Scribd
Upload
128 views3 pages

Determining Matchdays in Sports League Schedules To Minimize Rest Differences

Uploaded by

Andrea Audine
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
128 views3 pages

Determining Matchdays in Sports League Schedules To Minimize Rest Differences

Uploaded by

Andrea Audine
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

Determining matchdays in sports league

schedules to minimize rest differences


Burak Çavdaroğlu, Tankut Atan, Determining matchdays in sports league schedules to minimize rest
differences, Operations Research Letters, Volume 48, Issue 3, 2020, Pages 209-216, ISSN 0167-6377,
https://doi.org/10.1016/j.orl.2020.03.001.
(https://www.sciencedirect.com/science/article/pii/S0167637720300304)

Abstract
Many sports leagues first announce the games to be played in each round and then
determine their matchdays as the season progresses. This study focuses on the fairness
criterion of minimizing the total rest difference among opposing teams to find the
matchdays for an announced schedule. We show that the problem is decomposable into
optimizing the rounds separately. We also provide a polynomial-time exact algorithm
for canonical schedules.

Introduction
League scheduling is one of the popular applications of operations research in sports. In
most relevant research, the focus is on determining the games in the rounds of the
competition. Several fairness criteria may be considered in this timetable construction.
While the literature has mainly focused on fairness issues such as minimizing the
number of breaks and balancing the carryover effects, fairness criteria regarding the
team rest durations between the rounds have not received much attention by
researchers.

An (opponent) schedule is an assignment of games to rounds such that each game is


assigned to some round and each team plays at most one game per round. Once a
schedule is constructed, it is announced ahead of the competition season. Since each
round may actually consist of several days, tournament organizers determine the days of
the individual games of each round as the season progresses. Often, these matchdays are
criticized by team managers when their teams have fewer days to rest than their
opponents [6], [25]. Moreover, Scoppa [27] shows statistical significance of additional
rest days when the games are three or less days apart when investigating the role of
fatigue in FIFA World Cups and UEFA Championships. Scoppa’s work suggests the use
of a cut-off point in football such that when a team has ample rest days, the rest days
difference is weighted less or even ignored all-together. However, this would not stop
some team managers from complaining. For example, before Spain played against
Portugal at the semi-finals stage of UEFA Euro 2012, Spain’s manager Vicente del
Bosque had moaned about having two fewer days to rest than Portugal even though
Spain had played its quarter-final game four days ago [18]. Therefore, a fair schedule
should have as little rest difference between opposing teams as possible when
determining the matchdays. In this study, we concentrate on the problem of
determining the matchday of each game given a tournament schedule so that total rest
difference among teams is minimized.

The organization of this paper is as follows. After reviewing relevant literature in


Section 2, we first formally describe the problem of minimizing the sum of rest
differences of teams from each game for a given schedule in Section 3. Section 3 also
decomposes the problem into iterative optimizations of each round and shows that this
scheme actually provides an optimal solution. In Section 4, a polynomial-time algorithm
that decides the matchday of each game in canonical schedules with minimum rest
difference is described and proven to be optimal. Our computational experiments and
their results are given in Section 5 followed by a conclusion.

Section snippets

Previous work
A round robin tournament is a common format for organizing sports events. In a round
robin tournament, every team plays every other team a fixed number of times. Briskorn
and Drexl [4] provide integer programming models for round robin tournaments with
several externally given constraints such as disallowing forbidden matches, observing
regional capacities and respecting teams’ preferences for matchdays, and also fairness
constraints such as having few breaks and balancing the carryover

Total rest difference problem


Imagine a single round robin tournament with 𝑛 teams where 𝑛 is an even number.
There are |𝑅|=𝑛−1 rounds in the tournament with 𝑛∕2 games in each round where 𝑅 is the
set of all rounds. All games in each round are to be played in 𝑝 consecutive periods with
a (predetermined) non-negative number of games in each period. We use the
term period as a generic term for a time duration which potentially holds zero or more
games. This time duration can be referred as a matchday if it is equal to a day.

An exact polynomial-time algorithm for canonical schedules


Some top division professional football leagues in Europe such as Premier League
(Russia) and Süper Lig (Turkey) use a schedule best known as canonical schedule. The
main reason for using a canonical schedule is that it is capable of having a home/away
assignment with 𝑛−2 breaks in single round robin tournaments, which is the theoretical
lower bound of the break minimization problem [5]. In a canonical schedule, for 𝑛 even,
the games of each round 𝑟∈{1…𝑛−1} are decided by 𝐺𝑎𝑚𝑒𝑠𝑟={[𝑛,𝑟]}∪{[𝑟+𝑗,𝑟−
Computational experiments
Computer runs were executed on an Intel i5-4570 CPU 3.2 GHz computer with 8 GB of
RAM. Exact solutions to the reported problem instances were obtained with GAMS
using Gurobi 8.1 solver [14] for a Windows 64-bit operating system. The decomposed
problems for each round can be hard to solve exactly with commercial solvers, even if
the schedules were generated using the circle method. For instance, such a single round
robin problem instance with 30 teams and 15 periods (which corresponds to the

Conclusion
When opposing teams in matches have had rest durations of different lengths after their
games in the previous round, the team with less rest is disadvantaged physically which
makes the competition unfair. Thus, it is important to minimize such differences. In
many leagues, the specific dates for playing the games in each round are determined
after the game schedule is announced. In this article, we looked at minimizing the sum
of rest differences for a given round robin tournament schedule.

You might also like