Collusion in MLB

class: center, middle, inverse, title-slide

.title[
# Collusion in MLB
]
.author[
### Toby Yanchunas
]
.date[
### 4/17/2024
]

---

# Overview of Porters Five Forces
### Rivalry
- Teams compete for players, TV deals, sponsorship
  - Offer better contracts
  - Field competitive teams

### Substitutes
- No great substitutes
  - Nippon (NPB): 13.3% of MLB Revenue
  - KBO: 3.5% of MLB Revenue

---

# Overview of Porters Five Forces 2
### Suppliers (Players)
- Increasing Bargaining power
  - MLBPA
  - Arbitration, prospect promotion incentives
  - The salary cap and service time still exist

### Buyers
- TV Providers
  - Local TV deals account for 23% of revenue
- Fans
  - Oakland Athletics: Fans protesting

---

# Overview of Porters Five Forces 3
### Potential New Entrants
- 1930: 2 new teams were founded
  - $130 million each
- Now: Estimated $2 billion franchise fee per team
  - Unlikely to see any new franchises this decade

### Government intervention
- 1922: Supreme court ruled MLB was exempt from the Sherman Antitrust Act
  - Ruling was upheld in 1953 and 1969 (Curt Flood) 
- Fans
  - Oakland Athletics: Fans protesting

---

# Important, Potentially Collusive Topics

### Reserve Clause

- Players were bound to their original team until 1975

- 1922 MLB was ruled exempt from the Sherman Antitrust act

### Service Time

- A club has strong control of a player for their first 6 Years

- Arbitration attempts to compensate the players for this

---

# Continued

### Luxary Tax

- No hard salary cap

- Progressive system

### Revenue Sharing

- As of 2023, 48% of each teams revenue is pooled
  - "Improve performance on the field"

--- 
---
# History of Collusion

### 1985, 1986 & 1987
  
- 1985: Owners met and agreed to limit players contracts in length
  - Agreed not pursue other teams free agents
  - “Hold down salaries by any means necessary”
  - Under Commissioner Peter Ueberroth
  
- 1987: Owners had an "Information Bank"
  - Stored contract offers
  
- Total fines came to $280 million
  - Settled in Nov 1990
  - Certain players were granted free agency

---

# Accusations

### 2002 & 2003
- Players accused the owners of colluding
  - In 2006, players were awarded $12 million
  - No admission of guilt

### 2007 & 2008
- Alex Rodriguez
- Barry Bonds

### 2024
- Kiké Hernández 
- MLB teams may be sharing an algorithm to evaluate players
---

# Filtering Players in a certain WAR range

### What is WAR?

- WAR is described as "Wins Above Replacement"
  - Cumulative 
  
- It is a measure of a players value, relative to an average minor league player
  - As described on Baseball Reference: "8+ MVP, 5+ A-S, 2+ Starter, 0-2 Sub, < 0 Repl"
  
- We will filter players in the 2-4 WAR range

---

# Differences in Indexes

![](449finalmarkdown2_files/figure-html/unnamed-chunk-3-1.png)

---
# Graph of Indexes 
- Filtered Data
![](449finalmarkdown2_files/figure-html/unnamed-chunk-4-1.png)

---

# Actual Data, not Indexes
- Filtered Data
![](449finalmarkdown2_files/figure-html/unnamed-chunk-5-1.png)

---

# Cost of a Win 
- Filtered Data
![](449finalmarkdown2_files/figure-html/unnamed-chunk-6-1.png)

---
# One last graph
- Filtered Data
![](449finalmarkdown2_files/figure-html/unnamed-chunk-8-1.png)
---
#Notes on Variables

## Treat
- Variable of interest
- 1 if Year is 2013 - 2019 (orange boxes on prior graphs)

## Dummy2020
- Dummy Variable for the year 2020, when data was irregular

## sal_per_total
- Salary divided by total revenue in each season for each player

---

# 2 - 4 War Players

```
## 
## Call:
## lm(formula = sal_per_total ~ War + dummy2020 + treat, data = dfilt)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -937979 -613578 -276540  392439 4038673 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)   343718      72962   4.711 2.57e-06 ***
## War           162264      24726   6.563 6.17e-11 ***
## dummy2020     275419     104541   2.635  0.00847 ** 
## treat        -135753      31291  -4.338 1.48e-05 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 811300 on 3120 degrees of freedom
## Multiple R-squared:  0.02006,    Adjusted R-squared:  0.01912 
## F-statistic: 21.29 on 3 and 3120 DF,  p-value: 1.195e-13
```

---
# 2 - 4 War, robust standard errors

```
## 
## t test of coefficients:
## 
##             Estimate Std. Error t value  Pr(>|t|)    
## (Intercept)   343718      73722  4.6624 3.257e-06 ***
## War           162264      25961  6.2503 4.652e-10 ***
## dummy2020     275419      93007  2.9613  0.003087 ** 
## treat        -135753      29689 -4.5725 5.008e-06 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
```
---

# 2 - 4 War, cluster robust SE's

```
## 
## t test of coefficients:
## 
##             Estimate Std. Error t value  Pr(>|t|)    
## (Intercept)   343718     101564  3.3843 0.0007225 ***
## War           162264      32631  4.9728 6.957e-07 ***
## dummy2020     275419      67856  4.0589 5.052e-05 ***
## treat        -135753      40059 -3.3888 0.0007107 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
```
---
# All players (positive war)

```
## 
## Call:
## lm(formula = sal_per_total ~ War + dummy2020 + treat, data = dfilt2)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1835341  -431812  -247763   242558  4597572 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)   387639      11016  35.189  < 2e-16 ***
## War           144764       3813  37.966  < 2e-16 ***
## dummy2020      81194      44726   1.815   0.0695 .  
## treat        -112734      14359  -7.851 4.46e-15 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 738200 on 12242 degrees of freedom
## Multiple R-squared:  0.1115, Adjusted R-squared:  0.1113 
## F-statistic: 512.1 on 3 and 12242 DF,  p-value: < 2.2e-16
```
---
# All players (positive war)
### Robust Standard Errors

```
## 
## t test of coefficients:
## 
##              Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  387639.0    10606.2 36.5483  < 2e-16 ***
## War          144764.0     5095.5 28.4100  < 2e-16 ***
## dummy2020     81194.4    46952.7  1.7293  0.08378 .  
## treat       -112734.2    13577.3 -8.3031  < 2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
```
---

# All players (positive war)
### Cluster Robust SE's

```
## 
## t test of coefficients:
## 
##              Estimate Std. Error t value  Pr(>|t|)    
## (Intercept)  387639.0    19079.4 20.3171 < 2.2e-16 ***
## War          144764.0     8612.5 16.8086 < 2.2e-16 ***
## dummy2020     81194.4    25763.4  3.1515  0.001628 ** 
## treat       -112734.2    26075.5 -4.3234 1.549e-05 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
```