#### Date of Award

5-2010

#### Document Type

Thesis

#### Degree Name

Master of Arts (MA)

#### Legacy Department

Economics

#### Advisor

Warner, John T.

#### Committee Member

Tollison , Robert D.

#### Committee Member

Maloney , Michael T.

#### Abstract

Income differences are normally explained by theory in terms of marginal productivity. However, how can one explain salary enormous jump when a vice-president is promoted into a CEO position? It would be too naive for one to believe his/her productivity increased at the same rate as his/her salary. Differences in wage in terms of relative performance is what economists call tournament theory. What we normally see is that workers are underpaid when they enter a company, and then become overpaid after a promotion. The main point behind such theory is that beginners work hard in order to enjoy the fat paycheck later on. Hence, the large increase in a CEO's salary covers all the effort he/she supplied when underpaid.

Tournament theory has been tested and applied in many professional sports. This thesis will analyze the data gathered from the Association of Tennis Professionals website. Its focus will be on the outcomes of a tennis match. Six main hypotheses will be tested: (1) expected marginal payoffs is expected to increase at a linear or at an increasing rate up to the final round, and then have a distinct jump in the final match; (2) best 4 players of a tournament (champion, finalist, and 2 semi-finalist) are expected to receive 50% of total prize money; (3) the probability a player wins a match is expected to increase if he wins the first set. The effect of winning the first set is expected to be more important for lower ranked players; (4) upsets are more likely to occur in ATP tournaments (played in best of 3 sets) when compared to Grand Slam tournaments (played in best of 5 sets); (5) a change in spread level is expected to directly influence the outcome of a match. As spread level increases more wins coming from lower skilled players is expected; (6) total prize money will not affect the outcome of a match but will change the entry in a tournament. Better players are expected to enter tournaments where total purse is higher; (7) The effect of outcome is expected to depend on players' ranking level. Less competition is expected to be seen when ranking difference gets wider.

The construction of marginal payoff in professional tennis tournaments does not follow the theory. Marginal payoffs do increase, but at a decreasing rate, and an even larger decrease in the final round. Also, marginal payoff became negative, sometimes. In most cases, top four players do not receive 50% of total purse, but around 43%.

The probability a player wins a match does increase after he wins the first set. The effect of winning the first set was the same in percentage terms for both better and worse players; yet, in terms of ratio difference, first set is doubled important for worse players. Results from binomial distribution showed that upsets are more likely to occur in matches played in best of 3 sets when compared to matches played in best of 5 sets. An increase in amount of set played does give an advantage for the better player.

Output from regressions on outcome show that a change in spread level will increase the number of upsets in a tournament, meaning that payoff seems to be more important for worse ranked players. Regression results on average rank weight pointed out that an increase in tournament total purse will increase the quality of the draw, which is consistent with theory. In terms of ranking, variables Top Ranking and Bottom Ranking explain results in outcome in a more efficient way, rather than variable Ranking Difference. Lastly, results lead to the conclusion that less competition is seen as differences in ranking increase. Dominance does exist for the better player ranked in the top 10 when compared to the rest subgroups of better ranked players.

#### Recommended Citation

Salge, Carolina, "THE ATP WORLD TOUR: HOW DO PRIZE STRUCTURE AND GAME FORMAT AFFECT THE OUTCOME OF A MATCH?" (2010). *All Theses*. 803.

http://tigerprints.clemson.edu/all_theses/803