Moneyball Project

MoneyballProject

Thisproject aimed at assessing the MLB value for non-pitcher free agentsfor the 2015 season. Simplified methods were used and the data wasanalyzed and presented accordingly.

TeamSheet Analysis

Inthis analysis, data was gotten from the MLB database for the thirtyteams. The data considered included On Base Percentage (OBP) andSlugging Percentage (SLG), attendance, the price for the tickets,winning percentage and ticket revenue. The presentation in Excelworksheet followed the order presented below

Team Name

Winning Percentage

OBP

SLG

Offensive Index

Attendance

Ticket Price

Total Ticket Revenue

Allthe other values were obtainable directly from the database exceptfor Index and total ticket revenue. The offensive index wascalculated using the below formula

Multiplyingthe formula by 100 was to enhance readability and interpretation ofthe resulting numbers.

Onthe other hand, the total revenue was calculated by multiplying thetotal attendance by the ticket price. The formula is presented below.

Inthis case, total attendance was is the product of average attendanceand total games player (GMS). The data is presented in Table 1 Below.

Table1: Data

Team Name

Winning Percentage

OBP

SLG

Offensive Production

Attendance

Ticket Price

Total Ticket Revenue

Arizona

0.5

0.323

0.391

103.7

4812210

$16.89

$ 81,278,227

Atlanta

0.833

0.321

0.402

104.4

5047110

$17.32

$ 87,415,945

Baltimore

0.5

0.313

0.431

105.7

4676292

$23.89

$ 111,716,616

Boston

0.667

0.349

0.446

114.4

5509944

$53.38

$ 294,120,811

Chicago Cubs

0.6

0.3

0.392

99.2

5247504

$44.55

$ 233,776,303

Chicago Sox

0.333

0.302

0.378

98.2

3982335

$21.35

$ 85,022,852

Cincinnati

0.667

0.327

0.391

104.5

5165524

$19.59

$ 101,192,615

Cleveland

0.333

0.327

0.41

106.4

3831520

$23.65

$ 90,615,448

Colorado

0.667

0.323

0.418

106.4

5439960

$26.36

$ 143,397,346

Detroit

1

0.346

0.434

112.6

5353290

$30.09

$ 161,080,496

Houston

0.5

0.299

0.375

97.3

3885246

$19.83

$ 77,044,428

Kansas City

1

0.315

0.379

100.9

3972726

$27.54

$ 109,408,874

LA Angels

0.333

0.329

0.414

107.2

5253984

$22.37

$ 117,531,622

LA Dodgers

0.5

0.326

0.396

104.8

6606684

$29.27

$ 193,377,641

Miami

0.167

0.293

0.335

92.1

3993605

$24.95

$ 99,640,445

Milwaukee

0.167

0.311

0.398

102

5049702

$32.59

$ 164,569,788

Minnesota

0.167

0.312

0.38

100.4

4666410

$27.73

$ 129,399,549

NY Mets

0.5

0.306

0.366

97.8

4616514

$25.30

$ 116,797,804

NY Yankees

0.333

0.307

0.376

99

5970202

$51.55

$ 307,763,913

Oakland

0.429

0.327

0.419

107.3

4183164

$22.12

$ 92,531,588

Philadelphia

0.5

0.306

0.384

99.6

5479812

$37.42

$ 205,054,565

Pittsburgh

0.333

0.313

0.396

102.2

4969265

$17.21

$ 85,521,051

San Diego

0.571

0.308

0.378

99.4

4775760

$15.99

$ 76,364,402

San Francisco

0.429

0.32

0.381

102.1

6159860

$30.09

$ 185,350,187

Seattle

0.5

0.306

0.39

100.2

4065552

$28.45

$ 115,664,954

St. Louis

0.6

0.332

0.401

106.5

5959576

$33.11

$ 197,321,561

Tampa Bay

0.5

0.329

0.408

106.6

3982905

$20.39

$ 81,211,433

Texas

0.429

0.323

0.412

105.8

5456099

$22.54

$ 122,980,471

Toronto

0.667

0.318

0.411

104.7

4807674

$32.98

$ 158,557,089

Washington

0.333

0.313

0.398

102.4

5130702

$35.24

$ 180,805,938

Thereafter,three scatter plots were created. These are

  1. Scatter plot of total revenue of the team versus the Index. This is presented as shown in the Figure 1 below

Figure1: ScatterPlot of Total Revenue versus Index

Fromthe figure 1, the linear equation is expressed as Y= 3×106X- 108.Therefore, from the definition of linear equation, the slope isequivalent to 3×106.This implies that each additional point on the index of offensiveproduction will be worth 3×106revenue in dollars.

  1. Scatter plot of Winning Percentage versus the Index of offensive production. This is presented as shown in the Figure 2 below

Figure2:Scatter Plot of Winning Percentage versus Index

Fromthe scatter plot of the team winning percentage against index, thelinear equation is Y=0.0193x -1.4841. Therefore, the slope is 0.0193.This implies that for a unit change in Index, the winning percentagechanges by 1.93 percent.

  1. Scatter plot of Scatterplot of total revenue versus win percentage. This is presented as shown in the Figure 3 below

Figure3: TotalTicket Revenue vs. Winning Percentage

Fromthe linear equation, y=1×107x108.This implies that a change in X by 0.01 leads to a change in y, teamrevenue, by (1+E7x0.01) + 1+E8 = $100,100,000.

PlayerSheet

Inthis exercise, the player batting was related to player salaries. Theplayers were chosen randomly from the list of player battling. In mycase, the player battling section composed of 41 respondents.According to the requirement of this paper, only 30 players who wereto be chosen. Stratified sampling technique was applied to get thesample of 30 players. Stratified sampling is a form of randomsampling where the population is first divided into strata (classes)to give a representative sample. The sample was first divided intofive classes based on the annual salary of the players after which arepresentative sample was drawn for this project.

Thereafter,the data for the respective players was inserted into the excelworksheet in this order.

Player

OBP

SLG

Actual Salary

Index

Marginal Product

MRP (Salary Prediction)

Allthe data required was obtainable from the database except forplayer’s index, Marginal Product and Marginal Revenue Product(MRP)[ CITATION Kur14 l 1033 ]. Players Index was calculated as in the above equation of teamsheets.

Marginalrevenue refers to the increase in revenue resulting from sale of anadditional unit of the output. In this project, marginal product wasexpressed using the formula below.

Themarginal product reflects the amount that enhanced the team’s indexrather than player at Mendoza Line. The expression for Mendoza lineis shown below.

MarginalRevenue Product (MRP) refers to the change in quantity of totalproduct that results from unit change in variable input keeping allthe other inputs constant. In MLB, MRP reflects the extra revenuebrought in by the player to the team[ CITATION McL12 l 1033 ].

Thisis expressed as the value of the sum of marginal product, value ofincreases of the marginal product and the league minimum of $400,000that any team is assumed to get a player within the Mendoza line.From the previous worksheet, the additional point on the index ofoffensive production is worth 3×106revenue in dollars. Hence, MRP can be expressed using the formulabelow[CITATION Iac12 l 1033 ].

Table2 below shows the data collected and entered into cells

Table2:Data

PLAYER

OBP

SLG

Actual Salary

Index

Marginal Product

MRP (Salary Prediction)

Adam Jones

0.50

0.88

$85,500,000

188

18.7611

$ 3,400,018.76

Alcides Escobar

0.643

1.333

$154,000,000

261.9

26.133935

$ 3,400,026.13

Alejandro De Aza

0.394

0.467

$10,500,000

125.5

12.52614

$ 3,400,012.53

Alex Rios

0.441

0.548

$30,000,000

143

14.272535

$ 3,400,014.27

Alex Rodriguez

0.481

0.524

$504,700

148.6

14.832255

$ 3,400,014.83

Anthony Gose

0.444

0.938

$400,000

182.6

18.22076

$ 3,400,018.22

Asdrubal Cabrera

0.4

0.679

$500,000

147.9

14.75963

$ 3,400,014.76

Avisail Garcia

0.367

0.483

$515,000

121.7

12.146335

$ 3,400,012.15

Billy Butler

0.517

0.586

$501,000

162

16.169495

$ 3,400,016.17

Carlos Santana

0.36

0.375

$400,000

109.5

10.92975

$ 3,400,010.93

Hanley Ramirez

0.444

0.44

$500,000

132.8

13.2557

$ 3,400,013.26

Ian Kinsler

0.471

0.552

$75,000,000

149.4

14.911665

$ 3,400,014.91

Ike Davis

0.407

0.6

$53,000,000

141.4

14.111825

$ 3,400,014.11

Jason Kipnis

0.478

0.5

$1,650,000

145.6

14.53305

$ 3,400,014.53

Jose Iglesias

0.387

0.429

$106,000,000

120.3

12.007455

$ 3,400,012.01

Jose Reyes

0.5

0.724

$400,000

172.4

17.20578

$ 3,400,017.21

Kendrys Morales

0.379

0.586

$400,000

134.4

13.412945

$ 3,400,013.41

Kevin Kiermaier

0.448

0.333

$120,000,000

122.9

12.26881

$ 3,400,012.27

Lorenzo Cain

0.394

0.516

$160,000,000

130.4

13.01467

$ 3,400,013.01

Mark Canha

0.5

0.64

$549,000

164

16.3683

$ 3,400,016.37

Miguel Cabrera

0.464

0.565

$1,000,000

149.3

14.90145

$ 3,400,014.90

Mike Moustakas

0.36

0.417

$9,000,000

113.7

11.34849

$ 3,400,011.35

Mike Trout

0.4

0.704

$500,000

150.4

15.00888

$ 3,400,015.01

Prince Fielder

0.429

0.438

$214,000,000

129.6

12.936135

$ 3,400,012.94

Salvador Perez

0.433

0.759

$7,000,000

162.5

16.216405

$ 3,400,016.22

Sam Fuld

0.4

0.571

$800,000

137.1

13.68287

$ 3,400,013.68

Stephen Vogt

0.423

0.609

$400,000

145.5

14.521155

$ 3,400,014.52

Travis Snider

0.481

0.524

$1,200,000

148.6

14.832255

$ 3,400,014.83

Xander Bogaerts

0.5

0.533

$517,000

153.3

15.30151

$ 3,400,015.30

Yoenis Cespedes

0.48

0.429

$980,000

138.9

13.86513

$ 3,400,013.87

Figure4: SalaryPredicted vs. Actual Salary

Afterobtaining the data, below scatter plot was drawn for salaryprediction (MRP) values versus actual salaries[ CITATION Bra13 l 1033 ].This is shown in the Figure 4 below.

Fromthe correlation coefficient, R2is considerably less than one. This implies that the MLB players arenot paid in accordance to MRP.

Justification

Forthis project to be successful, use of excel formulas was necessary.This project is allows for exploratory data analysis and datawrangling to establish the effectiveness of the Money ball task[ CITATION Hop08 l 1033 ].As such use of formulas is critical in determining the worth ofcertain players. For instance, when calculating the salary of theplayer, the on-base and slugging percentages are used. These must addup to the runs for each team. Theseimplies that use of formula is inevitable and that Ms Excel providesa perfect alternative to carry out the analysis[ CITATION Sch99 l 1033 ].Some of the main critical features of excel include the in-buildformula,auto-calculate tool and data representation using charts and graphs.

Conclusion

Thisproject was critical in evaluating performance of MLB. The data wasgotten from the MLB database and presented in tabular form usinggraphs. The scatter plots were extensively used to linearly presentthe data and get the correlation between them. For instance, throughthe scatter charts, it was identified that the predicted salariescannot be correlated with the actual salaries for the players. Thisis an indication that MLB overpays some players and at the same timeunderpays them.

References

Bradbury, J. C. (2013). What Is Right With Scully Estimates of a Player`s Marginal Revenue Product. Journal of Sports Economics 14(1), 87-96.

Hopkins, T., &amp Rhonda, C. (2008). lugging Percentage in Differing Baseball Counts. Journal of Quantitative Analysis in Sports 4(4).

Iachan, F. (2012). Essays in Macroeconomics: Liquidity and Taxation. Massachusetts: Massachusetts Institute of Technology.

Kuratko, D. (2014). Entrepreneurship: Theory, Process, Practice. Mason, OH: South-Western Cengage Learning.

McLaney, E. J., &amp Peter, A. (2012). Accounting: An Introduction. Harlow: Financial Times/Prentice Hall.

Schell, M. (1999). Baseball`s All-time Best Hitters: How Statistics Can Level the Playing Field. Princeton, NJ: Princeton UP.