CrispAds Blog Ads

Statistically Speaking: I Feel Lucky

Tuesday, January 31, 2006

I Feel Lucky

After spending a few days in Las Vegas, I thought it would be interesting to look at which BCS football teams were the 'luckiest' and 'unluckiest' last season. To do this, I will examine which teams performed substantially different in conference play based on their expected record from their points scored and allowed. I am only using data in conference games because some blowouts over non-conference cupcakes can distort a team's point differential and prevent us from getting a true view of their abilities. A previous post explains the Pythagorean Theorem in detail (it should be noted that I am using 2.37 as the exponent now instead of 2 as it is more accurate with a smaller sample size of data).

First up, here are the teams that were 'luckiest' last season based on difference in actual and expected conference record, as well as brief commentary on each team.

UCLA went: 6-2, expected record: 3.43-4.57, difference +2.57 wins
UCLA was actually outscored in conference play, but went 4-0 in conference games decided by 4 points or less. They also added a a close 10 point win over Arizona State to their ledger. In their 2 conference losses, they were beaten by 38 (Arizona) and 47 (Southern Cal). Karl Dorrell may have coached one of the worst 10 win teams of all time. With Maurice Drew (early entry) and Drew Olson (graduation) leaving, the 2006 incarnation of the Bruins may fall on hard times.

South Carolina went: 5-3, expected record: 3.54-4.46, difference +1.46 wins
Another team with a winning conference record that was actually outscored in conference play. To be sure, Spurrier's first season was a success, but the Cocks were far from a dominant team. Like UCLA they had a good record in close games (4-1 in conference games decided by 8 points or less) and got thumped twice (by 23 to Alabama and 41 to Auburn). South Carolina was downright lucky in 2005, and any change for the worse in their fortuitiveness could mean the difference in a bowl bid and another losing season. On a side note, if Brad Scott had gotten outcoached by Gary Pinkel like that, he would've been hung in effigy. Why has Spurrier gotten off scott-free?

Oregon went: 7-1, expected: 5.71-2.29, difference + 1.29 wins
The term lucky may not be apropos for Oregon. The Ducks did slip by several teams late in the year (Arizona by 7, Cal by 7, and Washington State by 3), but that was only after quarterback Kellen Clemens was injured and lost for the year in the Arizona game. With Clemens, the Ducks were a legitimate top-10 team, and deserved a BCS bid. Without him, they were top-15, maybe top-20 (remember, they only lost to Oklahoma by 3).

Georgia Tech went: 5-3, expected: 3.76-4.24, difference +1.24 wins
The Jackets were one of the most sporadic teams in college football in 2005, winning at Auburn and Miami, and losing to Utah in their bowl game. Like their predecesors on this list, the Jackets had a good record in close conference games (3-1 in games decided by 7 points or less), and they were also blown out at least once (by 44 to Virginia Tech). Tech has 4 straight seasons of exactly 7 wins and between 5 and 6 losses. Expect more of the same in 2006.

Kansas went: 3-5, expected: 1.86-6.14, difference +1.14 wins
The Jayhakws are anomally on this list. They went only 1-0 in conference games decided by 7 points or less. Their disparity in actual and expected wins stems from two blowouts, one to Colorado (31 points) and one to Texas (52 points). They were relatively competitive in their other losses, and if the defense continues to remain solid could contend for the Big 12 North title next season.

North Carolina went: 4-4, expected: 2.90-5.1, difference +1.10 wins
The Heels went 4-2 in conference games decided by 7 points or less. Their other 2 losses were pretty convincing, but hardly thumpings (18 to Miami and 27 to Virginia Tech). The Heels rough non-conference slate (Wisconsin, Utah, and Louisville) kept them from being bowl eligible where they would made a solid representative.

Northwestern went: 5-3, expected: 3.98-4.02, difference +1.02 wins
As usual, when the 'Cats go bowling, they win a lot of close games. They were 3-1 in conference games decided by 7 points or less. In their other recent bowl years of 2000, 1996, and 1995, the 'Cats were a combined 10-0 in conference games decided by 7 points or less. Northwestern did return to a bowl after their magical 1995 season, but following their luck-induced 1996 and 2000 seasons, they slipped back to 5-7 and 4-7 respectively. Expect a repeat of this backsliding in 2006.

Now here are the Frank Grimes of the BCS conferences, the most unlucky teams of 2005.

Arkansas went: 2-6, expected: 4.11-3.89, difference -2.11 wins
The Hogs were 0-4 in conference games decided by 7 points or less. With returning quarterback Casey Dick and running back Darren McFadden only sophomores, the Hogs could be surprise contenders in the SEC West next year.

Washington State went: 1-7, expected: 2.84-5.16, difference -1.84 wins
The Cougars were 1-5 in conference games decided by 7 points or less. Washington State will be an interesting team to watch next season. They lose 1900 yard rusher Jerome Harrison, but quarterback Alex Brink, who improved as the conference season wore on could have the Cougars on the brink of a bowl bid next season.

Michigan State went: 2-6, expected: 3.80-5.20, difference -1.80 wins
If a season truly can turn on one play, then it happened to Michigan State last season. The blocked field goal that was returned for a touchdown by Ohio State turned that game around, and possibly destroyed Michigan State's season. The Spartans were 0-2 in conference games decided by 7 points or less. They whipped Illinois and Indiana, and were whipped by Northwestern and Minnesota. The future for the Spartans, as always, is very cloudy. Always one of the more inconsistent teams, don't expect a truly breakout season, until it happens.

Iowa State went: 4-4, expected: 5.70-2.30, difference -1.70 wins
For the second season in a row, the Cyclones blew their chance at a Big 12 North Division title by losing in their final regular season game. In 2004, it was a 3 point home loss to Missouri. In 2005, it was a 3 point road loss at Kansas. Overall, Iowa State was 0-3 in conference games decided by 7 points or less. After beginning conference play 0-3, the Cyclones reeled off 4 straight double-digit victories and were poised to be the sacrificial lamb for Texas in the Big 12 Championship Game. The finale against Kansas ended those hopes, but Iowa State played well in their bowl game (another close loss) against a top-15 TCU team. Iowa State will return their top passer (Bret Meyer), rusher (Stevie Hicks), and receiver (Todd Blythe), so another shot at the Big 12 North is within reach.

Clemson went: 4-4, expected: 5.45-2.55, difference -1.45 wins
The Tigers went 1-4 in conference games decided by 7 points or less. As they are want to do, they got hot at the end of the season, concluding with a 4 game winning streak. Clemson does lose quarterback Charlie Whitehurst, and the road schedule is pretty brutal (BC, Florida State, and Virginia Tech), but if they can win the majority of their home games, they could have a special season.

Tennessee went: 3-5, expected: 4.30-3.70, difference -1.30 wins
Tennessee went from being one of the luckiest teams in 2004 (6-0 in conference games decided by 7 points or less) to one of the unluckiest teams in 2005 (1-3 in conference games decided by 7 points or less). Did they forget how to 'win' close games? Probably not. More than likely, a few bounces that went their way in 2004, went the other way in 2005. The Vols will be back in a bowl game in 2006.

Purdue went: 3-5, expected: 4.22-3.78, difference -1.22 wins
Purdue was 1-2 in conference games decided by 7 points or less. Their disparity between actual and expected record can be explained by the fact that 2 of their conference wins were blowouts, and none of their conference losses was by more than 18 points. Surprisingly, Purdue had 3 players from their 5-6 squad opt for early entry into the NFL draft. Were I a betting man, I'd say Joe Tiller rights the Boiler's ship.

Cal went: 4-4, expected: 5.13-2.87, difference -1.13 wins
Cal went 1-3 in conference games decided by 7 points or less. In their other 4 games, they blew out Washington, Arizona, and Stanford, and were dismantled by Southern Cal. Jeff Tedford's team had a solid season despite losing starting quarterback Nathan Longshore for the entire season in the first game against Sacramento State. They should return to the upper-eschelon of the Pac 10 next season.

2 Comments:

Blogger Kevin McCarrell said...

I would argue with you about Spurrier getting outcoached. Spurrier's offense put up what should have been enough points on the board, although the three turnovers hurt us.
But in the end, it was the defense's inability to stop Brad Smith that killed us. Yes, Spurrier is head coach and is ultimately responsible for both sides of the ball, but you can't say Pinkel "outcoached" him. Pinkel just let Brad Smith do what he does.

12:00 PM  
Blogger matt said...

True, the offense did score enough to win. However, they only scored 3 points in the 2nd half. Although it probably had a bit to do with regression to the mean (SC played so well offensively in the 1st half they were bound to come back to Earth), coaching probably also played a role. Pinkel allowed Brad Smith to improvise in the second half, and after scoring only 7 first half points, they scored 24 in the 2nd. If any coach not named Spurrier (with the exception of maybe Bryant or Stagg) had lost a game in that fashion, they would have heard it from the media.

11:06 AM  

Post a Comment

<< Home

Free Website Counter
Free Website Counter