Abstract
The selection of a cricket team cannot be fair unless the best available performance measures are used. The traditional batting average can be very unrealistic, especially in the case of a small number of scores with a high proportion of not out
scores. In the present study the focus is on using the most suitable measures for the selection of a team after a small number of matches had been played. Provision is made for the fact that match conditions may influence the scoring rate of batsmen.
These measures are used for illustration purposes to select a team from the players who played in the International Cricket Council (ICC) Champions Trophy 2009 One-Day International (ODI) Series. It is shown how an integer programming method can be used for the selection process. The approach is that a well balanced cricket team should include different kinds of specialists, namely batsmen, bowlers, all-rounders and a wicket-keeper. A selection committee may be able to rank batsmen in order of batting ability and bowlers according to bowling ability, but when it comes to all-rounders it is not so simple. The fact that an all-rounder is, by definition, a good batsman and also a good bowler, makes it difficult to rank allrounders. Furthermore, how many of each specialist type should be selected? The purpose of this paper is to show how integer optimisation, an objective scientific method, can be used to aid in selecting a cricket team. Guidelines are also given for the selection of a team if career performance data have to be used.