Firing Coaches and Performance

The BBC has a nice piece on whether statistically it matters (in terms of performance) if the head coach (manager) is fired (sacked).  I wonder if there is any similarities in US sports?

NCAA Athletic Department Revenue Inequality

Last year I looked at NCAA FBS bowl game payout inequality to bowl participants, and found that there is a substantial level of payout inequality.  Now I am turning my attention to income (or revenue) inequality at FBS and non-FBS schools that are in USA Today's NCAA athletic department database.  As I stated yesterday, not all schools are in the USA Today's database.

Later I will investigate the Equity in Athletics data.  The advantage is that I can include more schools and look at total revenue by sport, but for now, I using USA Today's database since I am interested in revenue inequality among various revenue streams in athletic departments, not solely total revenue at athletic departments.

Economist's measure income inequality using a Gini coefficient.  The Gini coefficient measures the total difference between perfect income equality and the actual income distribution which is called a Lorenze curve.  For those interested - here's a step-by-step guide to calculate the Gini coefficient.

Given we have an income inequality measure and NCAA athletic department income, let's take a look at the various athletic department's revenue streams in the USA Today database from 2006 to 2011 by FBS and non-FBS schools.  The table below has the Gini coefficients.  As I expected FBS schools in general have greater revenue inequality than non-FBS schools except for ticket sales for each of the seasons during this time period and contributions in 2007, 2009 and 2011.  Otherwise FBS schools have greater income inequality than non-FBS schools.

		Gini		Gini
2006		FBS		Non-FBS
Ticket Sales		0.4988		0.5693
Student Fees		0.5708		0.4753
School Funds		0.5310		0.4361
Contributions		0.5852		0.5315
Rights & Licensing		0.4390		0.3919
OtherRevenue		0.3882		0.3637
Total Revenue		0.3492		0.2540
Generated Revenue		0.4685		0.3809
		Gini		Gini
2007		FBS		Non-FBS
Ticket Sales		0.5067		0.5610
Student Fees		0.5641		0.4882
School Funds		0.5137		0.4441
Contributions		0.4971		0.5398
Rights & Licensing		0.4474		0.3790
OtherRevenue		0.3750		0.3629
Total Revenue		0.3293		0.2601
Generated Revenue		0.4469		0.3747
		Gini		Gini
2008		FBS		Non-FBS
Ticket Sales		0.4964		0.5818
Student Fees		0.5767		0.4917
School Funds		0.5108		0.4299
Contributions		0.5165		0.5121
Rights & Licensing		0.4525		0.3710
OtherRevenue		0.4046		0.3273
Total Revenue		0.3333		0.2525
Generated Revenue		0.4503		0.3648
		Gini		Gini
2009		FBS		Non-FBS
Ticket Sales		0.5061		0.5704
Student Fees		0.5832		0.5102
School Funds		0.4983		0.4037
Contributions		0.5108		0.5631
Rights & Licensing		0.4465		0.3617
OtherRevenue		0.4946		0.3212
Total Revenue		0.3219		0.2496
Generated Revenue		0.4485		0.3640
		Gini		Gini
2010		FBS		Non-FBS
Ticket Sales		0.5074		0.5771
Student Fees		0.5817		0.5127
School Funds		0.5361		0.3992
Contributions		0.5394		0.5295
Rights & Licensing		0.4538		0.3539
OtherRevenue		0.4071		0.2982
Total Revenue		0.3380		0.2433
Generated Revenue		0.4602		0.3491
		Gini		Gini
2011		FBS		Non-FBS
Ticket Sales		0.5131		0.5687
Student Fees		0.5901		0.5071
School Funds		0.5449		0.4017
Contributions		0.4962		0.5309
Rights & Licensing		0.4544		0.3377
OtherRevenue		0.4073		0.2891
Total Revenue		0.3250		0.2400
Generated Revenue		0.4521		0.3415

NCAA Athletic Department Revenues

Today I am looking at the financial side of NCAA athletic departments using USA Today's NCAA athletic department database from 2006 to 2011.  USA Today reports ticket sales revenue, student fee revenue, school fund revenue, contribution revenue, rights & licensing revenue, other revenue and total revenue during this time period.

Please note that the USA Today NCAA athletic department database does not cover all FBS or non-FBS schools.  One of the reasons is that private schools like Notre Dame and Stanford do not have to publish their financial athletic department data like most public athletic departments.

I am interested in whether two different types of athletic departments - athletic departments in the Football Bowl Subdivision (FBS) and athletic departments outside of the Football Bowl Subdivision (non-FBS) - have different levels of revenue among other things.  Thus I have separated the data into these two groups for the analysis below.

Here are a few questions (and answers) I want to investigate about NCAA athletic departments covered in the USA Today database.

What is the average revenue at FBS and non-FBS athletic departments for the 2011 season?

First, I have added one more revenue variable to the USA Today NCAA Athletic Department Financial Database - Generate Revenue.  Generated Revenue is the amount of revenue that athletic department create by creating athletic events as opposed to revenue that athletic departments receive by being part of a college or university.  In the latter case, schools transfer funds from one part of the university to the athletic department independent of economic activity produced by the athletic department.  Economist usually refer to this as a subsidy.  For my calculations, the NCAA athletic department subsidy is equal to Student Fees plus School Funds.  Thus Generated Revenue = Total Revenue minus Student Fees minus School Funds.

Focusing on the most recent time period in the database, we see that average revenue for these various revenue streams are quite a bit different for the two types of NCAA schools.

2011	FBS		non-FBS
Ticket Sales	$12,866,227		$651,601
Student Fees	$4,053,653		$4,368,336
School Funds	$5,843,586		$4,997,387
Contributions	$11,511,685		$959,824
Rights & Licensing	$16,337,623		$1,028,285
Other Revenue	$4,125,049		$943,802
Total Revenue	$54,737,824		$12,949,236
Generated Revenue	$44,429,462		$3,583,512

How much are athletic departments subsidized at FBS and non-FBS athletic departments?

As you can see in the table above, for the two revenue streams that are non-generated (or a subsidy):  student fees and school funds have similar average revenues between FBS and non-FBS schools.  Combined FBS schools receive an average of $9,897,240 of non-generated revenues and non-FBS schools receive an average of $9,365,724 of non-generated revenues.

Which (if any) revenues streams at FBS and non-FBS athletic departments are significantly different from each other?

The numbers in the table above are average revenues for each revenue stream for FBS and non-FBS schools in the USA Today database.  While many of the averages between FBS and non-FBS schools are different from each other, what I want to know is which (if any) revenue streams are different in a statistical sense.  To do that, I will use a t-test assuming that FBS and non-FBS average revenues have unequal variances, which is called a Welch's t test.

Running the Welch's t test, I find that all of the generated revenues are statistically different from each other in FBS and non-FBS schools at the 99% confidence level.  This result holds for all generated revenue streams (Ticket Sales, Contributions, Rights & Licensing, Other Revenue, Total Revenue - including the subsidies, and Generated Revenue - the sum of Ticket Sales, Contributions, Rights & Licensing and Other Revenue) since 2006!  FBS and non-FBS schools in the USA Today database average generated revenues are different from each other at the 99% confidence level!

With respect to the non-generated revenues (Student Fees and School Funds) I find that BOTH of the non-generated revenues are NOT statistically different from each other between FBS and non-FBS schools at the 95% confidence level and this same result holds since 2006 between the two groups of athletic departments.

Athlete Markets

Fantex has recently started a market in professional athletes, where one can buy stock in a player's future earnings.  As we note in our book The Wages of Wins, professional athletes tend to be "consistently inconsistent" in terms of on-field performance, but in terms of salaries they tend to be rather more consistent - generally.  One of the problems with this model is that NFL players do not have guaranteed contracts like NBA players, so this might be one of the incentives that NFL players like Arian Foster have to sell off part of their future earnings.

Fantex plans on setting up a trading market where current holders of a players stock can sell to anyone willing to buy that players stock - just like any other stock exchange.  I guess next will be futures options (calls and puts) and then derivatives.

Fantex signs a second athlete - Vernon Davis.

2013 NCAA FBS Top 25 Ranking for Week 10

We have a new number one team in the Top 25 NCAA FBS Ranking in terms of productivity using the Complex Invasion College Football Production Model - Oregon has jumped over Baylor to take the top spot, but it is still rather close.

Some surprises with respect to the conventional wisdom: Texas A&M, Missouri, and LSU are all ranked ahead of Alabama; Wisconsin's has dropped with a win over Iowa; ditto with Michigan State defeating Michigan; and Northern Illinois, Ball State and Utah State are ranked above Clemson in terms of productivity.

These NCAA FBS Top 25 rankings are based on regression analysis using the data posted at College Football Statistics based on the Complex Invasion College Football Production Model.

Rank	Team
1	Oregon
2	Baylor
3	Florida State
4	Ohio State
5	Louisville
6	Texas A&M
7	Missouri
8	LSU
9	Alabama
10	Northern Illinois
11	Houston
12	Wisconsin
13	Ball State
14	Utah State
15	Michigan State
16	Arizona State
17	Auburn
18	Cincinnati
19	Clemson
20	Marshall
21	Oregon State
22	Miami (Florida)
23	Washington
24	Fresno State
25	UCLA

Previous Top 25 Ranks for 2013
2013 NCAA FBS Top 25 Ranking for Week 9
2013 NCAA FBS Top 25 Ranking for Week 8
2013 NCAA FBS Top 25 Ranking for Week 7
2013 NCAA FBS Top 25 Ranking for Week 6
2013 NCAA FBS Top 25 Ranking for Week 5
2013 NCAA FBS Top 25 Ranking for Week 4
2013 NCAA FBS Top 25 Ranking for Week 3
2013 NCAA FBS Top 25 Ranking for Week 2

MLB Payroll Tax for 2013

The USA Today reports that only the New York Yankees and Los Angeles Dodgers are subject to MLB's payroll tax for this past season.  Also as you read farther in the article, the author gives a lot of anecdotal evidence as to the limited link between payroll and performance - a subject that Dave, Marty and I have written about.

Speculation on Houston Nutt Coaching at UConn

According to College Football Talk, UConn is interested in Houston Nutt as the new head football coach. So given there has been some interest among readers of this blog about the head coaching situation at UConn, I thought that I would at a look at the Mississippi (or Ole Miss) Rebels while Houston Nutt was the head coach.  Unfortunately the data that I use only goes back to the 2008 season, so this is the exact time that coach Nutt was with the Mississippi Rebels, but does not cover his entire coaching career.  Thus with that caveat on the table, here is the last four seasons of Houston Nutt's coaching career in reverse chronological order using the Complex Invasion College Football Production Model.

2011
In Houston Nutt's last season as head football coach with the Rebels, Ole Miss finished 2-10.  Mississippi was the #115 (out of 120) ranked NCAA FBS team in overall production, with the #117 ranked offense and the #107 ranked defense.  Mississippi's best victory was a 38-28 victory over #92 ranked Fresno State University and the Rebel's worst performance was a 13-30 loss to #102 ranked Kentucky Wildcats.  The Rebels played against a "tougher" strength of schedule (SOS) of 49.33 given the Rebels schedule was between one and two standard deviations below the "league" SOS of 63.24.

2010
The Rebels finished the regular season at 4-8.  Mississippi was the #79 ranked team in terms of overall productivity; with the #74 ranked offense and the #86 ranked defense.  The Rebels best game was a 35-28 victory of #59 ranked Kentucky and their worst game was a 48-49 loss to FCS Jacksonville State.  In terms of their schedule, Mississippi played against a SOS of 59.00, which is "average" as compared to the "league" average SOS of 63.10.

2009
Mississippi again finished the regular season at 8-4 and bowl eligible defeating #42 Oklahoma State to finish the season at 9-4.  The Rebel's best win was over #22 ranked Tennessee and of their worst loss was to #87 Mississippi State all against a SOS that was average compared to the "league".  In terms of overall production, Mississippi was the #29 ranked team with the #47 ranked offense and the #17 ranked defense.

2008 
In Houston Nutt's first season at the helm of the Mississippi Rebels they finished the regular season bowl eligible at 8-4 and won their bowl game over #10 ranked Texas Tech to finish with a winning record at 9-4.  The Rebels were the #18 ranked team in overall productivity with the #26 ranked offense and the #22 ranked defense, while playing against an average SOS as compared to the "league".  The Rebels best performance was against #1 ranked Florida in the end of season rankings and their worst loss was to #67 ranked South Carolina the following week.

As you can see there were some very good years under coach Nutt.  Whether that was due to the hand that he inherited or was due to his coaching skill is an open question.  The dramatic fall of the team does raise some questions on his coaching resume and a serious question for an administration looking to hire a new head coach.  Yet, the positives are there and for a team that was willing to take a chance on a well known individual, this might be a chance worth taking.  In today's coaching carousel, given the UConn's program current state a few years with a coach that has the skills that Nutt has may not be that great of a risk.

Here is my an analysis of UConn up the the firing of coach Pasquolini.