How to Pick College Football Bowl Games

As promised, here is my wrapup column with all the lessons I learned throughout bowl season. Timely, I know. This guide would have been extremely helpful a month ago.

However, if I don’t write these down, I won’t remember them next season. This way, I’ll only have to remember to look at my blog archives. I will almost certainly forget to do that, but there’s always a chance that I remember.

Some of these are rules that we were all well aware of before the games, but were emphatically reassured with this season’s games. Others are somewhat new thoughts. In no particular order, here are eight lessons I learned from this year’s bowl slate:

1. Do not bet against the SEC in the National Championship Game

Yeah, we already knew this one. SEC teams had won four straight BCS championships heading into this season. But Auburn really hammered this point home with their victory over Oregon.

In many ways, this was the SEC’s biggest challenge to their supremacy. The previous four titles came against the Big 12 and Big Ten. This was the Pac-10’s first crack at the SEC. The Pac-10 is widely viewed as the second fastest conference after the SEC, and Oregon dominated that conference like no team since USC six years ago. Though Auburn finished undefeated, they were viewed as one of the weaker SEC champions of recent times because of their propensity to do juuuust enough to win games. If a team was going to end the SEC’s reign of dominance, this was the year.

It was not to be. Oregon only lost by a field goal, but the difference was apparent. Auburn and Oregon were both fast, but only Auburn was big and fast. From now on, don’t pick against the SEC in the championship game – wait until another conference shows that they can compete with the best the SEC has to offer.

2. Trust the good SEC teams, but not the average SEC teams

The five best SEC teams to make bowl games (Auburn, Arkansas, LSU, Alabama, and Mississippi State) went a combined 4-1 with 3 blowouts. The next five teams (South Carolina, Florida, Georgia, Tennessee, and Kentucky) went 1-4. Last year, SEC teams that finished 8-4 or better went 3-1 in bowl games; teams 7-5 or worse went 3-3.

Between 1998 and 2008, SEC teams that entered bowl games with 8-4 records or better went 31-24 (.564). Teams with 7-4 or worse records went a 18-10 (.643). Way too small of a sample size? Absolutely. But it’s worth keeping an eye on. For many years, the strength of the conference was in the middle – the conference won only two of the first eight BCS titles. Since then, the conference has become top-heavy with the best schools getting the best recruits and coaches. It would surprise no one if the SEC became a conference of haves and have-nots like the Big 12 or Big Ten.

If so, watch out for the 6-6 and 7-5 schools in bowl games. If 6-6 SEC teams continue lose to mediocre teams from the Big East, Conference USA, and ACC, you might as well throw the “always bet on the SEC” rule goes out the door.

#3. Watch out for unmotivated teams

My full post on this is here. In that post, I identified the eight games this season with an unmotivated team that was favored by a touchdown or more. The games resulted in four blowouts for the favored team and four straight-up wins for the underdog.

In my original post, I compared these games to the 5 vs. 12 games in the NCAA Basketball Tournament. Often in these games, you have an underdog that enters the game with something to prove versus a favorite that finds themselves in a less desirable bowl game because of a poor finish to the season. The lesson: beware of putting high confidence values on these teams in pools. It’s a pretty empty feeling when you know within the first five minutes that a team you put a high value on didn’t bother to show up for the game.

#4. Conference USA sucks

I know, I know, Central Florida beat Georgia after I swore up-and-down that a Conference USA team couldn’t beat a BCS conference team. In my Military Bowl preview, I pointed out that since 2005 Conference USA teams were 0-10 against BCS conference teams in bowls (0-11 after East Carolina was pummeled by Maryland).

Finally, C-USA champion Central Florida ended that streak with a 10-6 win over Georgia in the Liberty Bowl. Call me crazy, but I’m not impressed. I watched parts of the Liberty Bowl and I don’t think I’ve ever seen a team care about winning as little as 6-6 Georgia did in that game. And the C-USA champion still barely slipped by.

SMU, Southern Miss, East Carolina, and UTEP all showed how overmatched mediocre teams from C-USA are against mediocre teams from other conferences. It might be okay to bet on a really good C-USA team, but resist the urge to pick 7-5 and 6-6 teams. And I’m well aware that Tulsa destroyed Hawaii, which brings me to my next point…

#5. Don’t trust Hawaii at home in bowl games

In this post, I listed Hawaii’s performance at home in bowl games since 1999. Long story short: you never know what you’re going to get with the Warriors at home in bowl seasons. They win games that they’re not supposed to and lose games that they’re heavy favorites in.

#6. Do stick with teams you like in the season and don’t over-think match-ups

If I made a list of teams that I really liked after the season, it would have included these teams: Stanford (maybe the best team in the country by season’s end); LSU (they just win baby); Boise State (drastically undervalued after losing to a very good Nevada team on the road); Notre Dame (ended season on a hot streak); and Nevada (see Boise State). The list of teams I didn’t like included: Kansas State and Georgia (both burned me late in the season); Michigan (0-8 ATS in their last eight games); Nebraska (peaked way too soon); and South Carolina (insanely overrated based on win over Alabama).

Then the matchups got in the way. I picked all five of those teams I liked, but for various reasons, only put 29, 17, 26, 18, and 24 confidence points on them because I was scared of their matchups. Amazingly, I picked Kansas State (28), Georgia (30), and Nebraska (34) for waaaay too many points because I thought they fell into favorable matchups. I also only picked a solid Florida State team for 14 against the South Carolina team that I thought was overrated. In the Michigan game, I did pick Mississippi State for 31, so at least I followed my instincts once.

The moral of the story? Stick with teams you like and teams you don’t like. Don’t let unfavorable matchups sway you from teams that you liked during the season. And definitely don’t let favorable matchups trick you into picking teams that just aren’t playing all that well.

7. Sun Belt > MAC

For some reason, the NCAA likes pairing up these two conferences against each other. Maybe it’s a conspiracy to keep these also-rans away from other teams. Whatever the reasons, I’m sure the bowls that have to host these teams absolutely love it.

The standard theory is that the Sun Belt is the worst conference in the country. This is wrong – the MAC is worse. Even picking up the scraps that SEC and ACC teams leave behind in the fertile recruiting territory in the South, Sun Belt teams are still way faster than MAC teams. Sun Belt and MAC teams have met in bowls five times in the last three years. Each time, the MAC team had the better record. They have gone 2-3. Their only two wins were 11-2 MAC champion Central Michigan over 9-3 Troy in double overtime last year and 9-4 MAC champion Miami over 6-6 Middle Tennessee this year.

The very best MAC teams might be better than the top Sun Belt teams. But if the team’s records are within a game of each other, trust the speed of the Sun Belt.

8. Some coaches get their teams up for bowl games; others fail miserably

The Missouri/Iowa game was a good microcosm of this theory. I read an argument on a message board on this game. Angry poster #1 argued that Iowa’s Kirk Ferentz gets his guys up for bowl games and Missouri’s Gary Pinkel struggles to do the same. Without doing any research, this seemed correct based on my memory. Then angry poster #2 pointed out that, before this game, Ferentz was 5-3 in bowl games and Pinkel was 4-3. Well great, now I didn’t know what to think.

Turns out, angry poster #1 was emphatically right. Of Ferentz’s three losses, only one was a blowout (the 2002 Orange Bowl to a USC team that would win the next two national titles). The other two were to Florida in 2005 (31-24; Florida won the national title the next year) and to Texas in 2006 (26-24; Iowa was 6-6 and Texas was 10-2). Pinkel’s three losses came in a 35-13 beatdown against Navy last year, a 39-38 loss to Oregon State in the 2006 Sun Bowl, and a 27-14 loss to Arkansas in the 2003 Independence Bowl. Really, the only bowl game that Missouri has looked impressive in is when an 11-2 Tiger team dominated an 8-4 Arkansas team 38-7 in the 2007 Cotton Bowl.

Records aside, not all bowl performances are created equal. In retrospect, I might still have picked Missouri, but not for 23 confidence points against a team that has historically always been ready to play. Same with Fresno State; as I pointed out here, Pat Hill has struggled getting his team ready for bowl games. For some reason, I picked them anyway. I shouldn’t have been surprised when Northern Illinois beat them 40-17 in a game that wasn’t even as close as the score indicates.

Cornerbacks and the NFL Draft

Sometimes I have ideas for posts that simply don’t pan out when I do my research. Generally, I’ll come up with a theory that turns out to be wildly wrong, so I scrap the whole thing.

That’s what happened on this post. I had a theory that cornerbacks drafted in the first round tend to be busts a high percentage of the time and the best cornerbacks are drafted later in the draft. This turned out to be incorrect (sorta), but for reasons that were entirely unexpected by me. The research turned out to be pretty interesting though, so I decided to post it anyway.

I had the idea for this post right about the time that Packers cornerback Tramon Williams made the second half of Saturday night’s Packers/Falcons a mere formality with a devastating TAINT on the last play of the first half. For those not well-versed in Packers defensive backs, Williams was an undrafted rookie out of Louisiana Tech in 2007. He has become the best cornerback that the Packers have developed since they drafted Mike McKenzie in 1999. It’s not that they haven’t tried either – since 1999, they have drafted 15 cornerbacks. Two are still with the team – Brandon Underwood (drafted last year, primarily a special teams player) and Pat Lee (drafted in 2008, the dime back and part-time punt returner). Their top three cornerbacks are Charles Woodson (first rounder in 1998, came to Packers in 2004 via trade), Williams, and Sam Shields (an undrafted rookie).

My theory was that, judging from the Packers’ experience, it just isn’t worth it to draft a cornerback in the first couple of rounds. The short answer? I was wrong. The long answer? I might be right for a completely different reason.

****************

Here are the cornerbacks that were drafted in the first round between 1998 and 2008:

1998 (1 Pro Bowler, 2 average players, 1 bust) –
Charles Woodson (4) – Future Hall of Famer made the Pro Bowl in his first four seasons, was adequate for six seasons, then made four more Pro Bowls between 2008 and 2011.
Duane Starks (10) – Intercepted 25 passes and was a solid player for four teams in ten-year NFL career.
Terry Fair (20) – Lasted only four seasons before crashing out of the league.
R.W. McQuarters (28) – Intercepted 14 passes and was a decent player for four teams in eleven-year NFL career.

1999 (2 Pro Bowlers, 2 semi-busts) –
Champ Bailey (7) – 9-time Pro Bowler has 48 interceptions and is still one of the premier corners in the league.
Chris McAllister (10) – 3-time Pro Bowler was a shutdown corner for Ravens in eleven-year career.
Antuan Edwards (25) – Played for five teams in lackluster seven-year career.
Fernando Bryant (26) – Played for four teams in slightly less lackluster nine-year career.

2000 (1 Pro Bowler, 2 busts) –
Deltha O’Neal (15) – 2-time Pro Bowler. Had 34 interceptions in nine years, but was out of the league after the 2008 season.
Rashard Anderson (23) – Lasted only two seasons in the league due to a combination of substance abuse and lack of talent.
Ahmed Plummer (24) – Decent player for four years before injuries forced him to retire in 2005 after playing in only nine games in the previous two seasons.

2001 (1 Pro Bowler, 1 above average player, 2 busts) –
Nate Clements (21) – 1 Pro Bowl. Very good player, but 49ers questionably made him highest paid defensive player in history in 2007.
Will Allen (22) – No Pro Bowls, but has reputation as one of the best corner covers in league. Currently playing with Dolphins.
Willie Middlebrooks (24) – started one game in five years. Currently plays in Canada.
Jamar Fletcher (26) – played for five teams in eight years. Out of the league by age 29.

2002 (1 Pro Bowler, 2 average players, 1 bust) –
Quentin Jammer (5) – No Pro Bowls, only 14 interceptions in nine years with the Chargers.
Phillip Buchanon (17) – No Pro Bowls, 20 interceptions. Has played for five teams; was released outright twice.
Lito Sheppard (26) – 2 Pro Bowls, 1 All-Pro. Carved out a solid career playing for Eagles, Jets, and Vikings.
Mike Rumph (27) – Was called the worst corner in the league before the 49ers mercifully moved him to safety. Still lasted only three healthy seasons in league.

2003 (3 Pro Bowlers, 2 busts) –
Terence Newman (5) – 2 Pro Bowls, 26 interceptions. Has spent entire career as starter for Cowboys.
Marcus Trufant (11) – 1 Pro Bowl, 20 interceptions. Has spent entire career with Seahawks.
Andre Woolfolk (28) – Rarely played in four seasons with Titans. Out of league by 2006.
Sammy Davis (30) – Rarely played in five seasons with three teams. Out of league by 2007.
Nnamdi Asomugha (31) – 4 Pro Bowls. Premier shutdown corner in NFL. Amazingly, the ball was thrown his way only 27 times in 14 games in 2010. Signed to largest contract for cornerback in NFL history.

2004 (1 Pro Bowler, 2 above average players, 1 bust) –
DeAngelo Hall (8) – 3 Pro Bowls, 32 interceptions in seven seasons.
Dunta Robinson (10) – Was a starter for six seasons for Texans. In 2010, Falcons signed him to second largest contract ever for a cornerback.
Ahmad Carroll (25) – Weaknesses included not being able to cover receivers. Played parts of four seasons with three teams. Last seen as practice squad member of UFL’s Hartford Colonials.
Chris Gamble (28) – Has started at cornerback for last seven seasons for Panthers and grabbed 24 interceptions. One of the highest paid defensive players in NFL.

2005 (1 average player, 3 busts plus 1 bust at cornerback position) –
Pacman Jones (6) – Weaknesses included sobriety and not getting arrested. Actually had two decent seasons for Titans in 2005 and 2006. Last seen as member of Bengals, displaying little of the athleticism that made him the sixth overall pick.
Antrel Rolle (8) – Injuries and lack of coverage ability cost Rolle to lose his starting job after three seasons. Converted to safety in 2008, and has made two Pro Bowls since then.
Carlos Rogers (9) – Solid six-year starter for Redskins.
Fabian Washington (23) – Really fast, but not all that good. Has been mainly a backup for career.
Marlin Jackson (29) – A poor man’s Antrel Rolle. Mainly a backup cornerback, he has also filled in at safety at times, where he has done well. Eagles signed him in 2010 to be a safety, but a ruptured Achilles ended his season in June.

2006 (1 Pro Bowler, 1 average player, 2 busts) –
Tye Hill (15) – Part-time player has played for four teams in five years. Was waived by Titans before 2010 season.
Antonio Cromartie (19) – 1 All-Pro, 1 Pro Bowl, 18 interceptions in five seasons.
Johnathan Joseph (24) – Solid, if unspectacular five year starter with Bengals.
Kelly Jennings (31) – Still hanging around with Seahawks, despite being only a nickel or dime back.

2007 (1 Pro Bowler, 1 above average player, 1 TBD) –
Darrelle Revis (14) – 3-time Pro Bowler. Along with Asomugha, has reputation as premier shutdown corner in the league.
Leon Hall (18) – No Pro Bowls, but has been a standout corner with Bengals. Has 18 interceptions in just four seasons.
Aaron Ross (20) – Decent, injury-prone cornerback with Giants. Has started 24 games in four seasons.

2008 (2 Pro Bowlers, 1 above average, 2 TBD) –
Leodis McKelvin (11) – A good return man, but has been largely ineffective at corner in three seasons.
Dominique Rodgers-Cromartie (16) – A standout with 13 interceptions. Selected to 2010 and 2011 Pro Bowls.
Aqib Talib (20) – Another standout, has grabbed 15 interceptions in three seasons.
Mike Jenkins (25) – Shutdown corner for Cowboys was selected to 2010 Pro Bowl, although he seemingly took a step backwards in 2010 season.
Antoine Cason (27) – Intercepted four passes in 2010 season, his first as a starter.

****************

For those counting at home, that’s 14 Pro Bowl cornerbacks, 16 busts, and only eleven that fell in between (I excluded Ross, McKelvin, and Cason because each has spent only one full season as a starter so far).

Generally, if a team has an almost 40% chance of essentially wasting a draft pick if they draft a player from a certain position, you’d think they’d stay away. But clearly teams are swayed by the 35% chance of hitting a Pro Bowl pick. And there’s very little in-between – they either waste a first round pick for a bust or get a Pro Bowl cornerback.

Six cornerbacks were selected to the Pro Bowl this season; five were drafted in the first round (fourth round pick Asante Samuel was the lone non-first rounder). That was a higher percentage than any position other than inside linebacker (all four are first-rounders).* It is hard to fault teams for looking for another one of these Pro Bowl cornerbacks, but I can’t help but think that using a first round draft pick on a 50/50 shot is something that the Detroit Lions would do.

* I quickly looked up all linebackers (the NFL doesn’t differentiate between OLB and ILB in  the draft, and I was too lazy to break them down myself) from 1998 to 2008. The result: 16 Pro Bowlers, 13 average/TBD, and just 6 busts. And this is at a position significantly more likely to face career-ending injuries than cornerbacks. Needless to say, this is a way better success rate than cornerbacks.

So what explains the results of first round cornerbacks? I have a few theories. I suspect it’s a combination of one or more of these:

1. It’s extremely difficult to identify first-round talent for the cornerback position. College receivers are just that much easier to defend. This year’s top CB prospect, LSU’s Patrick Peterson, faced the #2 WR prospect Julio Jones of Alabama…and that’s it for receivers projected to go in the draft. The second best CB prospect, Nebraska’s Prince Amukamara, faced exactly zero receivers projected to go in the draft.

More than any other defensive position, a cornerback faces a series of one-on-one matchups. Good college linebackers and safeties have to make reads each play. Good defensive lineman face double teams if they are dominant players. Good college cornerbacks, on the other hand, simply aren’t targeted. The offensive team will gladly leave the cornerback to cover a receiver and make the game 10-on-10. There’s a reason that Darrelle Revis’s nickname is “Revis Island” after all. It’s hard to figure out just how well a cornerback’s skills will translate to the pros if they don’t get the reps against the best receivers.

2. NFL teams look at the wrong numbers. The most important attribute for a pro cornerback is instinct. And yet NFL teams still fall for cornerbacks who perform well at the draft combine. Nebraska’s Fabian Washington was drafted in the first round by the Raiders in 2005. Washington was a solid cornerback for the Huskers, but no Husker fan realistically viewed him as a standout. Then he ran a 4.25 40 at the combine, the fastest time for any player that year, and jumped 41 1/2 inches in the vertical jump. Somewhat predictably, he didn’t crack the starting lineup for the Raiders and was traded to the Ravens in 2008. He started for parts of 2008 and 2009 before the Ravens benched him for good early in 2010.

Nnamdi Asomugha is probably the best shutdown cornerback in the league. He ran a comparatively slow 4.45 and jumped only 37 1/2 inches. Closing speed helps, but it doesn’t help that much. Yet since that’s all teams can measure in the combine, they fall for guys like Washington’s pure speed and neglect guys like Asomugha’s “quickness.”

3. Teams don’t know how to develop cornerbacks. For reasons that aren’t really clear to me, teams tend to thrust rookie cornerbacks right into the starting lineup. Because of the 1-on-1 nature of the position, a rookie cornerback is extremely easy to exploit. They also standout far worse when they have bad games; it’s easier to point to the one cornerback covering the 200-yard receiver than it is to point out just one of the seven lineman and linebackers charged with stopping the 200-yard rusher.

Fans jump on the rookie cornerback, and his confidence is shot. The most famous example of this is Broncos’ rookie Roc Alexander, who was charged with covering the Colts’ Reggie Wayne in the 2005 NFL Playoffs. Colts quarterback Peyton Manning mercilessly picked on Alexander. Wayne finished with 10 catches for 221 yards; Alexander played one more season with the Broncos and was out of the league for good by 2007.

Other than quarterback, cornerback is probably the hardest position for a rookie to play for all the reasons that I described above. And yet teams mostly refuse to keep a roster spot for a rookie backup cornerback. On top of that, only 18 of 32 NFL teams even bothered to keep a cornerback on their practice squad this season. Wouldn’t it be a good idea to line up opposite an NFL receiver every week in practice, instead of being thrown to the wolves each week in a game?

Look no further than Asomugha for support. He sat on the bench for the better part of his first two seasons in the NFL. Now he’s one of the best cornerbacks in the league. Sure, Revis started right off the bat and developed into a premier shutdown corner. But for every Revis, there’s an Ahmad Carroll – the Packers first round pick in 2004 that was such a train wreck that he was abruptly cut four weeks into the 2006 season.

4. Reputation. I think this plays a small part in how long a cornerback stays in the league. Cornerbacks are a lot like offensive lineman in that we tend to notice them only when they do something bad. Occasionally a cornerback will come up with a good interception, but more likely, you’ll only notice them if they get burned on a long pass or get an illegal contact or pass interference penalty called on them. Revis and Asomugha combined for zero interceptions this year, yet we know them as the best shutdown corners in the league. Every week, we see that the receiver that they cover is out of the picture. Like an offensive lineman who doesn’t allow sacks, we just know that they are good.

But this can also turn into a self-fulfilling prophecy. Quarterbacks see that other teams don’t throw towards those good corners, so they don’t throw their way either. But the cornerbacks that get burned once or twice a game…well, those corners will be getting picked on. Just like that, the gulf between good cornerbacks and bad cornerbacks grows wider, simply because that’s who quarterbacks choose to target.

*****************

The schism between Pro Bowl and bust cornerbacks is probably a combination of those four theories. Nebraska’s Prince Amukamara is projected to go in the top five of the draft. I watched every Husker game this year and I couldn’t tell you if he’ll be a Pro Bowler or a bust. He was great for Nebraska, but it wasn’t so much for what he did – he had zero interceptions – but for what other teams didn’t do against him. Who knows if that will cut it against far more talented quarterbacks and receivers.

So what does this mean for NFL teams that draft a cornerback in the first round? Simply, beware: you have a 50/50 shot at getting a franchise cornerback and a 50/50 shot of wasting your first round draft pick.

If your team has a lot of needs, it’s probably not worth it to draft a cornerback. Your fans will remember. Trust me, Packers fans remember the name Ahmad Carroll. And not for good reasons.