Sunday, January 28, 2007

More on Returning Quarterbacks

A few days ago, I took a look at how teams who returned their starting (or at least experienced)quarterback performed compared to teams that did not in 2006. First I looked at overall records. However, those can be misleading for several reasons. First and foremost, all teams played at least 12 regular season games in 2006. Teams that played in bowl games could play up to 14 games. The most games any team played in 2005 was 13 games. Often times this extra game is a non-conference cup cake against a Sun Belt or non-Division IA school. To make the comparison more fair, this time we will look at each team's record only in conference games. With the big time conferences having as many as 12 teams, this is not totally unbiased because some teams can still have substantially easier conference slates based on since they do not play everyone in the conference. Still, 2006 is a good year to examine this phenomenon because no teams changed conferences between 2005 and 2006, keeping ot such lurking variables as conference strength. For example, it would not be fair to look at compare conference records between 2004 and 2005 because a number of teams changed conferences (Boston College went from a watered down Big East to the ACC, while Cincinnati, Louisville, and South Florida went from Conference USA to a BCS league, TCU went from Conference USA to the Mountain West, SMU, Rice, UTEP, and Tulsa went from the WAC to Conference USA, and Central Florida and Marshall went from the MAC to Conference USA). I think that covers them all. Some final notes: Conference title games are excluded so that teams from the same conference all have the same number of games played, and 4 independents (Notre Dame, Army, Navy, and Temple) are not included in this data set. With that out of the way, here is how the two groups of teams performed.

The teams (62 total) that returned an experienced quarterback (at least 100 pass attempts the previous season) and had him play significantly (at least 100 pass attempts this season) in 2006 went a collective 238-248 in conference play in 2005. This is a winning percentage of .490 and equates to just under a 4-4 record (3.91-4.09) in a standard 8 game conference season. In 2006, those same teams improved to 269-223 in conference play. This is a winning percentage of .547 and equates to a 4.37-3.63 record in a standard 8 game conference season. This is an improvement of roughly 1/2 game in the conference standings.

The teams (53 total) that did not return an experienced quarterback in 2006 went a collective 214-204 in conference play in 2005. This is a winning percentage of .512 and equates to a conference record of 4.10-3.90 in a standard 8 game conference season. In 2006, those same teams regressed to 188-234 in conference play. This is a winning percentage of .445 and equates to a conference record of 3.56-4.44 in a standard 8 game conference season. This is a regression of a little more than 1/2 game in the conference standings.

21 teams (33.9%) that returned an experienced quarterback improved by at least 2 games in the conference standings. 8 teams (12.9%) improved by at least 3 games in the conference standings. 13 teams (21%) declined by at least 2 games in the conference standings. 5 teams (8.1%) declined by at least 3 games in the conference standings.

10 teams (18.9%) that did not return an experienced quarterback improved by at least 2 games in the conference standings. 5 teams (9.4%) improved by at least 3 games in the conference standings. 21 teams (39.6%) declined by at least 2 games in the conference standings. 10 teams (18.9%) declined by at least 3 games in the conference standings.

Those are the results. Dissect them at your leisure. I will say this, it appears that it may not be as valuable to return your starting quarterback (12.9% that returned theirs improved by at least 3 games and 9.4% that did not improved by at least 3 games) as it is damaging to have him leave (more than double the chance--18.9% to 8.1% of declining by at least 3 games). As promised, in a few more days, I'll look at how BCS and non-BCS schools compare when they return and lose their quarterback.

Wednesday, January 24, 2007

How Important is the Quarterback Position in College Football?

Way back in August, I made a bet, or a wager if you will, with a friend of mine. Now my friend happens to be a huge Clemson supporter. Our bet was simple: If Clemson won 10 games, I would be forced to pay up, and if Clemson did not win 10 games, he would be forced to pay up. My main rationale for going against the Tigers despite a favorable schedule, was that they would be breaking in a new starting quarterback. After manning the position for what seemed like a decade, Charlie Whitehurst had been drafted by the San Diego Chargers and a senior named Will Proctor would be taking over. Clemson began the season 7-1, but imploded down the stretch thanks in part to the lackluster play of their signal caller. Certainly the defense and special teams were also to blame, but Proctor played poorly in two of the losses (Virginia Tech and Maryland) and for over 3 quarters in the bowl game against Kentucky.

What I want to do is see what happens to teams that break in a new starting quarterback. To do this I divided all 119 Division IA teams into two groups. Those that returned 'experienced' quarterbacks and those that did not. I defined an experienced quarterback as one that threw at least 100 passes in the preceding season (2005). Additionally, those quarterbacks also had to have significant playing time in 2006, lest their experience just rot on the bench. So teams that are coded as 'experienced' had the same quarterback throw at least 100 passes in both 2005 and 2006. I know 100 is just an arbitrary cutoff point, but it had to be somewhere. 63 teams returned experienced quarterbacks this past season and 56 did not. Here's what happened.

Teams with a returning experienced quarterback had a collective record of 375-357 (.512) in 2005. When their experienced quarterbacks returned in 2006, their combined record jumped to 469-337 (.582). That's an increase of roughly 7 percentage points in winning percentage.

Teams who lost their quartebacks after 2005 had a collective record of 341-309 (.525) in 2005. When they lost their quarterbacks, they regressed to a combined 316-384 (.451) in 2006. That's a decrease of roughly 7.4 percentage points in winning percentage. Interesting how the gain in winning percentage among teams that returned their quarterback is almost equal to the losses in winning percentage of teams that lost their quarterback.

Here's more fuel to the fire that returning your quarterback is very important. Random fluctuation is common in sports, so there is not a great deal of difference in a team going 7-5 one season and 6-6 the next, at least in the quality of play. However, for teams that return an experienced quarterback, there is a very good chance they will improve by at least two games. 23 of the 63 teams (36.5%) improved by at least two games in 2006. To be fair, nearly a fourth, 14 of the 63 teams (22.2%) regressed by more than 1 game (at least one and a half games). Since schedule length varies in college football with conference title games, bowl games, and the 'extra' Hawaii games, an example for a one and a half game difference would be going 8-4 one year and 7-6 the next (1 in the win column and 2 in the loss column).

For teams that had to start over at the quarterback position, only 9 of 56 (16%) improved by more than two games in 2006. Plus a whopping 25 of the 56 teams (44.6%), or more than double the percentage for teams that returned an experienced quarterback, declined by more than 1 game.

That's all for now. Stay tuned unil this weekend or early next week, when we'll look at conference record (perhaps a better indicator of team strength with no Texas State cupcakes) for teams that do and do not return experienced quarterbacks as well as a comparision in what happens to BCS and mid-major schools when they lose quarterbacking experience.

Saturday, January 13, 2007

Ranking the SEC Coaches

1-2-3-4. We lost to Wake Forest by 24.

I got dressed all by myself today.

Recently my friend and I got into a discussion about college coaches and we soon realized how loaded the SEC is with quality coaches. Every coach in the league with the exception of the above pictured gentlemen have a pretty good track record when it comes to coaching em' up, either at their current school or somewhere in the recent past. Now to be fair, both Ed Orgeron and Sylvester Croom stepped into tough situations at their respective universities. And if you believe what you read, Orgeron is a great recruiter so maybe Ole Miss will rise up in the near future. Croom's teams have improved (at least in regard to point differential) every season so maybe the Bulldog faithful have hope too. However, since those two have the least impressive credentials, I'm leaving them out of the discussion. Below the other ten coaches in the SEC are listed alphabetically along with their resumes.

Rich Brooks
22 seasons
Record: 106-141-4 .430
Brooks’ winning percentage is not that good, but he has taken Oregon and Kentucky (two programs that were not exactly elite when he got there) to 5 combined bowl games, going 2-3. He also won a conference title at Oregon in 1994, the schools first conference title in 37 years.

Phillip Fulmer
14.5 seasons
Record: 137-41 .770
Fulmer has had only one losing season in his 14+ seasons in Knoxville. He has won a national title (1998) as well as two conference titles and 4 SEC Eastern Division titles. He has taken the Vols to 14 bowl games going 7-7.

Bobby Johnson
13 seasons
Record: 73-77 .487
Johnson has made the Vanderbilt program competitive. His teams were a combined 6-29 his first three seasons, but have made strides the past two seasons going 5-6 (narrowly missing out on a bowl) and 4-8. Two seasons ago they beat Tennessee for the first time since Reconstruction and this past year they upset Georgia in Athens. Johnson was 58-34 (.630) in 8 seasons at Furman. He won one conference title, made 4 playoff appearances, and participated in the national championship game in his final season (2001).

Urban Meyer
6 seasons
Record: 61-12 .836
Meyer just won the national championship at Florida in only his second year. His teams have never lost more than 3 games in any one season. He has taken his teams to 4 bowl games, never losing a single one. He has 3 conference titles and the first ever BCS bowl appearance by a program outside the BCS leagues.

Les Miles
6 seasons
Record: 50-25 .667
Miles spent 4 years at Oklahoma State rebuilding that program and just finished his second season at LSU maintaining that program’s status as one of college football’s elite. His only losing season was his first in Stillwater. His teams have gone to 5 bowl games, winning 3, and he has an SEC Western Division title.

Houston Nutt
14 seasons
Record: 102-67 .604
Nutt has taken the Razorbacks to 7 bowl games in 9 years, winning two of them, and also taking two SEC Western Division titles. Prior to coming to Fayetteville, he spent 4 seasons at Murray State amassing two playoff appearances. He also spent one year (1997) at Boise State, coaxing the Broncos to a 4-7 record in only their second year of play in Division IA.

Mark Richt
6 seasons
Record: 61-17 .782
Richt has never had a losing season at Georgia nor has he lost more than 4 games in any one season. The Dawgs have gone to 6 bowl games (4-2 record), and have 3 SEC Eastern Division tiles to go along with 2 wins in the SEC Championship Game.

Nick Saban
11 seasons
Record: 91-42-1 .683
Little Nicky has never had a losing season and has coached his teams 9 bowl games where they have posted a 4-5 record. He has 2 SEC Western Division titles and 2 wins in the SEC Championship game. He also won a MAC Championship in his only year at Toledo. Oh and he won it all in 2003.

Steve Spurrier
17 seasons
Record: 157-50 .756
Spurrier has had but one losing season in his career (his first year at Duke). He also won a conference championship at Duke to go along with the 6 he won at Florida. His teams have gone to 14 bowl games winning 8. He also won Florida’s first (and until this season only national title) in 1996.

Tommy Tuberville
12 seasons
Record: 96-49 .662
Tuberville has had just two losing seasons in his 12 years roaming the sidelines at Ole Miss and Auburn. His teams have participated in 8 bowl games, winning 5. He has 2 SEC Western Division titles and one win in the SEC Championship Game. He also had an undefeated season in 2004, but did not get to participate in the BCS title game (Orange Bowl) because there were 3 undefeateds that season. He’s 6-2 against archrival Alabama (including 5 straight wins). And his record since 2000 is 66-23 (.742).

Now, I'm leaving it up to you. Who is the SEC's best coach? Use any definition of coaching you want (getting the best out of what you have, best X's and O's guy, best winning percentage, most wins, best Under Armor commercials, etc.) and let me know how you would rank these 10 gentlemen.

Tuesday, January 09, 2007

Championship Game Fall-Out

Well, that went about as expected. Not even the most ardent UF supporter could have seen that coming. If you introduced one of your foreign friends to this game and told them before hand that one team had the Heisman winner, one of the nation's top defense, a very good big game coach, and an undefeated record while the other team had a somewhat shaky quarterback, an offense that couldn't generate much offense against good teams, a very good defense themselves, and a good record built up by several close wins here's guessing they would have thought Florida was the undefeated team and Chris Leak was the Heisman winner. Ohio State would have been the team with the shaky offense, and Troy Smith would have been, well Reggie Ball? Smith's final stat line read 4-14 for 35 yards and one pick. In backyard football, those numbers will only net you two first downs.

Sporting events like this are good because they let you know a few things. First, you can never know everything there is to know about football. Every day you learn just a bit more. Secondly, statsitcis can only tell you so much. Every stat in the book predicted an OSU win, if not a blowout. Yet, the Gators were the team that thoroughly dominated--and yes they dominated, the Buckeyes. They held OSU to 82 total yards and one offensive touchdown. This wasn't a New England Patriots/St. Louis Rams upset where the underdog held on for dear life to win down the stretch. No the Gators dominated from just after the opening kickoff. You would have made a ton of money laying 27 points for the Gators about 16 seconds into the game after Ted Ginn Jr. ran the opening kick back.

Why didn't we see this coming? Florida barely survived Vanderbilt and South Carolina, yet here they were marching up and down the field on Ohio State while stifling their vaunted offense. Everyone knew the Gators came in with a superb defense, but to suggest beforehand they would humble the Buckeyes in this manner was unimagineable. Was it the month Urban Meyer had to prepare that made the Gators so unstoppable? Was it Tressel's refusal to adjust his defensive strategy? The Buckeye defensive backs were hardly ever in the picture when Florida snapped the ball allowing countless short completion after short sompletion. Was it motivation? Did Ohio State get full of its own hubris? Were the Gators bristling from all the disrespect that they were unbeatable? Probably a combination of all these things and a few others. The important thing is that although this site is called Statistically Speaking, sometimes statistics just get it wrong. I saw the Gators as a flawed offensive team that squeaked by several mediocre at best foes. And the Gators may still yet have to pay the piper for all their close wins. I saw the Buckeyes as a great team that obliterated everyone in their path. Statistics aren't perfect, just a tool for making rational decisions. Picking Ohio State to win this game was a very rational decision, but as they say: That's why you play the game.

Monday, January 08, 2007

Fies--Err BCS Championship Game Preview

Record: 21-10

BCS Championship Game: Florida versus Ohio State

So after folks have had two months to talk themselves into a Gators upset, reality will set in about 9:25 EST tonight. Does Florida have a chance? Sure they do. I'd say probably 30-35%. But seriously, lets look at the facts. The Gators scored 17 against a South Carolina team that couldn't stop Houston. They scored 21 on Florida State. Wake Forest scored 30 against the Noles. Chris Leak will certainly throw at least two picks off his back foot, the game will be close at halftime, and the Buckeyes will roll in the second half.

Sunday, January 07, 2007

GMAC Bowl Preview

Record: 20-10

GMAC Bowl: Ohio versus Southern Miss

The GMAC Bowl will serve as a nice ending to a day filled with football. After two do-or-die NFL playoff games, fans can sit back and enjoy an exhibition. And if the game sucks, and you end up going to bed early? So what, you have work tomorrow.

The work that Frank Solich has done at Ohio in just his second season is outstanding. Though he has been overshadowed by Jim Grobe, Greg Schiano, and even Dick Tomey, its time to give the man his due. Unjustly fired from Nebraska three seasons ago, Solich could actually finish this season with a better record (10-4) than Bill Callahan (9-4). To do so, the Bobcats will have to defeat Conference USA runner-up Southern Miss.

In a flip-flop more egregious than those of John Kerry or even Jay Mariotti—here’s why Ohio doesn’t stand a chance. The Bobcats are 9-4, but they only do three things particularly well—return punts, cover them, and cover kickoffs (12th nationally in returns, 6th in covering punts, and 3rd in covering kickoffs). They are also 34th in kickoff returns which is good but not great. They averaged only 4.02 yards per rush and Ohio quarterbacks managed only 8 touchdown passes against 13 interceptions on the season. Defensively, opponents averaged 4.10 yards per rush against Ohio and threw 14 touchdown passes against 12 interceptions. Without those stellar special teams, Ohio probably wins 5 or 6 games.

Southern Miss is pretty good offensively (4.54 yards per rush to go along with 14 touchdown passes and just 6 interceptions) and defensively (4.01 yards allowed per rush with 13 touchdown passes allowed and 15 interceptions). The Eagles are average in covering punts (50th in the country) and pretty bad at covering kickoffs (117th). Similarly, they are average in returning punts (49th) and a touch worse at returning kickoffs (79th). Still, their advantage on offense and defense should provide them with a substantial edge.

Prediction: Ohio can only win this game if they return a considerable portion of the Golden Eagles punt and kickoffs back a long way, don’t commit any turnovers on offense, and get at least three for themselves from Southern Miss. Ohio has had a nice season, but will finish the year on a two-game losing streak.

Saturday, January 06, 2007

International Bowl Preview

Record: 20-9

International Bowl: Cincinnati versus Western Michigan

For those of you who don’t know, there are two more bowl games to be played before the BCS Championship Game on Monday night. On of them is Saturday at Noon in the International Bowl. If you’re a fan of the Broncos or Bearcats, or if you’re hankering for an upset check this one out.

The story of this game is the Western Michigan Broncos and their coach Bill Cubit, who has proven that he is no zirconium in his two seasons at the school. The Broncos were 1-10 in 2004, but Cubit has gone an outstanding 15-8 in his two plus seasons on the job. The Broncos win games with their passing attack and their defense. Senior quarterback Ryan Cubit completed over 62% of his passes and while throwing 15 touchdowns and 11 interceptions. Defensively the Broncos are led by Ameer Ismail and his 17 sacks which oh by the way led the nation. With Ismail, the Broncos were very stout on defense. They held opponents to only 2.49 yards per rush and also played effectively against the pass (12 touchdowns and 22 interceptions against).

The Bearcats were very average on offense, something that does not bode well when facing the defense of Western Michigan. They averaged only 3.36 yards per rush and threw the same number of interceptions (15) as touchdowns. Defensively, the Bearcats put up very similar numbers. Opponents averaged 3.69 yards per rush and threw 12 touchdown passes against 12 interceptions.

Cincinnati also went 1-4 away from home this season (though to be fair the 4 losses were to Ohio State, Virginia Tech, Louisville, and West Virginia). On the other hand, Western Michigan was able to go a respectable 3-4 on the road including winning at Virginia and playing Florida State down to the wire. So how is Cincy a 7 point favorite? Beats me. They have a new coach and seem to about even with Western Michigan when you adjust for the strength of their respective schedules. The only aspect of the game Western Michigan should be a little nervous about is punt coverage. The Broncos were 98th in the nation allowing 11.31 yards per return (and 3 touchdowns). However, Cincinnati was only 91st in the nation at returning kickoffs.

Prediction: The Broncos are about even with Cincinnati in terms of team strength plus they don’t have to deal with a new coach for the bowl game. Brian Kelly is a great hire for Cincy, but he will not win his debut.