Competition in Major League Baseball: the impact expansion
Abstract:
This article, written by Martin B. Schmidt, focuses on the impact of expansion of the leagues as a major influence on competitive balance in Major League Baseball (MLB). Over the last few decades there has been an increase in competitive balance and many articles have been written focusing on the impact of free agency and the amateur draft and the influence they had on competitive balance. Starting with the 1960s, before free agency (1965) and the amateur draft (1977), when teams expanded from the original 16 teams, he looks at time-series behavior and uses the Gini coefficient to calculate his data. His data suggests that the change in competitive balance occurred quickly after the two leagues began expanding in 1962 for the American League and 1963 for the National League.
Extended Abstract:
Gini coefficient (G):
N= number of teams
XN= winning percentage of team N*
µ= average value of x
I= time period
The expectations for an expansion team are relatively low in their first season, however after the first season of playing, it is expected that the team becomes competitive with the rest of the league. Expansion has been known to increase the competition among the existing teams. When an expansion team is created, along with it usually comes an expansion draft, where competitive teams may lose key players and other talent
A common way to measure competitive balance is to look at the distribution of team winning percentages and the correlation of team winning percentages season to season. However this study develops a different approach by looking at Gini coefficient. The formula is written above along with the variables used. Each league has expanded during different time periods, so it was necessary to change the value of N every time a team was added (See Figure 1). Also the value of µ must be 0.5 because each game, there must be a winner and a loser.
Number of Teams
Period / American League / Period / National League1901-1960 / 8 / 1901-1961 / 8
1961-1968 / 10 / 1962-1968 / 10
1969-1976 / 12 / 1969-1992 / 12
1977-1998 / 14 / 1993-1997 / 14
(Figure 1) / 1998 / 16
The Gini coefficients were calculated and the results show that the change in competitive balance occurred quickly after the two leagues began to expand in 1962 and 1963. For the American League an average Gini coefficient of 0.105 was found for a time period of 1901-1960, with a standard error of 0.022. For the same time period for the National League a value of 0.105 was also found, with a standard error of 0.022. For a time period of 1961-1998, the American and National Leagues had values of 0.077, but the American League had a standard error of 0.015 and the National League was 0.013. This value has gone down over the years leading to greater competitive balance.
The closer the value of the Gini coefficient is to 0, the more balanced the league. A value of 1 would be perfectly imbalanced. If each team won 50% of its games then a value of 0 would be given because it was perfectly balanced.
To find the effect of expansion on the two leagues, two tests, Dickey-Fuller and Phillips-Perron, were used on the Gini coefficients for each time period to determine the integrated level of variables. The results found to reject the unit root. Two additional tests were used which were the Chow breakpoint test and an omitted variable test. The Chow breakpoint test revealed that the pre-expansion periods differed considerably from post-expansion periods. The omitted variable test introduced dummy variables 0 and 1 for time periods of the first expansion year and a year after the first expansion year
In conclusion, this paper introduces the Gini coefficient in calculating competitive balance. The results show that competitive balance began to rise with the movement towards expansion in 1962 for the American League and 1963 for the National League.
*Data on team winning percentages were obtained from The Sporting News Complete Baseball Record Book (1998)
Source: Schmidt, Martin B. Competition in Major League Baseball: the impact expansion. Applied Economics Letters, 2001, 8, 21-26. 2001