ABSTRACT
This chapter details the flow of a cricket match and describes how performance analysis of batting and bowling tasks is undertaken. Cricket involves two teams of 11 players consisting of batsmen and bowlers, and each team is supposed to bat and bowl at least once. While batting, the team’s aim is to score runs and set a target for the opponent, while bowlers aim to take wickets of the opposition batsmen and limit the runs scored. Batting performance is indicated with runs, and bowling performance with wickets. Batting and bowling performance also needs to be contextualised based on the three formats of international cricket – five-day test matches, one-day internationals (ODI) or a Twenty20 (T20). Performance analysis in cricket is mostly based on widely available descriptive statistics of batting and bowling performance such as average runs and wickets, as well as runs and wickets per 100 balls faced or bowled. More recent updates in broadcast data also include projected scores and win predictors. These descriptive statistics have been used to simulate match outcomes, judge the best starting line-ups as well as gauge run differentials of players. However, on-field strategy often requires more detailed observations of responses and outcomes. For situation-specific analysis as well as individual opponent-based strategy, notational analysis of ball-by-ball data is performed. These notations include categorical data that describes different characteristics of every ball bowled in the match. It could include ball-pitching lengths and associated batting stroke types, but also positions of fielders and other measures of quality of bat on ball contact. Analyst, coach and player interviews suggest that such notational analysis informs on-field plans, while also mentioning how different captains may put varying degrees of faith in such data-driven insights. Examples of effective use of data-driven strategy are also provided.
