigan, and after this week it's possible that they lost to the #1 team, not #2. But since the assumpting is, unless Michigan or OSU are completely blown out of the water by the other, the loser will get a Rose Bowl at-large berth, Wisconsin will be left out no matter what they do.

by DGL (not verified) :: Tue, 11/14/2006 - 1:56pm

Of course, all this discussion of "...should Rutgers win out..." virtually guarantees they'll lose to either Cincinnati or WVU.

by BB (not verified) :: Tue, 11/14/2006 - 2:11pm

I don’t understand it. Isn’t the fact that UM handed two teams their only loss of the season make UM that much stronger?

Yes, which is why UM needs both of those teams to win out for there to be a high possibility of a rematch -- that was my point. One loss to the #1 team (assuming it's close) while giving two others their only losses of the season would seemingly make them the top one-loss team.

by Tom Kelso (not verified) :: Tue, 11/14/2006 - 2:23pm

#34 --

Not really arguing for Boise, but it is interesting to note they demolished that "team outside the top 25" that beat USC. Boise deserves the BCS slot that they will get -- what they do with it is another matter.

I love the tools (Golic, Patrick) on ESPN who keep saying that this is just like a playoff, because all the top eight teams still happen to have to play another top-eight team. This is a pure accident of chance, which just underscores the stupidity of the whole system anyway. It ranks right up there with changing the rules after the fact to mollify USC and the AP in 2003.

The worst thing about the whole scenario is that there will never be a playoff -- too many people have too much invested in the current situation to risk gambling the pile of money they already have for even the near-dertain potential of a larger pile down the road.

by Derek (not verified) :: Tue, 11/14/2006 - 2:39pm

The reason Wisconsin is in this position is their weak non-conference schedule. They played Bowling Green (4-6), Western Illinois (5-6), San Diego State (2-7), and Buffalo (2-8). That reminds me of Kansas State under Bill Snyder.

Considering that the Badgers missed Ohio State in conference this year, the schedule was just too soft...especially with the collapse of my Hawkeyes and most of the rest of the Big 10. The Badgers best win may have been at Purdue (they beat Penn State in Madison).

While I'm on the topic, I should point out that Iowa misses both Michigan and Ohio State in 2007 and 2008. If the Hawkeyes can rebound from this year, they will likely be in a similar position (on the outside looking in).

by Derek (not verified) :: Tue, 11/14/2006 - 2:43pm

I should also add that I think most Badger fans have to be ecstatic with how the season has turned out. One loss in a "rebuilding year" with a pretty nice recruiting class coming in. The coach with the TigerHawk tattoo on his leg has certainly impressed.

by Mike W (not verified) :: Tue, 11/14/2006 - 2:49pm

The following is a rant relating to how perceived conference strength evolves over the season, usually illogically. 1) Remember at the beginning of conference play, the CW was that the SEC was down because noone looked impressive. Now that they've all been beating up on one another, they're all supposed to be good. That may be true, but they're all flawed, and if one of them was a true national title contender, we might see an unbeaten from the conference, or only one one-loss team having achieved separation from a herd of SEC pretenders. If Arkansas gets through the SEC title game with one loss, good for them. But I've heard that LSU is the best team in the SEC (and country) and I've heard Tennessee is the best team in the SEC, and they have five losses between them (so far). Are we supposed to believe that all of these top SEC teams are THAT good? Seemingly what conference play has done for the SEC is that just by playing each other, Ark/Fla/Aub/Ten/LSU have all moved up in general esteem, win or lose. 2) On the Big Ten side, since a few teams are terrible, and the middle teams haven't been able to beat the top teams, then the feeling is that the top three teams haven't played anyone in conference. Now, we accept that OSU and Mich are terrific, but evidently not UW, because they haven't beaten anyone 'good.' Ok, fine, but meanwhile, all three top Big East teams are highly rated (as high or higher than UW). Why? Because they've played (or will play) each other, and recently. So it really seems that you don't have to beat good teams, you just have to play them. And they don't really have to be good either, as long as they in turn play a couple 'good' teams. Like they say at bad companies, "Just look busy."
This is not to say that Florida sucks, or that Wisconsin would crush Tennessee or Louisville - I don't know what would happen, and neither do you - but while we should not penalize teams for playing good teams, we MUST penalize them for losing. And if we only do so by docking them a small amount, and they then gain all of that back by some vague general ratcheting up of their conference's esteem, then winning isn't as important as it should be.

by Travis (not verified) :: Tue, 11/14/2006 - 3:37pm

Ok, fine, but meanwhile, all three top Big East teams are highly rated (as high or higher than UW). Why? Because they’ve played (or will play) each other, and recently. So it really seems that you don’t have to beat good teams, you just have to play them. And they don’t really have to be good either, as long as they in turn play a couple ‘good’ teams. Like they say at bad companies, “Just look busy.�

Again, look at Louisville's non-conference schedule. 5 games, 3-4 bowl teams (Miami needs to win 1 game), 3 BCS conference teams, 2 games on the road, 1 neutral.

Not one game was close. They beat Kentucky (4-3 in the vaunted SEC) by 31. They held Kansas State (who just beat Texas), at Kansas State, to 6 points (and Louisville's normal QB was out with an injury).

West Virginia's much-maligned schedule nontheless includes a win over #21 8-2 Maryland (28-0 in the 1st quarter, and 38-10 at halftime). Rutgers' schedule is certainly not strong, but includes wins over bowl teams Navy (on the road, 34-0) and Ohio (24-7), as well as 2 wins over bad BCS teams (Illinois and at North Carolina).

There's certainly a difference between, say, USC's non-conference schedule and a typical Big East team's. But there's also a difference between the Big East teams' schedules and those of teams (Texas A&M, Wisconsin) who scheduled 4 nobodies, all at home or neutral sites.

by Larry (not verified) :: Tue, 11/14/2006 - 3:48pm

Thanks, Pat. I'd say that Massey isn't well enough defined, as he discusses only his MOV based system and includes an arbitrary component (the time decay) which is described poorly [this component is the main reason for the differences with the other rankings]. So, it's really 3/6, which isn't very good. Also, it isn't clear to me how Wolfe differentiates undefeated teams in his Bradley-Terry method or what his actual ranking numbers are, since some of them are negative (My guess is they're log's of the P's, but I can't be sure).

Compare this to DVOA. Sure, none of us know the precise components or formulas, but we know what the final numbers are supposed to represent (the increase in success points [which represent real points] over the situational average) and the criteria by which the components of the system are determined (to maximize correlation with wins). That's better than what I get from half the BCS ratings. I'd prefer to go with the Simulated Voting Monkeys, though they actually rank OSU 5th.

by Pat (not verified) :: Tue, 11/14/2006 - 4:04pm

Thanks, Pat. I’d say that Massey isn’t well enough defined, as he discusses only his MOV based system

Massey's non-MOV based system is the same thing, replacing the game-output function with a 1 or a 0 depending on whether or not a team won or lost.

and includes an arbitrary component (the time decay) which is described poorly

Yeah, that's why I said it would take a bit of work, but given that the time decay is the only open variable, you could figure it out without *too* much effort.

I have no idea how Massey justifies using a time-weighting on a pure win/loss ranking system. On the margin of victory based rankings, sure, but on the W/L rankings, that's just retarded.

by dryheat (not verified) :: Tue, 11/14/2006 - 4:18pm

You know, if Rutgers had been ranked in the pre-season rankings, not highly...let's say in the 17-22 range, they'd be ranked #3 and nobody would be questioning it.

Did they have to go 4-0 to crack the top 25, or 5-0? How high did a mediocre Miami of Florida team debut, and how many weeks did it remain in the top 25 despite losing?

It seems to me that the most unfair aspect of the BCS rankings is the arbitrary starting point for the schools, which is largely based on name-recognition/previous year's success.

by Travis (not verified) :: Tue, 11/14/2006 - 4:35pm

Jeff Bowden has resigned as Florida State's offensive coordinator, effective at the end of the season.

by Larry (not verified) :: Tue, 11/14/2006 - 4:37pm

Re: Massey's time-weighting
Considering most of the major non-conference games take place early in the season, he's throwing away the most important pieces of information for ranking teams. I mean you could argue that those games don't represent teams' current skill level, but if that's so, you might as well designate them exhibition games.

by Mike W (not verified) :: Tue, 11/14/2006 - 5:10pm

Travis #49 - everybody plays a couple 'bowl' teams. Penn St., Purdue, and Iowa (Indiana?) are going to bowls, too. I don't mean to defend Wisconsin; I think the Capitol One Bowl is entirely appropriate and their opponent will probably be about as good as they are. I thought that was clear. What I was objecting to is the reciprocal legitimizing of teams by beating each other, which is a problem separate from the often commented on (and very real) issue of preseason rankings poisoning late-season rankings.

by BadgerT1000 (not verified) :: Tue, 11/14/2006 - 5:23pm

While nobody can ever be certain it is believed in Badger country that Barry/Pat Richter eased off on the non-conf scheduling to ease the transition for a new coach. If you look back at Wisky's schedule you will find legit opponents in Oregon, Colorado (when they were good), and Fresno State (when they were really good). But Alvarez had a plan and was going to help the next guy move into his new role. At least that is what more then a few Badger fans suspect.

The next few years have more of the same though Fresno St. does pop up. As long as Pat Hill is around one has to think they will be more good then bad in most years.

I would also point out that just last year the MAC was a really good conference and that Bowling Green almost BEAT Wisconsin in Madison with Omar Jacobs at QB. So to rip them now ignores the very recent and successful past of the MAC teams.

But I understand folks looking in just the context of this year and declaring their schedule soft.

Though I think any team that gets Penn State in a bowl better be ready to rumble. If they get just a hint of offense they are going to absolutely rock somebody's world. The defense and special teams are darn good.

by Travis (not verified) :: Tue, 11/14/2006 - 5:29pm

Re: 55

Travis #49 - everybody plays a couple ‘bowl’ teams. Penn St., Purdue, and Iowa (Indiana?) are going to bowls, too.

Not everybody plays bowl teams out of conference, though. Pitt, South Florida, and (possibly) Cincinnati are going to bowl games, too, but I didn't bring them up.

What I was objecting to is the reciprocal legitimizing of teams by beating each other, which is a problem separate from the often commented on (and very real) issue of preseason rankings poisoning late-season rankings.

This seems like largely an SEC problem, though, and has little to do with the Big East. West Virginia and Louisville were already thought of (by many, if not all) as legitimate teams before the season.

by Travis (not verified) :: Tue, 11/14/2006 - 5:34pm

Re: 56

Just to clarify, I have no problem with Wisconsin scheduling as soft as they did, especially in a transition season, so long as they're satisfied with a Capital One Bowl berth and a poll ranking in the 11-15 range.

by Sophandros (not verified) :: Tue, 11/14/2006 - 6:23pm

If the SEC is "serious" and full of "serious" teams, then what' Notre Dame? Gothic? They're pretty good, have history, and all of that church imagery going on, so if it's serious SEC competition, then teams who play ND have to call them Goth ND.

I would also qualify Army as Goth, as well, for similar reasons PLUS the fact that their nickname is The Black Knights of the Hudson. Besides the color scheme which is totally goth, they bring in the Holy Grail Factor. Then again, I guess that means that they can never be serious, and thus, they can never be in the serious SEC.

by Derek (not verified) :: Tue, 11/14/2006 - 9:09pm


All the Big Ten teams schedule the MAC and you have to admit that losses are pretty rare, even to the best of MAC teams (which Bowling Green is currently not). Decent Big Ten teams expect to win against the generally well-coached but physically inferior MAC squads. In Ben Roethlisberger's final season, I saw him lose 21-2 in Kinnick Stadium against Iowa...it was the only loss Miami of Ohio had that season.

I wasn't ripping Wisconsin for the scheduling decision nor was I suggesting that they are habitual offenders. Alvarez certainly had his reasons and it has worked out quite well but it is hard to feel too badly for the Badgers given the schedule they have played. They deserve to be highly ranked...I just don't see them as a BCS team. I'll even be pulling for them in the Capital One Bowl.

by MRH (not verified) :: Thu, 11/16/2006 - 4:19pm

A little late to the party but some of the general comments out in the blathosphere inspired this analogy:

Billy Packer:George Mason
SEC fans:Rutgers

The ACC and North Carolina are almost always better than the Colonial and GMU. But not always. And the SEC and Florida are almost alwasy better than the reconstitued Big East and Rutgers. But not always.

Don't discount an inspired team believing in their coach.