Abstract
We study a model of two interdependent contests and heterogeneous players with commonly known types. The winners of both contests have winning values that depend on the types (abilities) of both winners. Therefore, endogenous win probabilities in each match depend on the outcomes of the other contests through the identity of the winner. The designer seeds players according to their types in order to maximize (minimize) the total effort. For such interdependent contests, each of which includes two heterogeneous players, we consider two different types of a winning value function and demonstrate that for multiplicative value functions it is optimal to place the two highest type players in different contests. On the other hand, for additive value functions it is optimal to place the two highest type players in the same contest since otherwise they practically do not affect each other.
Similar content being viewed by others
Notes
A nice illustration of the interdependency of an elimination (knockout) tournament is the surprising fact that sometimes the second best player is more likely to win the tournament than to finish as the second one [see Kulhanek and Ponomarenko (2020)].
Csató (2021) is probably the first book that brings operations research into the realm of sports economics.
The equilibrium analysis holds even if the types are positive but smaller than 1. However, some of the results hold if these values larger than 1.
The players’ winning value function is derived from the ex-post interaction of the winners in both contests.
References
Briskorn, D., & Drexl, A. (2009). A brunching scheme for finding cost-minimal round-robin tournaments. European Journal of Operational Research, 197(1), 68–76.
Corona, F., Forrest, D., Tena, J. D., & Wiper, M. (2019). Bayesian forecasting of UEFA Champions League under alternative seeding regimes. International Journal of Forecasting, 35(2), 722–732.
Csató, L. (2020). The UEFA Champions League seeding is not strategy-proof since the 2015/16 season. Annals of Operations Research, 292(1), 161–169.
Csató, L. (2021). Tournament design: How operations research can improve sports rules. Palagrave Pivots in Sports Economics. Palagrave Macmilan.
Dagaev, D., & Rudyak, V. (2019). Seeding the UEFA Champions League participants: Evaluation of the reform. Journal of Quantitative Analysis in Sports, 15(2), 129–140.
Dagaev, D., & Suzdaltsev, A. (2018). Competitive intensity and quality maximizing seedings in knock-out tournaments. Journal of Combinatorial Optimization, 35(1), 170–188.
David, H. (1959). Tournaments and paired comparisons. Biometrika, 46, 139–149.
Della Croce, F., Dragotto, G., Scatamacchia, R. (2022). On fairness and diversification in WTA and ATP tennis tournaments generation. Annals of Operations Research, 316, 107–119.
Einy, E., Haimanko, O., Moreno, D., Sela, A., & Shitovitz, B. (2015). Equilibrium existence in Tullock contests with incomplete information. Journal of Mathematical Economics, 61, 241–245.
Engist, O., Merkus, E., & Schafmeister, F. (2021). The effect of seeding on tournament outcomes: Evidence from a regression-discontinuity design. Journal of Sports Economics, 22(1), 115–136.
Fu, Q., & Lu, J. (2009). The beauty of “bigness’’: On optimal design of multi-winner contests. Games and Economic Behavior, 66(1), 146–161.
Fu, Q., & Lu, J. (2012). The optimal multi-stage contest. Economic Theory, 51(2), 351–382.
Fu, Q., Lu, J., & Pan, Y. (2015). Team contests with multiple pair-wise battles. American Economic Review, 105(7), 2120–2140.
Glenn, W. (1960). A comparison of the effectiveness of tournaments. Biometrika, 47, 253–262.
Gradstein, M., & Konrad, K. (1999). Orchestrating rent-seeking contests. Economic Journal, 109, 536–545.
Groh, C., Moldovanu, B., Sela, A., & Sunde, U. (2012). Optimal seeding in elimination tournaments. Economic Theory, 49, 59–80.
Horen, J., & Riezman, R. (1985). Comparing draws for single elimination tournaments. Operations Research, 3(2), 249–262.
Hwang, F. (1982). New concepts in seeding knockout tournaments. American Mathematical Monthly, 89, 235–239.
Jian, l, Li, Z., & Liu, T. X. (2017). Simultaneous versus sequential all-pay auctions: An experimental study. Experimental Economics, 20, 648–669.
Juang, W. T., Sun, G. Z., & Yuan, K. C. (2020). A model of parallel contests. International Journal of Game Theory, 49, 651–672.
Karpov, A. (2016). A new knockout tournament seeding method and its axiomatic justification. Operations Research Letters, 44(6), 706–711.
Kräkel, M. (2014). Optimal seeding in elimination tournaments revisited. Economic Theory Bulletin, 2, 77–91.
Krumer, A., Megidish, R., & Sela, A. (2017). First-mover advantage in round-robin tournaments. Social Choice and Welfare, 48(3), 633–658.
Krumer, A., Megidish, R., & Sela, A. (2020). The optimal design of round-robin tournaments with three players. Journal of Scheduling, 23, 379–396.
Kulhanek, T., & Ponomarenko, V. (2020). Surprises in knockout tournaments. Mathematics Magazine, 93(3), 193–199.
Laica, C., Lauber, A., & Sahm, M. (2021). Sequential round-robin tournaments with multiple prizes. Games and Economic Behavior, 129, 421–448.
Levi-Tsedek, N., & Sela, A. (2019). Sequential (one-against-all) contests. Economics Letters, 175, 9–11.
Linster, B. G. (1993). Stackelberg rent-seeking. Public Choice, 77(2), 307–321.
Mago, S. D., & Sheremeta, R. M. (2019). New Hampshire effect: Behavior in sequential and simultaneous multi-battle contests. Experimental Economics, 22, 325–349.
Moldovanu, B., & Sela, A. (2006). Contest architecture. Journal of Economic Theory, 126, 70–96.
Rasmussen, R., & Trick, M. A. (2008). Round-robin scheduling—A survey. European Journal of Operational Research, 188(3), 617–636.
Rosen, S. (1986). Prizes and incentives in elimination tournaments. American Economic Review, 76(4), 701–715.
Sahm, M. (2019). Are sequential round-robin tournaments discriminatory? Journal of Public Economic Theory, 21(1), 44–61.
Schwenk, A. (2000). What is the correct way to seed a knockout tournament? American Mathematical Monthly, 107, 140–150.
Searles, D. (1963). On the probability of winning with different tournament procedures. Journal of the American Statistical Association, 58, 1064–1081.
Stracke, R., Hochtl, W., Kerschbamer, R., & Sunde, U. (2014). Optimal prizes in dynamic elimination contests: Theory and experimental evidence. Journal of Economic Behavior & Organization, 102, 43–58.
Suksompong, W. (2016). Scheduling asynchronous round-robin tournaments. Operations Research Letters, 44(1), 96–100.
Szidarovszky, F., & Okuguchi, K. (1997). On the existence and uniqueness of pure Nash equilibrium in rent-seeking games. Games Economic Behavior, 18(1), 135–140.
Tullock, G. (1980). Efficient rent-seeking. In J. M. Buchanan, R. D. Tollison, & G. Tullock (Eds.), Toward a theory of rent-seeking society. Texas A &M University Press.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This article does not contain any studies with human participants performed by any of the authors.
Appendix
Appendix
1.1 Proof of Proposition 1
By (5), the second-order conditions (SOC) of the maximization problems (1)–(4) are
We can see that the SOC are negative as we need\(\square \).
1.2 Proof of Proposition 2
The F.O.C. given by (5) can be rewritten as follows:
By dividing the first two F.O.C we obtain (7)
And, by dividing the last two F.O.C. we obtain (8)
We can now calculate the winning probability of each player using (8):
and
Similarly, by (7), we have
and
Now, inserting (22) and (23) into (21) yields
or, alternatively,
Then, by (7), we have
Thus, the equilibrium effort of the low-type player from set M is
and similarly, the other players’ equilibrium efforts are
\(\square \)
1.3 Proof of Proposition 3
By (9), the total effort in set M is
and in set W it is
Thus, the total effort in both sets is
\(\square \)
1.4 Proof of Proposition 4
Below, we calculate the terms \({\tilde{w}}\) and \({\tilde{m}}\) that appear in the equilibrium efforts given by (15). By definition,
Inserting (13) and (14) yields
By (15), we have
and
Inserting (27) and (28) into (26) yields
Thus,
or, alternatively,
Rearranging the last equation yields the following quardratic equation of the parameter \({\tilde{w}}\),
The solution of this equation is
Since \({\tilde{w}}\) is positive, we have only one feasible solution which is
Similarly, we obtain that
\(\square \)
1.5 Proof of Proposition 5
By (15), we obtain the total effort in set M is
Inserting (16) into the above equation gives us
Similarly, the total effort in set W is
Thus, the total effort in both sets is
\(\square \)
1.6 Proof of Proposition 7
Assume that \(\beta \) is any constant and that \(\alpha \) approaches infinity. Then, by (20), given that \( v_{1}>v_{2}>v_{3}>v_{4}\), we obtain that
-
(1)
$$\begin{aligned} \lim _{\alpha \rightarrow \infty }\frac{TE_{3}}{TE_{1}}=\lim _{\alpha \rightarrow \infty }\frac{\frac{\left( v_{1}^{\alpha }v_{3}^{\beta }+v_{4}^{\alpha }v_{2}^{\beta }\right) \left( v_{1}^{\alpha }v_{2}^{\beta }+v_{4}^{\alpha }v_{3}^{\beta }\right) }{\left( v_{1}^{\alpha }+v_{4}^{\alpha }\right) \left( v_{2}^{\beta }+v_{3}^{\beta }\right) }}{ \frac{\left( v_{1}^{\alpha }v_{4}^{\beta }+v_{2}^{\alpha }v_{3}^{\beta }\right) \left( v_{1}^{\alpha }v_{3}^{\beta }+v_{2}^{\alpha }v_{4}^{\beta }\right) }{\left( v_{1}^{\alpha }+v_{2}^{\alpha }\right) \left( v_{3}^{\beta }+v_{4}^{\beta }\right) }}=\frac{\left( v_{3}^{\beta }+v_{4}^{\beta }\right) v_{2}^{\beta }}{\left( v_{2}^{\beta }+v_{3}^{\beta }\right) v_{4}^{\beta }} >1. \end{aligned}$$
-
(2)
$$\begin{aligned} \lim _{\alpha \rightarrow \infty }\frac{TE_{3}}{TE_{2}}=\lim _{\alpha \rightarrow \infty }\frac{\frac{\left( v_{1}^{\alpha }v_{3}^{\beta }+v_{4}^{\alpha }v_{2}^{\beta }\right) \left( v_{1}^{\alpha }v_{2}^{\beta }+v_{4}^{\alpha }v_{3}^{\beta }\right) }{\left( v_{1}^{\alpha }+v_{4}^{\alpha }\right) \left( v_{2}^{\beta }+v_{3}^{\beta }\right) }}{ \frac{\left( v_{1}^{\alpha }v_{4}^{\beta }+v_{3}^{\alpha }v_{2}^{\beta }\right) \left( v_{1}^{\alpha }v_{2}^{\beta }+v_{3}^{\alpha }v_{4}^{\beta }\right) }{\left( v_{1}^{\alpha }+v_{3}^{\alpha }\right) \left( v_{2}^{\beta }+v_{4}^{\beta }\right) }}=\frac{\left( v_{2}^{\beta }+v_{4}^{\beta }\right) v_{3}^{\beta }}{\left( v_{2}^{\beta }+v_{3}^{\beta }\right) v_{4}^{\beta }} >1. \end{aligned}$$
-
(3)
$$\begin{aligned} \lim _{\alpha \rightarrow \infty }\frac{TE_{3}}{TE_{4}}=\lim _{\alpha \rightarrow \infty }\frac{\frac{\left( v_{1}^{\alpha }v_{3}^{\beta }+v_{4}^{\alpha }v_{2}^{\beta }\right) \left( v_{1}^{\alpha }v_{2}^{\beta }+v_{4}^{\alpha }v_{3}^{\beta }\right) }{\left( v_{1}^{\alpha }+v_{4}^{\alpha }\right) \left( v_{2}^{\beta }+v_{3}^{\beta }\right) }}{ \frac{\left( v_{3}^{\alpha }v_{1}^{\beta }+v_{4}^{\alpha }v_{2}^{\beta }\right) \left( v_{4}^{\alpha }v_{1}^{\beta }+v_{3}^{\alpha }v_{2}^{\beta }\right) }{\left( v_{3}^{\alpha }+v_{4}^{\alpha }\right) \left( v_{1}^{\beta }+v_{2}^{\beta }\right) }}=\lim _{\alpha \rightarrow \infty }\frac{ v_{1}^{\alpha }}{v_{3}^{\alpha }}\frac{\left( v_{1}^{\beta }+v_{2}^{\beta }\right) v_{3}^{\beta }v_{2}^{\beta }}{\left( v_{2}^{\beta }+v_{3}^{\beta }\right) v_{1}^{\beta }v_{2}^{\beta }}=\infty . \end{aligned}$$
-
(4)
$$\begin{aligned} \lim _{\alpha \rightarrow \infty }\frac{TE_{3}}{TE_{5}}=\lim _{\alpha \rightarrow \infty }\frac{\frac{\left( v_{1}^{\alpha }v_{3}^{\beta }+v_{4}^{\alpha }v_{2}^{\beta }\right) \left( v_{1}^{\alpha }v_{2}^{\beta }+v_{4}^{\alpha }v_{3}^{\beta }\right) }{\left( v_{1}^{\alpha }+v_{4}^{\alpha }\right) \left( v_{2}^{\beta }+w_{3}^{\beta }\right) }}{ \frac{\left( v_{2}^{\alpha }v_{1}^{\beta }+v_{4}^{\alpha }v_{3}^{\beta }\right) \left( v_{4}^{\alpha }v_{1}^{\beta }+v_{2}^{\alpha }v_{3}^{\beta }\right) }{\left( v_{2}^{\alpha }+v_{4}^{\alpha }\right) \left( v_{1}^{\beta }+v_{3}^{\beta }\right) }}=\lim _{\alpha \rightarrow \infty }\frac{ v_{1}^{\alpha }}{v_{2}^{\alpha }}\frac{\left( v_{1}^{\beta }+v_{3}^{\beta }\right) v_{3}^{\beta }v_{2}^{\beta }}{\left( v_{2}^{\beta }+w_{3}^{\beta }\right) v_{1}^{\beta }v_{3}^{\beta }}=\infty . \end{aligned}$$
-
(5)
$$\begin{aligned} \lim _{\alpha \rightarrow \infty }\frac{TE_{3}}{TE_{6}}=\lim _{\alpha \rightarrow \infty }\frac{\frac{\left( v_{1}^{\alpha }v_{3}^{\beta }+v_{4}^{\alpha }v_{2}^{\beta }\right) \left( v_{1}^{\alpha }v_{2}^{\beta }+v_{4}^{\alpha }v_{3}^{\beta }\right) }{\left( v_{1}^{\alpha }+v_{4}^{\alpha }\right) \left( v_{2}^{\beta }+w_{3}^{\beta }\right) }}{ \frac{\left( v_{2}^{\alpha }v_{1}^{\beta }+v_{3}^{\alpha }v_{4}^{\beta }\right) \left( v_{3}^{\alpha }v_{1}^{\beta }+v_{2}^{\alpha }v_{4}^{\beta }\right) }{\left( v_{2}^{\alpha }+w_{3}^{\alpha }\right) \left( v_{1}^{\beta }+v_{4}^{\beta }\right) }}=\lim _{\alpha \rightarrow \infty }\frac{ v_{1}^{\alpha }}{v_{2}^{\alpha }}\frac{\left( v_{1}^{\beta }+v_{4}^{\beta }\right) v_{3}^{\beta }v_{2}^{\beta }}{\left( v_{2}^{\beta }+w_{3}^{\beta }\right) v_{1}^{\beta }v_{4}^{\beta }}=\infty . \end{aligned}$$
Thus, when \(\alpha \) converges to infinity, the optimal seeding for a designer who wishes to maximize the players’ total effort is \(M:1{-}4\), \( W:2{-}3.\)
Likewise, by (20), given that \( v_{1}>v_{2}>v_{3}>v_{4}\), we obtain that
-
(6)
$$\begin{aligned} \lim _{\alpha \rightarrow \infty }\frac{TE_{4}}{TE_{5}}=\lim _{\alpha \rightarrow \infty }\frac{\frac{\left( v_{3}^{\alpha }v_{1}^{\beta }+v_{4}^{\alpha }v_{2}^{\beta }\right) \left( v_{4}^{\alpha }v_{1}^{\beta }+v_{3}^{\alpha }v_{2}^{\beta }\right) }{\left( v_{3}^{\alpha }+v_{4}^{\alpha }\right) \left( v_{1}^{\beta }+v_{2}^{\beta }\right) }}{ \frac{\left( v_{2}^{\alpha }v_{1}^{\beta }+v_{4}^{\alpha }v_{3}^{\beta }\right) \left( v_{4}^{\alpha }v_{1}^{\beta }+v_{2}^{\alpha }v_{3}^{\beta }\right) }{\left( v_{2}^{\alpha }+v_{4}^{\alpha }\right) \left( v_{1}^{\beta }+v_{3}^{\beta }\right) }}=\lim _{\alpha \rightarrow \infty }\frac{ v_{3}^{\alpha }}{v_{2}^{\alpha }}\frac{\left( v_{1}^{\beta }+v_{3}^{\beta }\right) v_{1}^{\beta }v_{2}^{\beta }}{\left( v_{1}^{\beta }+v_{2}^{\beta }\right) v_{1}^{\beta }v_{3}^{\beta }}=0. \end{aligned}$$
-
(7)
$$\begin{aligned} \lim _{\alpha \rightarrow \infty }\frac{TE_{4}}{TE_{6}}=\lim _{\alpha \rightarrow \infty }\frac{\frac{\left( v_{3}^{\alpha }v_{1}^{\beta }+v_{4}^{\alpha }v_{2}^{\beta }\right) \left( v_{4}^{\alpha }v_{1}^{\beta }+v_{3}^{\alpha }v_{2}^{\beta }\right) }{\left( v_{3}^{\alpha }+v_{4}^{\alpha }\right) \left( v_{1}^{\beta }+v_{2}^{\beta }\right) }}{ \frac{\left( v_{2}^{\alpha }v_{1}^{\beta }+v_{3}^{\alpha }v_{4}^{\beta }\right) \left( v_{3}^{\alpha }v_{1}^{\beta }+v_{2}^{\alpha }v_{4}^{\beta }\right) }{\left( v_{2}^{\alpha }+w_{3}^{\alpha }\right) \left( v_{1}^{\beta }+v_{4}^{\beta }\right) }}=\lim _{\alpha \rightarrow \infty }\frac{ v_{3}^{\alpha }}{v_{2}^{\alpha }}\frac{\left( v_{1}^{\beta }+v_{4}^{\beta }\right) v_{1}^{\beta }v_{2}^{\beta }}{\left( v_{1}^{\beta }+v_{2}^{\beta }\right) v_{1}^{\beta }v_{4}^{\beta }}=0. \end{aligned}$$
Since we already found that \(\lim _{\alpha \rightarrow \infty }\frac{TE_{4}}{ TE_{3}}=0\), and similarly it can be verified that
$$\begin{aligned} \lim _{\alpha \rightarrow \infty }\frac{TE_{4}}{TE_{2}}=\lim _{\alpha \rightarrow \infty }\frac{TE_{4}}{TE_{1}}=0, \end{aligned}$$we obtain that when \(\alpha \) converges to infinity, the optimal seeding for a designer who wishes to minimize the players’ total effort is \(M:3{-}4\), \( W:1{-}2.\)
\(\square \)
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Cohen, C., Rabi, I. & Sela, A. Optimal seedings in interdependent contests. Ann Oper Res 328, 1263–1285 (2023). https://doi.org/10.1007/s10479-023-05373-8
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10479-023-05373-8