This systematic review examines recent statistical methodologies for assessing cricket player performance across batting, bowling, and fielding. Traditional metrics such as batting averages, runs scored, and wickets taken often fail to capture a player's comprehensive contribution to the game. Recent approaches, however, employ advanced statistical techniques including Data Envelopment Analysis (DEA), Generalized Geometric Distributions (GGD), Bayesian methods, copula functions, and machine learning algorithms, to develop novel performance indices. These methods incorporate contextual factors like opposition quality, pitch conditions, and match situations, offering a better evaluation, particularly in fast-paced formats like T20 cricket. The review emphasizes the integration of metrics across all skill areas to achieve an overall assessment, highlighting the significance of opening partnerships, high-pressure bowling performances, and fielding efficiency in determining match outcomes. It also discusses the growing utilization of real-time data and sophisticated analytics in strategic decision-making and player selection. However, challenges remain such as the difficulty in collecting detailed performance data, particularly for fielding actions, and considering psychological factors like how different players handle pressure. The review concludes by underscoring the need for future research to improve evaluation methods, include more detailed data, and integrate psychological dimensions. As cricket analytics continue to evolve, embracing these new approaches will help teams make better use of data, leading to improved strategies, player performance, and a deeper understanding of the game.

This study provides an in-game prediction of the winner for Twenty20 (T20) cricket by focusing on the matches played in the Indian Premier League. For the analysis, data were collected from 812 completed matches played between 2008 and 2020. Initially, several candidate features were identified, and then the LASSO method was applied as a feature selection technique to identify the most important set of features. Based on the identified important features, predictions are provided for each stage of a match where a T20 match can consist of a maximum of 240 stages. For each stage, three classification models were formed using Naive Bayes, Logistic Regression and Support Vector Machines. The prediction accuracy was used for evaluating the findings of the study, and the prediction accuracy of each model indicates the ratio between the number of correctly predicted instances and the total number of predicted instances. Naive Bayes demonstrated prediction accuracies ranging from 53.08% to 91.76% between the first and the 240th stage of matches, whereas the accuracy of Logistic Regression varied from 56.92% to 97.65%. In comparison, Support Vector Machines also displayed comparable outcomes with a prediction accuracy of 55.00% at the first stage, and 90.59% at the 240th stage. Furthermore, a strategy generator that assists the competing teams in the second innings to devise winning strategies, is presented in this study alongside an interactive web-based application for making in-game predictions, and for assisting the end users (players and coaching/managing staff) in decision making, based on the generated winning strategies.

When a batter scores low in a series of innings, lack of self-confidence generated as a result coupled with poor batting strategies may lead to scoring even lower in the immediately preceding innings. In a team game like Cricket, the out-of-form status of a batter may affect the outcome of the game. This motivated us to explore a statistical method to detect batting inconsistency of a batter in Cricket. One-sided Exponentially weighted moving average (EWMA) control chart was employed due to its evident capability in rapid detection of upward or downward changes in the process mean. The analysis of this study considered ODI batters, who were engaged in the sport during the past decade belonging to seven different cricket nations. An observation of out-of-control state in a control chart was recognized as a batter experiencing an out-of-form performance phase and basic sensitizing rules were utilized to detect non-random and unnatural variation which were explained as warning signals of batting inconsistency. As a demonstration of the proposed method, EWMA control charts of a few popular batters were presented and analyzed which revealed some interesting findings related to their ODI career performance. When an occurrence of a clear downward sequence was detected in the control chart, it was decided that the batter has been out-of-form during that phase. The proposed method can be utilized to dynamically evaluate career batting performance through detection of out-of-form status of batsmen using their recent past scores and this would facilitate effective player selection for the next game.

This study applies Bayesian Network analysis to examine the probabilistic causal relationship between calendar effects and stock market anomalies in the Colombo Stock Exchange. While prior research has explored the existence of Calendar Anomalies in the Colombo Stock Exchange, few studies have examined the underlying cause-and-effect relationship between these anomalies and their associated probabilities. This study employs a Bayesian Network model using market data from 2007 to 2020 to investigate this relationship. The results indicate that calendar effects are prevalent in the market, and the analysis identifies a probabilistic causal relationship between abnormal market returns and Day-of-the-Week and Turn-of-the-Month calendar anomalies. The findings of this study enable investors to time their trades by assigning probabilities to positive or negative market returns on specific trading days, maximizing their returns and improving the efficiency of their trades in the Colombo Stock Exchange.

Cluster analysis can be defined as applying clustering algorithms with the goal of finding hidden patterns or groupings in a dataset. Different clustering methods provide different solutions for the same dataset. Traditional clustering algorithms are popular, but handling big data sets is beyond the ability of such methods. We propose three big data clustering methods, based on the Firefly Algorithm (FA). Three different fitness functions were defined on FA using inter cluster distance, intra cluster distance, silhouette value and Calinski-Harabasz Index. The algorithms find the most appropriate cluster centers for a given data set. The algorithms were tested with four popular synthetic data sets and later applied on two badminton data sets to identify different playing styles of players based on physical characteristics. The results specify that the firefly algorithm could generate better clustering results with high accuracy. The algorithms cluster the players to find the most suitable playing strategy for a given player where expert knowledge is needed in labeling the clusters. Comparisons with a PSO based clustering algorithm (APSO) and traditional algorithms point out that the proposed firefly variants work similarly as the APSO method and surpass the performance of traditional algorithms.