Monday, October 16, 2017

2017 NCAA FBS Top 25 Ranking for Week 7

Below is the Top 25 ranked teams using the Complex Invasion College Football Production Model from the data provided at www.cfbstats.com.  According to the Complex Invasion College Football production model, Alabama is back on top as the most productive team in all of the Football Bowl Subdivision as of the end of last week. Links to the previous weeks rankings for this season are at the bottom of the page.

Rank Team
1 Alabama
2 Ohio State
3 Washington
4 Penn State
5 South Florida
6 Wisconsin
7 Oklahoma State
8 UCF
9 North Carolina State
10 Oklahoma
11 TCU
12 Miami (Florida)
13 Auburn
14 Georgia
15 Texas Tech
16 Michigan
17 Southern Mississippi
18 Stanford
19 USC
20 Minnesota
21 Texas A&M
22 Clemson
23 Arizona
24 Iowa State
25 Virginia

Previous 2017 Top 25 Rankings
2017 NCAA FBS Top 25 Rankings for Week #6
2017 NCAA FBS Top 25 Rankings for Week #5
2017 NCAA FBS Top 25 Rankings for Week #4
2017 NCAA FBS Top 25 Rankings for Week #3
2017 NCAA FBS Top 25 Rankings for Week #2

Wednesday, October 11, 2017

2017 MLB Pay and Performance

Last week I blogged about competitive balance in MLB.  Now I want to turn to the relationship between MLB payroll and MLB performance.  I have looked at this using USA Today MLB team payroll data in the past (2015 and 2016), but now I want to use data from spotrac.com for MLB.  I find that the data from spotrac is excellent and probably more accurate than the USA Today MLB team payroll data.  The data from spotrac starts in 2011, so I have a smaller time frame to work with, but I thought it would be good to see if the relationship between payroll and performance differs using a different set of MLB team payroll data.  If you are interested in doing this type of analysis, here is a step-by-step guide to payroll and performance analysis.

So, after collecting and organizing the 2011 through 2017 MLB team payroll and team performance data, I ran the regression on team performance using relative payroll as the independent variable.  I use a software package called Stata for the statistical analysis and found that relative payroll is positive and statistically significant with respect to MLB team performance, just like I have found in each previous time I have done this type of statistical analysis using data from USA Today.  The coefficient on relative payroll is 0.06868, which means that a one unit increase in relative payroll will lead to a 0.06868 increase in team winning percent.  In terms of wins, the regression results show that a one unit increase in relative payroll would result in just over 11 additional wins.  In order to get those wins, a MLB team would have to increase team payroll by one unit of relative payroll, which is the same thing as average payroll.  For the 2017 MLB season, that is an increase of $152,327,084 using data from spotrac.  So those approximately 11 wins would cost about $13,690,875 per win.  That is a rather costly amount to pay for each win.

Finally, the regression results show that while the relationship between MLB team performance and relative payroll is positive and statistically significant, the amount that relative payroll "explains" of MLB team performance is not that much.  In this case, using data from spotrac, the explanatory power of the regression is rather weak, with an adjusted r-squared equal to 0.127.  Another way of saying that is that relative payroll only "explains" 12.7% of MLB team performance.  I will let you decide if you think that is good enough.  But will state that there are other variables that explain a greater amount of MLB team performance than relative payroll.

Tuesday, October 10, 2017

Gary Andersen Departs as Head Football Coach at Oregon State

Gary Andersen has departed as head football coach at Oregon State University yesterday.  Andersen was under contract until 2021, but decided it was best for the program to leave now, with the Beavers currently 1-5, their only victory over FCS Portland State in their second game of the season.  So, here is a look at Oregon State since Andersen's arrival.

Below is a chart of offense, defense and total production of the Oregon State Beavers football program during Andersen's tenure as head football coach, along with who would be the lowest ranked team during this time period (in purple) and the average team (sky blue).   Oregon State has been below average each season under his tenure. All rankings in this blog come from my Complex Invasion College Football Production Model.  More details about the program under Andersen are after the chart below as well as a link to his contract.


Gary Andersen [2015 - 2017*]

2015
In head coach Andersen's first year at the helm of the Beavers football team, Oregon State  finished the regular season overall at 2-10 (bowl ineligible).  Oregon State played against a "tougher" strength of schedule (SOS) as compared to the "league" average SOS, meaning that Oregon State's SOS was between one and two standard deviations lower than the "league" average SOS.  The Beavers best game again was their victory over #60 ranked San Jose State (35-21), and their worst loss was to #87 ranked Colorado (13-17).  Oregon State had the #120 ranked team in total production with the #113 ranked offense and the #113 ranked defense from the Complex Invasion College Football Production Model.

2016
At the end of the regular season the Beavers were again bowl ineligible with a 4-8 win/loss record, while playing against an "average" strength of schedule (SOS) as compared to the "league" average SOS, meaning that UTEP had an SOS plus or minus one standard deviation of the "league" average SOS.  The Miners best win was over #98 ranked California (47-44) and their worst loss was to #70 ranked UCLA by a score of (24-38).  Oregon State had the #95 ranked team in total production with the #90 ranked offense and the #74 ranked defense from the Complex Invasion College Football Production Model.

2017
Near the halfway mark of the regular season the Beavers are 1-5, while playing against a "much tougher" strength of schedule (SOS) as compared to the "league" average SOS, meaning that Oregon State's SOS is lower than two standard deviations of the "league" SOS.  The Beavers only victory to date was over FCS Portland State and their worst loss was to currently ranked #34 Minnesota by a score of (14-48).  At the time of Andersen's departure, Oregon State had the #123 ranked team in total production (that is the worst team), with the #116 ranked offense and the #118 ranked defense from the Complex Invasion College Football Production Model

Monday, October 9, 2017

2017 NCAA FBS Top 25 Ranking for Week 6

Below is the Top 25 ranked teams using the Complex Invasion College Football Production Model from the data provided at www.cfbstats.com.  According to the Complex Invasion College Football production model, Washington is now the most productive team in all of the Football Bowl Subdivision as of the end of last week. Links to the previous weeks rankings for this season are at the bottom of the page.

Rank Team
1 Washington
2 Alabama
3 Ohio State
4 Penn State
5 Washington State
6 Georgia
7 Clemson
8 Notre Dame
9 Wisconsin
10 Oklahoma
11 UCF
12 Virginia Tech
13 Auburn
14 Louisville
15 Wake Forest
16 Miami (Florida)
17 San Diego State
18 Oregon
19 TCU
20 Texas Tech
21 West Virginia
22 Michigan State
23 USC
24 Colorado State
25 Kansas State

Previous 2017 Top 25 Rankings
2017 NCAA FBS Top 25 Rankings for Week #5
2017 NCAA FBS Top 25 Rankings for Week #4
2017 NCAA FBS Top 25 Rankings for Week #3
2017 NCAA FBS Top 25 Rankings for Week #2

Friday, October 6, 2017

2017 MLB Competitive Balance

With the 2017 MLB regular season finished, let's take a look at competitive balance using the Noll-Scully measure of competitive balance.  The Noll-Scully uses the actual standard deviation of a league's winning percentage and compares it to a league if wins and losses were randomly determined in a statistical sense.  A Noll-Scully of 1.000 indicates wins and losses were randomly determined and thus the league is perfectly balanced in a competitive sense.  A Noll-Scully higher indicates that the league is less than perfectly balanced.  As the value increases the league is less competitive.

For the 2017 MLB regular season, we see that the American League had a competitive balance of 1.658 and the National League had a competitive balance of 1.897, meaning that the National League was less competitive than the American League.  Overall, MLB had a Noll-Scully of 1.783, which is slightly more competitive than the league average Noll-Sculy since 1982.

If you are interested in doing this on your own, here is a step-by-step guide to calculate the Noll-Scully measure of competitive balance using Microsoft Excel.  

Wednesday, October 4, 2017

2017 MLB Regular Season Attendance

Today I will look at regular season home attendance in MLB.  To do so I grabbed the data from ESPN. First thing to note is that overall regular season attendance has decreased by 488,645 fans from the 2016 regular season and the 2016 regular season dropped by about 600,000 fans from the 2015 season.  Yes, that is a million fewer people attended a MLB game in 2017 as compared to 2015.

The three largest attendance declines were Kansas City (-337,342), Pittsburgh (-329,574) and the New York Mets (-328,980), and the three largest attendance increases were Atlanta (484,338), Cleveland (456,471) and Colorado (351,126).

While total attendance and average attendance are lower this regular season as compared to the 2016 regular season, is it statistically different from the previous regular season?  We need a statistical tool to determine if attendance is different between the two regular seasons with some degree of confidence.  The statistical tool I use is a t-test as the means to perform the analysis.

A t-test looks at the differences between attendance for two regular seasons average home attendance and allows us to judge the difference between their means relative to the variability of their regular season average home attendance.  You can quickly perform a t-test in Microsoft Excel =t.test(...).  I choose to use a two tailed test, since regular season average home attendance can increase and decrease as compared to the previous regular season.

I am going to look at regular season average home attendance for the 2016 and 2017 MLB season, just like I did for the 2016 MLB regular season.  The reason I am looking at average attendance is that not all teams play the same number of home games from one season to the next and using the average is a much more accurate for comparing one season to another.

So after running the t-test between the 2016 and 2017 regular season home attendance using a paired t-test, a t-tests where the sample variance is assumed to be equal, and a t-test where the sample variance is not assumed to be unequal; in each case, the result of the t-test is that it fails to accept (rejects) that regular season average home attendance is different between the two seasons.

That might not be much consolation to the Royals, Pirates and Mets, but the decline in attendance is not significant as compared to the prior regular season.

Monday, October 2, 2017

2017 NCAA FBS Top 25 Ranking for Week 5

Below is the Top 25 ranked teams using the Complex Invasion College Football Production Model from the data provided at www.cfbstats.com.  According to the Complex Invasion College Football production model, Alabama is now the most productive team in all of the Football Bowl Subdivision as of the end of last week. Links to the previous weeks rankings for this season are at the bottom of the page.

Rank Team
1 Alabama
2 Washington
3 Penn State
4 Ohio State
5 Clemson
6 South Florida
7 Washington State
8 Oregon
9 Oklahoma
10 Louisville
11 Georgia
12 Wisconsin
13 Virginia Tech
14 Oklahoma State
15 Auburn
16 Wake Forest
17 TCU
18 Notre Dame
19 North Carolina State
20 Miami (Florida)
21 Kansas State
22 UCF
23 SMU
24 Texas A&M
25 Utah

Previous 2017 Top 25 Rankings
2017 NCAA FBS Top 25 Rankings for Week #4
2017 NCAA FBS Top 25 Rankings for Week #3
2017 NCAA FBS Top 25 Rankings for Week #2

Sunday, October 1, 2017

Sean Kugler Resigns at UTEP

Earlier today University of Texas at El-Paso (UTEP) head football coach Sean Kugler resigned.  The Miners are currently winless this season, with an 0-5 mark.  With his resignation, let's take a look at the UTEP Miners under Kugler's tenure as head football coach.

Below is a chart of offense, defense and total production of the UTEP Miners football program during Kugler's tenure as head football coach, along with who would be the lowest ranked team during this time period (in purple) and the average team (sky blue).   UTEP has been below average each season under his tenure. All rankings in this blog come from my Complex Invasion College Football Production Model.  More details about the program under Kugler are after the chart below as well as a link to his contract.


Sean Kugler [2013-2017*] (2015 amended contract)

2013
At the end of the Kugler's first season, the Miners finished 2-10.  UTEP played against an "average" strength of schedule (SOS) as compared to the "league" average SOS, meaning that their SOS was plus or minus one standard deviation from the "leagues" SOS.  The Miners best victory was against #117 New Mexico State (42-21) and their worst loss was to #106 ranked Tulsa by a score of (20-34).  UTEP had the #116 ranked team in total production with the #107 ranked offense and the #118 ranked defense from the Complex Invasion College Football Production Model.

2014
The Miners again finished the regular season at 7-5 and were bowl eligible, but lost their bowl game against #32 ranked Utah State by a score of (6-21).  UTEP played against an "average" strength of schedule (SOS) as compared to the "league" average SOS.  The Miners best regular season game was a victory (24-21) over #74 ranked Middle Tennessee State, and their worst loss was to #101 ranked Texas Tech by a score of (26-30).  Overall, the Miners had the #69 ranked team with the #75 ranked offense and the #69 ranked defense from the Complex Invasion College Football Production Model.

2015
UTEP finished the regular season overall at 5-7 (bowl ineligible).  UTEP played against an "easier" strength of schedule (SOS) as compared to the "league" average SOS.  The Miners best game again was their victory over #101 ranked Florida Atlantic (27-17), and their worst loss was to #105 ranked UTSA (6-25).  UTEP had the #115 ranked team in total production with the #119 ranked offense and the #98 ranked defense from the Complex Invasion College Football Production Model.

2016
At the end of the regular season the Miners were again bowl ineligible with a 4-8 win/loss record, while playing against an "easier" strength of schedule (SOS) as compared to the "league" average SOS, meaning that UTEP had an SOS between one and two standard deviations higher than the "league" average SOS.  The Miners best win was over #71 ranked UTSA (52-49) and their worst loss was to #124 ranked Rice by a score of (24-44).  UTEP had the #82 ranked team in total production with the #71 ranked offense and the #73 ranked defense from the Complex Invasion College Football Production Model.

2017
Near the halfway mark of the regular season the Miners were 0-5, while playing against an "average" strength of schedule (SOS) as compared to the "league" average SOS.  The Miners worst loss was to currently ranked #125 Rice by a score of (14-31).  At the time of Kugler's resignation, UTEP had the #130 ranked team in total production (that is the worst team), with the #120 ranked offense and the #128 ranked defense from the Complex Invasion College Football Production Model

Monday, September 25, 2017

2017 NCAA FBS Top 25 Ranking for Week 4

Below is the Top 25 ranked teams using the Complex Invasion College Football Production Model from the data provided at www.cfbstats.com.  According to the Complex Invasion College Football production model, Penn State is currently the most productive team in all of the Football Bowl Subdivision as of the end of last week. Links to the previous weeks rankings for this season are at the bottom of the page.

Rank Team
1 Penn State
2 Washington
3 Alabama
4 Oklahoma
5 Washington State
6 Virginia Tech
7 Clemson
8 Ohio State
9 Oregon
10 West Virginia
11 TCU
12 Wisconsin
13 Wake Forest
14 Mississippi State
15 SMU
16 Notre Dame
17 Texas A&M
18 South Florida
19 USC
20 Louisville
21 Texas Tech
22 Michigan
23 Virginia
24 Georgia
25 New Mexico State

Previous 2017 Top 25 Rankings
2017 NCAA FBS Top 25 Rankings for Week #3
2017 NCAA FBS Top 25 Rankings for Week #2

Wednesday, September 20, 2017

Chris Brown's Observation about East Carolina

Earlier this month Chris B. Brown tweeted about East Carolina being "atrocious".
Here is my analysis of the Pirates under McNeill's tenure as head football coach. What I found was that the team was on an upward trend during the first five years as head coach, and then had one poor season and was let go.  While I cannot answer why former head coach McNeill was let go, I can confirm that East Carolina is currently "atrocious".

In the following season (2016), East Carolina was 3-9 and in terms of the Complex Invasion College Football Production Model the Pirates were ranked #113 out of 128 at the end of the season, with the #114 ranked offense and #100 ranked defense.

And things have gotten worse at the beginning of the 2017 season.  East Carolina after both the second and third full weeks of the season ranked as the worst team in all of the FBS.  Next week is not encouraging as they will face South Florida, which is currently a top 25 team in terms of total production.

Monday, September 18, 2017

2017 NCAA FBS Top 25 Ranking for Week 3

Below is the Top 25 ranked teams using the Complex Invasion College Football Production Model from the data provided at www.cfbstats.com.  According to the Complex Invasion College Football production model, Penn State is currently the most productive team in all of the Football Bowl Subdivision as of the end of last week. Links to the previous weeks rankings for this season are at the bottom of the page.

Rank Team
1 Penn State
2 Wisconsin
3 Mississippi State
4 Oklahoma State
5 Oklahoma
6 Washington
7 Duke
8 Clemson
9 Oregon
10 Wake Forest
11 South Florida
12 Maryland
13 Minnesota
14 TCU
15 Vanderbilt
16 Washington State
17 Virginia Tech
18 North Carolina State
19 Kansas State
20 UCF
21 Air Force
22 Utah
23 Texas A&M
24 Arizona
25 Texas Tech

Previous 2017 Top 25 Rankings
2017 NCAA FBS Top 25 Rankings for Week #2

Monday, September 11, 2017

2017 NCAA FBS Top 25 Ranking for Week 2

Below is the Top 25 ranked teams using the Complex Invasion College Football Production Model from the data provided at www.cfbstats.com.  According to the Complex Invasion College Football production model, Air Force is currently the most productive team in all of the Football Bowl Subdivision as of the end of week #2.  It is still very early in the season, so much will be changing as the season moves forward.

Rank Team
1 Air Force
2 Oklahoma
3 Clemson
4 Colorado
5 Penn State
6 Texas Tech
7 Kansas State
8 Oklahoma State
9 Miami (Florida)
10 LSU
11 Mississippi State
12 Alabama
13 Ohio
14 Washington
15 UCF
16 Wisconsin
17 Central Michigan
18 Northern Illinois
19 Colorado State
20 Georgia Tech
21 UTSA
22 Mississippi
23 Navy
24 Wake Forest
25 Connecticut

Monday, July 24, 2017

Hugh Freeze Out at Mississippi

The University of Mississippi (Ole Miss) and head football coach Hugh Freeze have parted ways recently.  While much is being made of the circumstances regrading coach Freeze's departure, I want to look at the football programs productivity during coach Freeze's tenure.

Below is a chart of offense, defense and total production of the Mississippi Rebels football program during Freeze's tenure as head football coach, along with who would be the lowest ranked team during this time period (in purple) and the average team (sky blue).   Mississippi has been trending upward since 2012 except for last season under his tenure. All rankings in this blog come from my Complex Invasion College Football Production Model.  More details about the program under Freeze are after the chart below.


Hugh Freeze (2012 - 2016) [2015 contract]

2012
During Freeze's first year at the helm of the Rebels football team, Mississippi finished the end of the regular season at 6-6 (and were bowl eligible) and defeated to #31 Pittsburgh (38-17) to finish 7-6 overall.  Mississippi played against an "average" strength of schedule (SOS) as compared to the "league" average SOS, meaning that Mississippi was within one standard deviation of the "leagues" SOS.  The Rebels best win was over interstate rival #44 ranked Mississippi State (41-24) in the Egg Bowl, and their worst loss was to #53 ranked Texas by a score of (31-66).  Mississippi had the #45 ranked team in total production with the #48 ranked offense and the #39 ranked defense from the Complex Invasion College Football Production Model

2013
At the end of the regular season the Rebels were 7-5 (and were bowl eligible) and defeated #43 Georgia Tech (25-17) to finish 8-5 overall.  Mississippi played against an "average" strength of schedule (SOS) as compared to the "league" average SOS.  The Rebels best victory was against #11 LSU (27-24) and their worst loss was to #47 ranked Mississippi State by a score of (10-17).  Mississippi had the #45 ranked team in total production with the #53 ranked offense and the #53 ranked defense from the Complex Invasion College Football Production Model.

2014
The Rebels again finished the regular season at 9-3 and were bowl eligible, losing to #1 ranked TCU by a score of (3-42).  Mississippi played against a "much tougher" strength of schedule (SOS) as compared to the "league" average SOS, meaning that their SOS was more than two standard deviations below the "league" average SOS.  The Rebels best regular season game was a victory (23-17) over #7 Alabama, and their worst loss was to #31 ranked LSU by a score of (3-10).  Overall, the Rebels had the #16 ranked team with the #39 ranked offense and the #12 ranked defense from the Complex Invasion College Football Production Model.

2015
Mississippi finished the regular season overall at 9-3 (bowl eligible) and defeated by #40 ranked Oklahoma State in their post-season bowl game by a score of (48-20) to finish the season at 10-3.  Mississippi played against an "average" strength of schedule (SOS) as compared to the "league" average SOS.  The Rebels best game again was their victory over #7 ranked Alabama (43-37), and their worst loss was to #48 ranked Florida (10-38).  Mississippi had the #6 ranked team in total production with the #11 ranked offense and the #29 ranked defense from the Complex Invasion College Football Production Model.

2016
At the end of the regular season the Rebels were 5-7, and were bowl ineligible for the first time under Freeze's tenure, while playing against a "tougher" strength of schedule (SOS) as compared to the "league" average SOS, meaning that Mississippi had an SOS between one and two standard deviations lower than the "league" average SOS.  The Rebels best win was over #45 ranked Georgia (45-14) and their worst loss was to #85 ranked Arkansas by a score of (30-34).  Mississippi had the #83 ranked team in total production with the #56 ranked offense and the #93 ranked defense from the Complex Invasion College Football Production Model.

Wednesday, July 5, 2017

Ranking the Alabama Crimson Tide Under Head Coach Saban

The USA Today recently had an article ranking Alabama over the last ten seasons during Saban's tenure as head coach with the Crimson Tide.  So that prompted me to do the same, except that the Complex Invasion College Football Production Model that I use starts in 2008, so that I will not have anything to say specifically about the 2007 season.  The author of the article (Paul Myerberg) ranks the Crimson Tide teams [by season] as listed below.  I encourage you to read his thoughts in the article linked at the beginning of this blog, and here I will share mine.

Myerberg
2011
2009
2015
2012
2016
2013
2014
2008
2010
2007

First, let me briefly discuss the process about how I came up with the rankings.  To do this, I grabbed all the offense and defense data for all the FBS teams that I use for the Complex Invasion College Football Production Model from 2008 to 2016 and then I ran an offense regression and a defense regression.  I used the coefficient's from the offense and defense (which were all statistically significant and of the correct sign) and multiplied the actual on-field production to get a offense total value and a defense total value.  The higher the value for offense is a better offense and a lower value for defense is a better defense.  Then I subtracted the defense value from the offense value to get the total value.  The advantage here is that we are comparing those nine Crimson Tide seasons using the same "weights" for each season.

Second, I ranked offense, defense and total values since 2008.  Below is my rankings of the Alabama Crimson Tide (in terms of production) overall, with just offense and with just defense.

Total
Offense
Defense
2016
2016
2011
2012
2012
2009
2011
2014
2016
2009
2015
2012
2015
2013
2010
2010
2010
2008
2014
2009
2015
2013
2011
2013
2008
2008
2014

As you can see since 2008, last season's Crimson Tide was the most productive, and the 2008 team was the least productive (excluding 2007 since I don't have that data).  As you can see of the four national championships during Saban's tenure (2009, 2011, 2012 & 2015) that the 2016 team was actually more productive than the others and last year's team would have won except for that last exceptional drive by Clemson, that the 2016 team would also fit in that group.