Varsity Numbers: The Unexpected

Varsity Numbers: The Unexpected
Varsity Numbers: The Unexpected
Photo: USA Today Sports Images

by Bill Connelly

In 1999, Lou Holtz's first season in Columbia, the Gamecocks went 0-11 and scored 87 total points. The very next year, they went 8-4 and whipped Ohio State in the Outback Bowl. Turner Gill better hope for some Holtzian second-year magic, as his first season is becoming an amazing one for all the wrong reasons. Kansas' 59-7 loss to Kansas State Thursday (the same Kansas State team that got run off the field by Nebraska on October 7) was the third "Worst Kansas loss ever" candidate from this season alone. The Jayhawks' 6-3 loss to North Dakota State in Gill's debut was really bad, and the 55-7 road loss to Baylor was worse. But this could very well be the worst of all. The Jayhawks have now been outscored 114-14 by two average-at-best Big 12 programs, and they could very well go 0-8 in conference despite the lack of either Oklahoma or Texas on the schedule.

We are quickly seeing the biggest problem with preseason Outsiders projections. We knew that coaching changes cause uncertainty, and we made an attempt back in August to figure out the general impact of coaching changes. But a lot of the teams on which we have whiffed most severely so far -- Tennessee (projection: 20th, current rank: 77th), Kansas (projection: 51st, rank: 102nd before the KSU game), Texas Tech (projection: 18th, rank: 52nd) -- brought in new coaches.

(Two other major misses, BYU and Penn State, have given true freshmen serious playing time at quarterback. But that's for an entirely different column.)

Is Illinois ... Good?

One team on the Surprisingly Good list also brought in some new coaching talent -- it just wasn't in the top spot.

A couple of weeks ago, I took the time to point out a couple of early outliers, teams who seemed to have overachieved statistically in the first few weeks of the season. Needless to say, Kentucky and Virginia have not panned out well as sleepers. Kentucky might be alright in a different conference, but Virginia has done nothing but tumble. They were 25th in the F/+ rankings just two weeks ago, now they rank 60th.

But now we're almost halfway through college football's regular season. Should we make something of the fact that Illinois ranks 21st in F/+ and 10th in S&P+? Do the Illini, who have been reasonably steady all season, have staying power? Their two losses are to F/+ Top 20 teams, both by respectable margins. They just destroyed Penn State in Happy Valley (even though the Nittany Lions clearly are not a great team this year, that is still impressive). They are 3-2 with three extremely winnable home games (Indiana, Purdue, Minnesota) standing between them and bowl eligibility. A season that was supposed to be Ron Zook's final go-round is beginning to show promise.

Is Illinois ... good?

Let's start answering that question by looking at where the S&P+ rankings stood at this point last season. Because of the strength of their early schedule, Tennessee stood at No. 3 with a 3-3 record. Arkansas, at 3-2, was fifth. Nevada and Fresno State, both 2-3, ranked 19th and 22nd. Tennessee went 4-3 the rest of the way and made the Chick-Fil-A Bowl. Arkansas finished 8-5 and created strong buzz for 2010. Both Nevada and Fresno State won six of eight to end their seasons in bowls. Though none of them won out or played at a Top 10 level, they all ranked surprisingly high at midseason, and they all played reasonably well down the stretch. There is a precedent here.

Let's say for a moment that the Illini do continue playing well. Let's say they win the three easy home games, beat Fresno State at the end of the season and pull off a road win over Northwestern, Michigan, or (least likely) Michigan State. How will it happen?

Identity. Never underestimate the impact of committing 100 percent to a style and personality of ball. Illinois is exactly who they will be the rest of the season -- a team that runs constantly (76 percent of the time on standard downs and 45 percent on passing downs, eighth and 13th most in the country, respectively), both because they're good at it and because they are protecting their redshirt freshman quarterback, Nathan Scheelhaase, as much as possible. They alternate between pistol, shotgun and behind-center looks; they offer Scheelhaase on the option; they throw big, talented running backs (Mikel Leshoure is a very good back) at you; and they occasionally wing it to shifty receiver A.J. Jenkins in space. New offensive coordinator Paul Petrino has helped the Illini to establish a unique identity, and they have fully invested in it.

Timing. The Illini pick their spots well. They rarely throw, but when they do, it is effective. They rank first in the country in Passing PPP+, though that has a lot to do with the pass defenses they have faced. They will not finish the season that high, but their timing has been strong.

Tackling. They rank 10th in the country in Defensive PPP+, and despite playing two ranked teams (Ohio State and Missouri), they are allowing fewer than five yards per play. Led by an outstanding set of linebackers -- Martez Wilson, Nate Bussey and Ian Thomas have exceeded all expectations -- they are allowing just 3.7 yards per carry and a semi-respectable 6.1 yards per pass. New defensive coordinator Vic Koenning has installed a much higher level of discipline in this unit.

In the offseason, it was easy to see Ron Zook's moves -- hiring what seemed like an entirely new coaching staff -- as sure signs of desperation. It was the last-gasp effort to save his job, it seemed, and that rarely works. Five games into this new experiment, with both Petrino and Koenning are thriving, and it looks like it might have just worked.

So what happens if they lose to Michigan State and Indiana on their way to a 4-8 season? How will it happen if things quickly go sour, and I come to realize that I was severely jumping the gun here?

Running Game Could Falter. Looking at the raw play-by-play statistics, you see pretty quickly that their strong "+" rankings have been bolstered mostly by a strong schedule. Illinois ranks just 68nd in raw Rushing Success Rate and 80th in Rushing PPP. With connectivity still a problem this early in the season (and always), the schedule-adjusted numbers could still be telling an inaccurate tale, and if Illinois suddenly can't run, Scheelhaase could be exposed and forced to pass more. That probably will not end well.

No More Element of Surprise. When Illinois came out in a pistol-with-fullback formation against Missouri in the opening week of the season, the Tigers were caught off guard. There was no way to scout what Illinois was going to try to do. But at this point in the season, nobody is going to be surprised. As the "How to stop Scheelhaase" book becomes more and more clear, the second half of the season could kick-start a premature sophomore slump. Meanwhile, their defensive tendencies could be exposed the same way. It is really hard not only to improve, but also to sustain it. Maybe Zook, Petrino, and Koenning have pulled off the feat, but let's see how they do against Michigan State this weekend before jumping to conclusions.

Ask the Expert

Fascinated by Illinois' continued presence in the S&P+ rankings, I asked the Chicago Tribune's Shannon Ryan, who has been covering the Illini this season, to do an e-mail Q&A about the Illini.

Bill Connelly: It's still early, and the numbers are still fluctuating constantly, but Illinois currently has the 14th-best offense in the country according to our opponent-adjusted S&P+ measure. They are 21st rushing and 14th passing (it's a per-play measure, so it's not looking at the fact that Illinois passes fewer than 20 times per game). How convinced are you that they can keep up this level of performance for another seven games? (Disclaimer: at this point a couple of weeks ago, Virginia also ranked pretty high, but they were pretty clearly a statistical mirage.)

Shannon Ryan: Illinois has already played two of its three or four hardest games, facing Ohio State and Penn State with Michigan State and Michigan left on the schedule. Thanks to their schedule, the Illini don't face the Big Ten's top scoring defense (Iowa) or No. 4 team in total defense (Wisconsin). The Illini played well against the conference's top two pass defenses (Ohio State and Penn State) and No. 1 total defense (Ohio State).

Illinois is certainly a run-oriented offense with quarterback Nathan Scheelhaase's ability and star-in-the-making running back Mikel Leshoure. But Scheelhaase, a redshirt freshman, has shown progression in his passing and decision making from game to game. He's known for his maturity and for being studious in the film room, and it's showing. While the Illini aren't scoring a ton of points (other than the 33 they dropped on Penn State), it helps that they have a reliable kicker in Derek Dimke, who is 10 of 11 on field goals.

BC: Illinois' offense has improved in 2010 despite handing the reins to a redshirt freshman at quarterback (Nathan Scheelhaase). What do you feel has been the biggest cause for the improvement? Scheelhaase? The new offensive coaches?

SR: Offensive coordinator Paul Petrino has brought a dramatic change to Champaign. Players talk about his demand for perfection and more aggressive approach. Nathan Scheelhaase is more mature than most freshmen and drives himself to not repeat the same mistake twice. I think giving the ball more to Mikel Leshoure has been a significant part of the Illini's success as well.

BC: I was at the Missouri-Illinois game at the start of the season. I was struck by how few tackles the defense missed. They played disciplined football, and I loved the game Nate Bussey put together (he obviously had another very good one at Penn State). Illinois' defense currently ranks seventh in our Defensive S&P+ measure, third against the run. I'll ask the same question I did of the offense: Is this sustainable?

SR: The defense is the most surprising aspect of Illinois' early turnaround. They lost a significant amount of players from the secondary (Supo Sanni, injury; Walt Aikens, dismissed; Terry Hawthorne, injury), but it has not phased this group. New defensive coordinator Vic Koenning has stressed tackling, a sore spot last season for the Illini. Bussey has been a solid playmaker and the return of middle linebacker Martez Wilson from a neck injury cannot be overstated. After a few games, he returned to his aggressive form and is part of group of Illini defensive players who have 31 tackles for losses of 132 yards so far this season.

BC: I'm firmly of the belief that if some other Big Ten team (say, Michigan State) were putting together Illinois' current performance -- lose two tight games to undefeated teams, beat the best team in the MAC, kill Penn State in Happy Valley -- people might be taking them more seriously. But since this is Ron Zook and Illinois, they have a lot more to prove before people believe in them. Do you think this is an accurate assessment?

SR: I think that is true, but the skepticism may be warranted. After reaching the Rose Bowl after the 2007 season, expectations were high. For reasons that are tough to determine, the Illini failed miserably at matching those expectations. I think fans are just being cautious before getting too excited. A victory at No. 13 Michigan State would bring about well deserved credit.

BC: What do you think: Do they keep up the strong play at Michigan State this weekend?

SR: Obviously, Michigan State poses the biggest test on the schedule other than Ohio State. The Illini have high confidence after playing well against OSU and beating PSU, so they won't feel jittery heading into East Lansing. I think it's going to be a close game, but I think the Spartans offense (35.8 points) will be tough for the Illini to keep up with.

Is Oregon ... Overrated?

I had a week to prepare for Illinois' high presence in the rankings. Another team's ranking, however, caught me completely off guard. By simply letting the S&P+ kryptonite known as Washington State hang around for a half, Oregon fell 15 spots in the S&P+ rankings, even as they were vaulting to No. 1 in FEI. It was a perfect illustration of both the differing impact of strength of schedule on two different rating systems, and the hazards of drawing definitive conclusions about teams six weeks into the season, whether you are using numbers or eyeballs.

As I mentioned in the comments section in the early week Varsity Numbers, I think Oregon is one of the best teams in the country. They are frightening on offense and opportunistic on defense. They lost Jeremiah Masoli and, it seems, got better.

So what gives? What do the S&P+ ratings see that nobody else does?

What have they done that a few others couldn't? Two weeks ago, the Ducks absorbed a strong uppercut from Stanford (currently 11th in F/+) before applying a knockout blow of their own. Their 52-31 win over the Cardinal was easily their best outcome of the season. The problem is, most of their other outcomes have been rather ho-hum, at least as far as the ratings are concerned. They humiliated New Mexico (currently 119th out of 120 teams in the F/+ rankings) and Portland State (a "Tier 5" FCS team), and because of that, their ratings did not take much of a beating from playing such lowly teams. But they let Tennessee (current F/+ rank: 77th) hang around for three quarters in Knoxville before applying the knockout blow, and they let lowly Washington State (104th) hang around for two. The way the S&P+ opponent adjustment works, this hurt them quite a bit. Throw in a statistically unimpressive win over Arizona State, and you've got quite the mixed résumé. In fact, I'd be willing to bet that if Utah had put this exact set of wins together, we would be touting them as a Top 10 or Top 15 team ... but not a Top 5 team.

If Oregon is a national title-caliber team, their remaining schedule gives them more than enough opportunities to prove it. They'll face UCLA (current F/+ rank: 44th), USC (24th), Washington (45th), California (17th), Arizona (ninth), and Oregon State (12th). In other words, their strength of schedule is about to skyrocket. If they finish 12-0, it is a virtual lock that they will finish, at the very least, in the Top 5 of both S&P+ and F/+.

Logjam. When you take a look at the F/+ rankings below, you will see that Oregon ranks 18th, behind four two-loss teams. You can use this as evidence that the whole S&P+ method (FEI is in the clear, since Oregon ranks first) is a bunch of hooey. Or you can do the following: look at the logjam.

Let's divide the F/+ Top 25 below according to range.

Top F/+ Teams by Range
+30% or more
+24% to +30%
+18% to +24%
Ohio State Alabama
South Carolina
Boise State
Oregon State
Virginia Tech
Michigan State

Even with the weak schedule, Oregon is potentially just a handful of good or bad plays away from sixth in the current F/+ rankings. There is a cluster of close teams ranked between sixth and 18th, and Oregon just happens to rate slightly below the others. In fact, Oregon is about as close to No. 5 Nebraska as No. 2 Alabama is to No. 1 Ohio State at the moment. After the top four, it starts to get quite messy. With a huge win over UCLA next weekend, the Ducks could be right where we all think they belong in the ratings.

But if Oregon falters, then S&P+ is going to start to get a really big head.

F/+ Rankings

The last two weeks have uncovered a potential issue with the F/+ rankings: S&P+ has a much larger standard deviation than FEI, which has meant that teams at the top and bottom the S&P+ list seem to carry a little extra weight. When you see where Miami ranks this week, you'll see that this is still an issue. We might try to make a tweak with that in future weeks. Keep in mind that it's still early. (Remember, Tennessee was No. 3 in S&P+ a year ago. They clearly did not finish No. 3.)

F/+ Top 25 (After Six Weeks)
Rk Team F/+ Last
Chg. S&P+ Rk FEI Rk
1 Ohio State (6-0) +36.1% 2 +1 311.6 1 .216 11
2 Alabama (5-1) +28.3% 1 -1 275.3 2 .236 5
3 South Carolina (4-1) +27.4% 12 +9 266.4 4 .261 3
4 Boise State (5-0) +27.1% 5 +1 264.6 6 .264 2
5 Nebraska (5-0) +25.2% 13 +8 266.2 5 .219 9
6 Miami (3-2) +23.9% 3 -3 272.3 3 .163 17
7 Auburn (6-0) +23.4% 11 +4 255.6 9 .234 6
8 TCU (6-0) +22.4% 10 +2 252.6 12 .227 8
9 Arizona (4-1) +22.0% 6 -5 262.9 7 .171 16
10 LSU (6-0) +22.0% 17 +7 250.8 14 .229 7
F/+ Top 25 (After Six Weeks)
Rk Team F/+ Last
Chg. S&P+ Rk FEI Rk
11 Stanford (5-1) +21.7% 9 -2 244.5 19 .253 4
12 Oregon State (3-2) +20.7% 25 +13 251.7 13 .199 13
13 Virginia Tech (4-2) +20.6% 16 +3 248.9 15 .209 12
14 Oklahoma (5-0) +19.8% 15 +1 243.5 20 .219 9
15 Missouri (5-0) +19.3% 14 -1 254.3 11 .157 18
16 Michigan State (6-0) +18.9% 26 +10 244.9 18 .194 14
17 California (3-2) +18.9% 19 +2 257.8 8 .132 24
18 Oregon (6-0) +18.6% 8 -10 229.3 30 .264 1
19 Florida State (5-1) +17.1% 24 +5 246.1 17 .154 19
20 Iowa (4-1) +17.1% 7 -13 246.6 16 .151 20
F/+ Top 25 (After Six Weeks)
Rk Team F/+ Last
Chg. S&P+ Rk FEI Rk
21 Illinois (3-2) +15.2% 32 +11 254.5 10 .075 40
22 Arkansas (4-1) +14.3% 20 -2 236.1 23 .145 22
23 N.C. State (5-1) +13.3% 38 +15 225.0 38 .180 15
24 USC (4-2) +13.1% 18 -6 230.5 25 .149 21
25 Florida (4-2) +12.4% 6 -19 239.2 21 .092 34

26. West Virginia (4-1), 27. Utah (5-0), 28. Oklahoma State (5-0), 29. Notre Dame (3-3), 30. Clemson (2-3), 31. North Carolina (3-2), 32. Michigan (5-1), 33. Wisconsin (5-1), 34. Nevada (6-0), 35. Mississippi State (4-2), 36. Arizona State (3-3), 37. Kentucky (3-3), 38. Pittsburgh (2-3), 39. Georgia (2-4), 40. Ole Miss (3-2), 41. Texas A&M (3-2), 42. Texas (3-2), 43. Central Florida (3-2), 44. UCLA (3-3), 45. Washington (2-3), 46. Vanderbilt (2-3), 47. Georgia Tech (4-2), 48. Kansas State (4-1), 49. Air Force (5-1), 50. Colorado (3-2).

F/+ vs AP


Air Force (AP Rank: 23rd, F/+ Rank: 49th). I am beginning to notice a trend in looking at the F/+ vs AP differences. While mid-major teams seem to hit a glass ceiling in terms of how high they can be ranked -- hence Boise State's fall last week despite a 59-0 win -- they also seem to get a bit over-valued in terms of snagging low Top 25 votes. This makes sense, really. Voters want to give credit to teams for winning games, and programs like Air Force benefit from that. The Falcons looked great in challenging Oklahoma, but they gave up too many points to Colorado State (and didn't score enough against Navy). They certainly have a chance to prove their Top 25 bona fides against an angry San Diego State team this week, and against TCU the next.

Oregon (AP Rank: 2nd, F/+ Rank: 18th). Probably well-covered territory at this point.

Utah (AP Rank: 11th, F/+ Rank: 27th). I expected the Utes to rise more than two spots with their utter demolition of Iowa State in Ames last week.

Wisconsin (AP Rank: 18th, F/+ Rank: 33rd). It wouldn't be an "Over-Appreciated" list without a Wisconsin appearance. Like Air Force, though, the Badgers have ample opportunity to prove themselves -- Ohio State comes calling Saturday night.


Miami (AP Rank: 27th, F/+ Rank: sixth). As surprised as I was about Utah's lack of a rise, Miami's failure to drop in the rankings despite getting waxed by Florida State was potentially even more jarring. Oregon was this week's "What The ...?" team of the Week, but it's safe to say that Miami will be in the hot seat next week -- especially if they fail to sufficiently thump Duke this weekend.

Virginia Tech (AP 'Rank': 29th, F/+ Rank: 16th). Stop the presses! FO likes Philadelphia Virginia Tech! While we have been a bit self-deprecating about the fact that the Hokies seem to rank highly in our system no matter what they do on the field ... you have to admit that they have looked pretty damn good in recent weeks. They get to feast on Wake Forest and Duke now, with Georgia Tech coming to Blacksburg on November 4. In other words, they will likely be 7-2 and 5-0 in ACC play when they head to Chapel Hill on November 13. Are you ready for the possibility of an ACC champion that lost to James Madison?

Oregon State (AP Rank: 24th, F/+ Rank: 12th). The Beavers benefit from a nice strength of schedule -- they have now played three F/+ Top Ten teams away from home (No. 4 Boise State, No. 8 TCU, and No. 9 Arizona) and have acquitted themselves quite well in the process. As with the last couple of seasons, Oregon State should be a player in the Pac-10 race ... if they can overcome the loss of receiver James Rodgers to injury, anyway.

Arizona (AP Rank: 17th, F/+ Rank: 8th). This is still a strong team despite the loss to Oregon State.

"What The ...?" Team of the Week

This week: Oregon.

Next week: Miami. You're on notice, Hurricanes.

AP + F/+ = BCS?

Just for fun, here's what the BCS Top 10 would look like this week if it only took into account the AP Poll (60 percent) and F/+ rankings (40 percent). Consider it a thought experiment if nothing else.

1. Ohio State (6-0)
2. Boise State (5-0)
3. Nebraska (5-0)
4. Alabama (5-1)
5. TCU (6-0)
6. Auburn (6-0)
7. South Carolina (4-1)
8. Oregon (6-0)
9. Oklahoma (5-0)
10. LSU (6-0)

Upset Watch

Louisville over Cincinnati. Spread: Louisville +3 | F/+ Projection: Louisville by 2.9. As strange as it may sound, this is a pretty big game in the Big East race. If Cincinnati can continue its recent momentum by putting up a big score on Louisville, it will be a statement that new coach Butch Jones has figured out his team and that they are ready to challenge for their third straight conference title. Meanwhile, Louisville has quietly played competitive football this year. Their two losses were to F/+ Top 40 teams Kentucky and Oregon State, both by just seven points. They looked far from inspiring in wins over Arkansas State and Eastern Kentucky, but the margin to which they beat lowly Memphis (56-0) was impressive. They boast an F/+ rank of 53rd, and while they probably are not conference contenders by any means (though in the Big East, you never know), a win over the Bearcats would get them rolling back toward bowl eligibility in their first year under Charlie Strong.

Missouri over Texas A&M. Spread: Missouri +3.5 | F/+ Projection: Missouri by 8.1. This is a really interesting game in terms of "strength vs. weakness" matchups. Missouri ranks ninth in the country in turnover margin, while A&M ranks 116th, thanks primarily to quarterback Jerrod Johnson's propensity for interceptions and fumbles. On the other hand, Mizzou quarterback Blaine Gabbert has struggled in previous attempts against a three-man front, and A&M has the speed in their offensive backfield to which Mizzou might be vulnerable. Really, this is as even a matchup as you could imagine. Whichever team exploits the other's weaknesses the most will come away victorious.

California over USC. Spread: California +2.5 | F/+ Projection: California by 5.5. This matchup pits a team ready start living up to its lofty F/+ rankings (I'm sure everybody on the Cal football team knows exactly where they rank in S&P+, ahem) against a team trying desperately to keep things together after back-to-back last-second losses. Can USC avoid the negative momentum? Can Cal overcome the mental block that seems to strike them against the Trojans (four of their last five games against USC have resulted in losses by at least 14 points ... albeit against USC teams quite different from this one)?

Oklahoma State over Texas Tech. Spread: OSU +3.5 | F/+ Projection: OSU by 6.5. As the line shows, it is difficult to pick Oklahoma State to win in Lubbock when they haven't done so since Bing Crosby's "Swinging on a Star" was a hit (1944). But this is one of the Cowboys' better opportunities to do so. The game might last five hours with all the quick passing, but to the extent that recent performance matters, OSU certainly holds the edge in this one ... as long as they don't wait until the third quarter to get rolling like they did last week against La.-Lafayette.

The Playlist

In honor of South Carolina's big win last weekend ...

"Goin' Down South," by R.L. Burnside
"South Bound Suarez," by Led Zeppelin
"Southern Cross," by Crosby, Stills & Nash
"Southside," by Common
"Southtown Girls," by The Hold Steady
"The Carolina," by Chatham County Line
"Carolina," by Phish
"Carolina In My Mind," by James Taylor
"Oh My Sweet Carolina," by Ryan Adams
"Carolina Moon," by Thelonious Monk

Any excuse to reference both R.L. Burnside and Thelonious Monk is a good one.

Closing Thoughts

If you follow college football, it has been impossible to miss this week's release of Death to the BCS, an anti-status quo (to put it gently) publication by Yahoo! Sports' Dan Wetzel, Josh Peter, and Jeff Passan. Early reviews (and there are plenty of them at the moment) point to exhaustive research and investigation, and I am all for it. I have not gotten a chance to read it yet (How do I get on the "Send him a free copy of anything related to college football" list?), but it has me curious.

I have never been a heavy-duty playoff proponent. I will not object if or when a playoff is instituted, but I don't significantly mind the "two-team playoff" structure that we have now. (For one thing, playoffs do not always crown the most "deserving" champion either. We want a playoff because of excitement as much as anything ... not that there is anything wrong with that.) Sometimes four or six teams deserve a legitimate shot at the national title, but at other times (2007, for instance), nobody really does. My imagination goes wild when thinking about a playoff (in fact, I've already drawn up the perfect one), but my eyes have always glazed over when I see somebody going on an anti-BCS diatribe. What I hope to see from Death to the BCS, when I get the chance to read it, is simply why "the BCS" is the enemy instead of the overall college football structure. To me, the BCS has long been an improvement over the previous system. We may disagree on which two teams should rank No. 1 and No. 2 at the end of the regular season (and to say the least, the rankings process itself could be improved drastically) ... but at least we actually get a No. 1 versus No. 2 matchup this way. It very rarely happened before the BCS (and its Bowl Alliance predecessor) came into play.

If you want to rage against college football's status quo and the people who ensure that it remains the status quo, then I'm all ears. But labeling the BCS itself as the enemy has always struck me as extremely misguided -- as if the higher-ups were almost ready to put a playoff together before the BCS folks stormed the gates and got their way. I look forward to giving Wetzel, Peter, and Passan a chance to change my mind on that.


38 comments, Last at 18 Oct 2010, 10:36am

1 Miami

Can you talk a bit about why Miami is so high in S&P? If Oregon is this week's "what the" team on the low end, the 'Canes look like the "what the" team on the high end.

If (for the sake of the argument) Miami is really about the 15th best team, I would think that waxing Pitt and getting waxed by FSU roughly cancel relative to what you'd expect for a 15th place team. I would expect that giving Ohio St something of a game wouldn't move the dial much, and neither would a 9-point win over a Clemson team that hasn't been so great thus far.

Also, why is their defense rated #2? Holding Pitt to 3 was quite good... but what else was? 36 points and 414 yards to Ohio St isn't horrible but isn't a great showing either. Ditto for 21 points and 311 yards to Clemson, and 45 points / 471 yards to FSU looks like a bad showing to me. Did the model just totally love the Pitt game and the AA game (where they completely shut them down)? Or am I unfairly maligning their other defensive performances?

2 Re: Varsity Numbers: The Unexpected

What's crazy is that Turner Gill built Buffalo into a respectable program. He looked like an excellent pickup for KU based on both that and his conference ties dating back to his playing at Nebraska. It's not just that they've been losing but that they've just been so erratic. (They managed to win at Georgia Tech before giving up 55 at Baylor?!) The K-State is a good example: It was only 3-0 after a quarter then they completely stopped playing--it was 54-0 early in the 4th before K-State started substituting.

Regarding Air Force: It's hard to argue with Air Force being in the Top 25--their lone loss was a fantastic effort at Oklahoma. But the bigger point is it is pointless to debate about the the bottom part of the Top 25 because those issues fix themselves--the teams either keep winning and prove their worth or they lose and fall out. Over-ranking teams at the top of the rankings can be a problem because it impacts the BCS and invites to the bigger bowls. But the bottom of the rankings mean very little.

4 Re: Varsity Numbers: The Unexpected

I would make a couple caveats to your point about the bottom of the rankings not meaning much:

1) It's possible that a team that is supposed to be much higher is at the bottom of the top 25, in which case that would matter.

2) If there's a flaw (or seeming flaw) that shows up in the bottom of the top 25, that may also be impacting other teams' ratings, though perhaps to a lesser degree. IMO any rating which seems substantially off is worth looking into to figure out if there's a problem or oversight, in which case a useful model adjustment can come out of the analysis. I've done that myself a couple of times. IMO it's good to always be keeping an eye out for ways to make things better, especially since everyone else (esp. Vegas) seems to be getting better at making picks as time goes on.

8 Re: Varsity Numbers: The Unexpected

The thing with Gill is that he only had one winning season out of his four as a head coach. He wasn't exactly Brian Kellyesque coming out of the MAC.

2006- 2-10
2007- 5-7
2008- 8-6, MAC champs, bowl loss to UCONN
2009- 5-7

His Buffalo teams never beat a BCS conference team, either. He took over a team that was 1-10, so he improved 3 of the 4 years.

26 Re: Varsity Numbers: The Unexpected

Gill took over a Buffalo team that had gone 5-41 the previous 4 seasons and were losing big to FCS teams Lehigh and Colgate. Going 20-30 over the next four years was a HUGE step forward.

And Brian Kelly was 10-12 his first two years at Central Michigan before his breakthrough campaign in 2006.

6 Re: Varsity Numbers: The Unexpected

My complaint with the BCS versus the old system is that it's pretending to something it's not.

The old system was a totally arbitrary pick by voters. The BCS is a totally arbitrary pick by voters plus some window dressing to make it look slightly less arbitrary. In addition, we lost the tradition of Big 10/Pac 10 Rose Bowl in exchange for that slight change in arbitraryness.

My problem with the BCS system is that not every team has a chance to be national champion. The rise of Boise State, TCU and Utah (and sometimes BYU) points this out. And I'm a Michigan State fan and have not allegiance to any of those teams. It irks me that they call something a championship when about half the teams are effectively ineligible to win it. I'd like to see a playoff with all the conference champs, plus an at-large team or two, just to even it out. And I won't feel bad for anyone who's 12-1, looks like the second-best team, and gets excluded because they lost in their conference championship game.

Or a return to old bald-faced arbitrary system.

7 Re: Varsity Numbers: The Unexpected

Not taking too much issue with what you have added here. But I don't think Boise State, TCU and Utah have been one of the top-two teams in college football (pre- or post-bowl) yet, so I don't subscribe to their having had no chance. This year may be different, perhaps due to unprecedented poll position, but perhaps due to their being deserving for the first time. If Boise State gets to the BCS championship this year, does the entire argument against the BCS as somehow preventing half of FBS from winning a title go out the window?

Second, is this just a question of semantics? If the BCS championship was named the Terrific Bowl and we crowned a Terrific Bowl champion every year via the same methodology, does that eliminate the "window dressing"? And who is fooled by what the BCS is and what it is not?

9 Re: Varsity Numbers: The Unexpected

Well, Boise State in the BCS championship would certainly take some wind out of my argument's sails. But, they were anointed by the voters preseason and they were the only non-BCS team who had the chance to make the championship game if they went undefeated in an impressive fashion. Nobody from the Mountain West, (Unless TCU does) Sun Belt, C-USA or MAC has the same chance. That means half the teams minus one have no chance of playing for the championship, no matter what they do.

And, yeah, it's sorta semantics and, yeah, I'd be happier if they called it the Terrific Bowl, but the main thing that bugs me is the unequal oppurtunity. Even if the non-BCS teams get blown every year like the conference champs who end up as 16 seeds in the NCAA b-ball tourney, at least they got a chance to play.

11 Re: Varsity Numbers: The Unexpected

I guess what I'm saying is that Sun Belt, CUSA, MAC teams have as much chance of participating in the BCS title game as Minnesota, Iowa State, Washington State, Vanderbilt, Syracuse and Duke. None. None of those teams are anointed at the beginning of the year either, none of them have gone undefeated in impressive enough fashion.

Now I agree that Boise State, Utah and TCU have had undefeated seasons and have not participated in the championship. But neither did Cincinnati last year. The non-AQ teams have to be more impressive than they have been, and they have to have the good fortune of having that success in a year in which others are not more impressive. Had Boise State or Utah had their run in 2007, we may already have had the glass ceiling argument shattered.

This year might be the year. Not just because of Boise State's anointing at the beginning of the season, but because their victories over Virginia Tech, Oregon State and Nevada, combined with obliterations of everyone else, combined with no other teams proving they deserve it more might make it happen.

23 Re: Varsity Numbers: The Unexpected

I don't agree that the Sun Belt, CUSA, or MAC teams have as much change of participating in the BCS title game as the lower-level BCS conference teams. A BCS conference team, no matter who they are, will play enough quality opponents that if they go undefeated (and there aren't more than two undefeated BCS conference teams) then they're all but assured of their shot. Boise, Utah, and TCU might have one or even two quality wins per year, but the impact of those games is diluted by playing 8 or 9 games against teams like San Jose State and Idaho.

For an example, look at Northwestern's 1995 Rose Bowl season (it was before the BCS, but it serves our purposes):
They started out unranked, of course, and after beating Notre Dame, dropped a game to Miami of Ohio. But by going undefeated through the Big Ten schedule, even though they missed playing Ohio State, they rose to #3 in the polls before the Rose Bowl. And that was WITH a loss to a MAC team early on. If they hadn't lost to a MAC team, they almost certainly would have been in the top two.

24 Re: Varsity Numbers: The Unexpected

This is generally correct. Both 2007 Kansas and 2006 Rutgers would have played for the BCS title had they won out. In 2009, Cincinnati would have likely done so had Nebraska upset Texas (and if Cincinnati hadn't, a non-AQ team would have).

27 Re: Varsity Numbers: The Unexpected

I agree with this completely. If Minnesota goes undefeated, they would have a legitimate shot at being in the BCS title game. Same with all those lower level teams in a major conference (the Big East is probably excepted from this). There's a big difference between being in a BCS league and not in terms of opportunity.

18 Re: Varsity Numbers: The Unexpected

Part of it of the non-AQ's own making. Look at Boise's 2009 schedule:

Yes, Boise plays in the WAC, but they also ELECTED to play a AA team (UCD), TWO MAC teams (MiamiOH, BG), and Tulsa (they had 13 games b/c of playing @ Hawaii). They could have tried to schedule TCU, Utah, BYU, various Big East teams, etc. Would some have said no? Of course? Would all have said no? I seriously doubt it. Now, in terms of last year's title chase, they still wouldn't have jumped Bama or Texas, but at least if they'd had an OOC of (for instance): Oregon, TCU, BYU, Tulsa, Rutgers; they'd have been taken a bit more seriously.

Similar story for TCU, though to a lesser extent:

They did play AT two ACC schools, but they also played a AA team and SMU. When you're a non-AQ, playing a AA school is virtual suicide for the resume. For that game, they could have (and should have) dialed up other good non-AQ's (maybe Boise could have moved the Fresno game; maybe there was a CUSA team w/ a spot open), or various AQ teams (Oklahoma had recently played them a couple times, Texas had played them once, the BE has LOTS of open schedule spaces, etc.).

Obviously both Boise and TCU are at a structural disadvantage, but they also haven't done everything they could to maximize their resume. At least some of it is, quite frankly, their own fault.

For an even better example, Utah in 2008 ran the table but didn't get a title shot, even though the 2 teams who did had a loss.

Perhaps if they'd played tougher games instead of Utah St and a AA team, an undefeated run would have been taken more seriously. Unfortunately for the, Michigan was way down, but then again, if Michigan hadn't been way down, they might well have lost that game, given how close it ended up being (which is a decent argument for them not deserving the title shot).

28 Re: Varsity Numbers: The Unexpected

" . . . Would all have said no? I seriously doubt it."

And you know this how?

Do you have any idea of how a football schedule is structured at the major college level? Boise's not traveling for free and like many other programs they support other sports on campus.

My understanding is that they will go ANYWHERE for a million dollars (which would be a loss of money for them as they make nearly 2 million per home game) or take a home and home with ANYBODY in the country. Quality teams have the money and would rather give it to FCS teams instead to get a win.

31 Re: Varsity Numbers: The Unexpected

Their current approach is that they'll go anywhere for $1M+ or take a home and home.

FWIW, I have no idea how this meant that it was a good idea to engage in a bunch of home and homes (~ 2 games / year) with the MAC instead of the upper-level non-AQ's. I'm sure this is somehow the BCS's fault, though. Those sly, manipulative bastards!

37 Re: Varsity Numbers: The Unexpected

Well, presumably they'd tell you they didn't have offers from upper-level BCS conference teams. Maybe they're lying, but most BCS conference teams play jokes of nonconference schedules. (Michigan State is my team and they played Western Michigan, Florida Atlantic, Notre Dame and Northern Colorado. They used to at least play ND and another BCS conference team and they'll go back to that in 2013) Presumably, that's on purpose, as to get eight home games and fund the athletic department. It's not in a BCS conference team's interest to play tough non-conference games, so we should probably be happy they do it at all.

5 Re: Varsity Numbers: The Unexpected

I think it's obvious to anyone who's watched the games that VT is a top-15 team hiding behind a bizzare loss to JMU.

10 Re: Varsity Numbers: The Unexpected

The N.C. State game showed me what I needed to see from them -- they were playing a hot, confident team on the road and fell behind late, but still rallied to win. I think they cruise from here on out and go about 7-1 in the ACC. Don't know why they start so poorly, but we forget about the starts when the finish is strong (case in point: Colorado 2001).

16 Re: Varsity Numbers: The Unexpected

Besides the disastrous first quarter of self-inflicted wounds and during Boise's final drive they played Boise well. They came back on the road to beat a good NC State team and handled Central Michigan, East Carolina, and Boston College. Other than a weird loss to JMU, Va Tech looks like a top 15 team.

19 Re: Varsity Numbers: The Unexpected

They do look like they could be a top 15 team, but I don't think the resume demands that rating just yet. The NC St win was good, the JMU loss was very bad, and hanging with Boise was solid for about the 15th rated team. BC apparently sucks, as does CMU, so those wins strike me as not meaning all that much.

So overall, MAYBE they're a top 15 team, but I'm holding judgment at least through the end of their 3-game homestand to see if they've got another bad gag in them.

33 Re: Varsity Numbers: The Unexpected

Oh I completely agree that they don't DESERVE to be in the top 15 right now. I'm saying that IF you treat the JMU loss as some sort of bizzaro-world fluke, all other signs point to VT being a top 15 team. The JMU loss is there and it can't be ignored and it isn't going away. But I think everyone should expect VT to "play like" a top 15 for the rest of the year. I do agree that we need to see 3 wins in the next 3 weeks, and the WF and Duke wins should be fairly convincing, before actually putting top 15 on paper. A decisive win this week probably gets them ranked.

12 Re: Varsity Numbers: The Unexpected

Oregon's win over ASU was as unimpressive in person as it was statistically. Yes, the Ducks scored a whole lot of points in the first three quarters. And yes, their defense has been "opportunistic," which might as well be a euphemism for bad (yet lucky).

Let's face it, though; the Ducks' D was getting shredded drive after drive in the 4th quarter, each of which ended in a horrifically thrown interception by Threet. Those were some *really* bad passes, seemingly unrelated to anything the Ducks' D was doing.

And yes, statistically, the Ducks were outgained by almost 200 yards and only won by 11 points despite getting *seven* turnovers.

14 BCS

The problem I've always had with the BCS is the window dressing. From my perspective, it seems like they simply want to pair up teams arbitrarily: these two in the Ultimate Bowl, these two in the Orange, etc. They seem to think that people might not take too kindly to that (even though that's what happened prior to this system), so they make a show of adding in objective measures to balance things out ... except that they keep jiggering things around to get the results they like, even if the only reason they're making a change is to "prevent" an outcome that may have looked bad due to chance. (The Ultimate Bowl is a blowout, which means that one of the teams didn't belong, which means we have to change the formula.)

The heaviest weights in the system go to subjective systems that can be hysterically inaccurate at times (there are plenty of stories about ballot mistakes if you want to find them). There are some objective components, sure, but they had to be modified to suit the BCS because I don't know why (dropping margin-of-victory adjustments). Yes, I know they gave a reason, to "prevent" teams from running up the score to influence the rankings, but the reason itself is stupid: not only is it a baby-plus-bathwater thing (because those systems aren't allowed to look at scores at all), but it doesn't stop coaches from running up the score to influence voters, who are the most important components of the formula anyway.

And even after all that, it was only supposed to provide us with a 1 vs. 2 matchup in the first place. Of course, to do that, it had to get all the bowls to cooperate, and to make up for moving those teams, the bowls had to accept teams outside their normal tie-ins ... so tradition took a pretty big hit there. (Not that bowls love tradition necessarily: they'll change like nothing if you put enough money on the table. However, some of us do appreciate tradition.)

Of course, if you don't have exactly two teams that stand out above the rest, the BCS can't really help you, but then they use it every year anyway. I feel like before, we didn't really know who was #1, and though we might care, that's just how it was. Now, people try to pretend that we do know who's #1, but most of the time, we still don't.

Is the BCS itself an obstacle to a playoff? Sure, in the sense that a playoff is a process to crown a champion, and the BCS thinks it does that already. The larger obstacle would be the conferences themselves, I think. Yes, a robust playoff (as opposed to the square-dance crap that's been thrown around before - in case you can't tell, I strongly oppose any idea that takes teams from the existing BCS bowl pool) is going to provide additional money to share, but some of that will end up outside the power conferences (quite a bit, if teams like TCU and Boise State are correct), and no one really likes to share.

21 Re: BCS

In reply to by zlionsfan

I think your last paragraph is my main point -- yes, the BCS stands in the way of a playoff. But only because the powers-that-be don't want a playoff. If they wanted a playoff, the BCS wouldn't be able to stand in the way (when the contract ran out, anyway).

15 Just for fun ...

if we took F/+, Sagarin, and Massey, and weighted them all equally, the top 10 would be this:

1. Boise State (.907)
2. TCU (.880)
3. Alabama (.840)
4. Nebraska (.800)
5. Oregon (.773)
5. Ohio State (.773)
7. Stanford (.760)
8. LSU (.747)
9. Arizona (.680)
10. South Carolina (.587)

1.000 would mean each system rated that team #1. I used their actual systems (as you might guess from my post above) and not the BCS-neutered ones.

Doesn't necessarily mean much, but it's a fun way to spend time ...

17 Re: Varsity Numbers: The Unexpected

The BCS is an attempt to implement a 1v2 game (itself a playoff, and by far the easiest one to achieve with maximum popular support and minimal disruption to the current profitable and generally beneficial system), but in determining the participants in the 1v2 game they (read: at the behest of then-SEC Commissioner Roy Kramer, I believe) devised a system taking into account the polls and the famous computer rating systems (as opposed to the good computer ratings, see Billingsley). Since then, any time teams other than AP#1 and AP#2 have met in the BCSCG, there's been a huge outcry (read: by the AP voters) as to how this is wrong, and the selection formula has been adjusted to "fix" the "problem" and guarantee a future BCSCG in the same circumstances will match AP#1 and AP#2. We've achieved a state of relative balance (one tilted IMO much too heavily toward the human polls) in this regard, at least until we get a result where AP#1 and AP#2 don't meet in the BCSCG (which could've happened in 2006 w/r/t UF v UM without pollster vote manipulation), but the fundamental issue is that the generally assumed correct system is created by people who aren't particularly good at picking the best team in a consistent and intelligent fashion, don't realize and/or care that they aren't particularly good at that because gosh darn it that's how it just to be, and have a very large soapbox.

Like Bill, I regard the BCS as far, far short of the ideal, but from a broad-scale perspective relatively close to the best possible system given the existing political interests of the teams and major conferences and the unwillingness to risk changing what has historically been important in CFB. I'll be reading Death to the BCS when the library gets it, but judging by the promotional stuff I've seen I'd bet it's a lot of (i) caterwauling about the big bowls buying off teams and preventing a playoff, (ii) how going to non-marquee bowls can be uneconomical for teams, (iii) glorious projections about how much more money a playoff would make in TV revenue, and (iv) caterwauling about how the marquee teams and conferences are screwing over other teams.

25 Re: Varsity Numbers: The Unexpected

The projections may be glorious, but there is also a decent chance that they are reasonably accurate, given the demand for football content on television. A Conference Championship Saturday, followed by one or two playoff Saturdays in December (I'd favor one), followed by semifinals at two traditional BCS bowl sites on January 1st, followed a week later by The Big Game, all sold together, and perhaps divvied up among the major networks, would likely command an enormous amount of money. It's the one reason I have some small optimism that it may happen. Somebody like Comcast is eventually going to wave such a huge bag of money that even the people who run college football will take notice.

32 Re: Varsity Numbers: The Unexpected

We'll see a playoff at some point after the top 40-80 programs walk out on either 1-A or the NCAA as a whole. It's ridiculous to waste a playoff spot and the cash that goes with it on the MAC/Sun Belt, but in the current 1-A setup it would be almost impossible not to give that to them.

Moreover, any playoff implementation would inevitably create even more pressure to pay players (longer season, more classroom disruption, WAY more cash floating around CFB, etc.), and moving in that direction would devastate the finances of most of the non-AQ's.

As long as major college football has far too many teams and leagues, I don't think there's any reasonable shot of seeing a playoff.

29 Re: Varsity Numbers: The Unexpected

My worthless, humanities driven $.02 on the failure of the BCS.

I realize that the following puts me on dangerously thin ice at FO, but fantasy leagues notwithstanding, statistically determined outcomes are only so relevant to the average sports fan. I enjoy my time at FO because you guys have tailored your statistics to be highly relevant to the sport. I invariably learn something useful about teams I may not have had the chance to see play. But there is not a single power rating, efficiency index, poll, projection, or amalgamation of the above that will ever replace what happens when two teams suit up and decide matters between themselves on the field of play.

At the end of the average CFB season, whatever numbers get kicked out, however the polls have shaken down, we are left with eight teams that can compete with each other at a very high level. (Let's just pretend that that UH/ UGA Sugar Bowl didn't happen.) Under the BCS we get to see these games played, but they amount to the most tantric experience in American sports. Imagine the NFL playoffs ending after the wild card round, with the two teams who compiled the best regular season records being given a bye and a free pass to the Super Bowl. Every year, even in years with fairly undisputed #1s and #2s, it is painfully obvious that the opportunity for teams to play exciting, relevant, championship caliber games to close the year is being lost. If you are a real CFB fan, and not just a homer for a perennial championship school, this never fails to be demoralizing.

We are all familiar with the apples and oranges nature of this comparison, but it's the elephant in the room of every BCS discussion. In college basketball, how much does the public actually demand a #1 vs. #2 matchup? Is there an armed revolt and a ratings drop when a #1 seed goes down in March? Butler can earn national respect in CBB, but most discussions of Boise or TCU's right to play for a championship start sounding like Tea Party rallies. So much of this "need" in CFB to see a preordained #1 vs. #2 is manufactured by never having had the chance to see the alternative in action, that I feel true CFB fans HAVE to demand better, whether we ever get it or not. The people behind the BCS know that once a playoff is instated, their numbers game is over, because no one will look back.

Finally, I can not stand the way the BCS inflates the value of an undefeated season. These are twenty year old kids playing an amazingly demanding sport. Win-loss perfection is great, but anyone who has been around sports for a length of time will tell you that perfection is a goal, rarely a reality. Teams grow and mature over the course of a season. I would have no problem with a team who lost in September winning the championship in January. The BCS doesn't make this impossible, but it puts a stigma on losing a game, ever, that makes me nauseous.

30 Re: Varsity Numbers: The Unexpected

The BCS does a good job getting two of the top four or five teams to play each other in a bowl each year. The problem is that they claim to get the top two teams, and that's impossible to truly assess without either massive scheduling reform or consistently having years like 2005 where the top two teams are obvious (note that getting Texas and USC in a 1 versus 2 matchup for the title is something that wouldn't have happened preBCS; USC would have played Penn State in the Rose Bowl while Texas got a matchup with somebody like West Virginia or Virginia Tech.)

It doesn't help that the BCS foolish depends on formulas to pick 1 vs 2, when the difference between teams is sufficiently subtle enough to be a bit of an art. In some sense, the BCS might give better results if they had a Selection Committee pick teams like in basketball (with some requirement that they look at things like polls, records, and strength of schedule.)

That said, I'm fine with 11 conference champions+5 wild cards, with a Selection Committee making the wild card selections.

34 Re: Varsity Numbers: The Unexpected


LOL even more @ Nebraska


Men are more ready to repay an injury than a benefit because gratitude is a burden and revenge a pleasure.

36 Re: Varsity Numbers: The Unexpected

Michigan has an offense but that team is still poorly coached in a lot of spots. It may be entertaining but as a long term strategy winning games 45-42 is ill-advised. Because if your offense makes one or more mistakes you get a game like yesterday. Or the prior week.

It's pretty interesting that NW/MSU is shaping up as a big game. The Wildcats are decent to good so Sparty should have its hands full. I find it incredible that Fitz cannot find a better field goal kicker. The little Greek kid is a train wreck.

I feel bad for Tim Brewster. Wisconsin and Iowa have raided the top in-state talent now for what seems like a decade and getting kids from North Dakota is small consolation.

Save 10%
& Support Bill
Support Football Outsiders' independent media and . Use promo code WRITERS to save 10% on any FO+ membership and give half the cost of your membership to tip Bill.