Hemanta Saikia

Predicting the Performance of Bowlers in IPL: An Application of Artificial Neural Network

Application of data mining tools is often used in professional sports for evaluating players/teams’ performance. Cricket is one of those sports where a large amount of numerical information is generated in every game. The game of cricket got a new dimension in April 2008, when Board of Control for Cricket in India (BCCI) initiated the Indian Premier League(IPL). It is a franchise based Twenty20 cricket tournament where teams are formed by competitive bidding from a collection of Indian and International players. Since, valuations of the players are determined through auction, so performance of individual player is always under scanner. The objective of this study is to analyze and predict the performance of bowlers in IPL, using artificial neural network. Based on the performance of bowlers in the first three seasons of IPL, the paper tries to predict the performances of those bowlers who entered in the league in its fourth season as their maiden IPL venture. The performances of these bowlers in IPL-IV are predicted, and the external validity of the model is tested using their actual performance in IPL-IV. This prediction can help the franchises to decide which bowler they should target for their team.

A Double Weighted Tool to Measure the Fielding Performance in Cricket

Though several statistics are used to quantify the batting and bowling performance of cricketers, there is no such measure for the fielders in cricket. This article introduces a measure that can be used to gauge the fielding performance of cricketers. The various parameters that are considered for fielding are quantified to scores based on the ball-by-ball information of a match for each fielder. Subjective weights are assigned to these parameters based on the relative importance of different alternatives under various fielding circumstances. The weights are then combined with the scores of each of the on-field performances of a fielder to get his corresponding fielding score. Two different measures of fielding performance are proposed and their relative differences are discussed. To demonstrate these measures, the final of the first Twenty20 World Cup tournament played on 24 September 2007 is considered. The outcomes obtained from the two measures are accordingly compared.

On Performance Measurement of Cricketers and Selecting an Optimum Balanced Team

A cricket squad that participates in a tournament generally comprises of fifteen players. A balanced squad contains players having different expertise like batting, bowling, wicket keeping, etc. Keeping in view the cricketing requirements, selecting an optimal squad is a difficult decision-making problem. In order to attain that, the study proposes a composite index to measure the performance of cricketers irrespective of their expertise. Then using a binary (0-1) integer programming method a balanced squad of 15 players is selected. To apply the tool to practice, optimum balanced squads are selected from amongst the Indian players who participated in the Indian Premier League (IPL), based on the proposed performance measure for different seasons. The team so selected is compared to the actual Indian team selected for the International Cricket Council (ICC) Twenty20 World Cups following the IPLs in the year 2009, 2010 and 2012. It is seen that the actual Indian team is not much variant to the optimal squad of ICC Twenty20 World Cup 2012 but things are different during the years 2009 and 2010. The optimization technique discussed in this study can be helpful for the balanced team selection of other team sports as well.

A Bayesian Classification Model for Predicting the Performance of All-Rounders in the Indian Premier League

The game of cricket got a new dimension, when the Indian Premier League (IPL), a competition of twenty over-a-side featuring eight teams named after various Indian cities/states started in 2008. The teams were franchisee driven and the players were selected via competitive bidding from a pool of available players. All-rounders i.e. players with the ability to both bat and bowl play a noteworthy part in cricket, whatever is the version of the game. The study measures the performance of all-rounders in Indian Premier Leagues (IPL) based on their strike rate and economy rate. The all-rounders are divided into four non-overlapping class viz. performer, batting all-rounder, bowling all-rounder and under performer. Stepwise multinomial logistic regression is used to determine the significant predictors responsible for such categorization. A Naive Bayesian classification model is developed that can use the significant predictors to forecast the class in which an incumbent all-rounder is expected to lie. The classifier is build based on the performance of all-rounders who participated in IPL-I and II, and the validity of the classifier is subsequently tested over the incumbent all-rounders of IPL-III. The classifier though moderately successful in predicting the appropriate class of the incumbent all-rounders in IPL III, is expected to perform better in future with increase in the size of training sample. This classification would be useful for the participating teams’ management while deciding about which all-rounder to be bided for and to what amount in the next addition of the league.

A New Model for Player Selection in Cricket

A balanced squad in cricket contains players with different expertise like batting, bowling, all-rounder and wicket keeping. Maintaining these cricketing requirements, selecting an optimal squad is a difficult decision making problem. The selectors have chosen cricketers using their own experience and based on the recent performances of available players from all the different expertise. The paper introduces a measure that can quantify the performance of the cricketers into a single numerical value, which is a measure of the player’s cricketing efficiency. The distributional pattern of the performance measure is identified and then used to recognize the performers under different expertise. The entire exercise simplifies the job of the selectors as they now have a smaller subset to choose from. To explain the model empirically, data of the Indian Premier League (IPL), 2012 is used to identify the performers of the game under different expertise. The actual Indian team selected just after the said season of IPL is then compared with the performers of IPL, 2012 as identified by the model. It was found that several non-performers were selected in the actual squad and some performers were left out. This model can be extended for the team selection in other versions of cricket and even in other team sports.

A Consistency Adjusted Measure for the Success of Prediction Methods in Cricket

Two models are used to predict the outcomes of matches in a Twenty20 cricket series. The success of prediction methods is hampered by the fact that if two teams play two or more matches against each other and each team wins some of the matches, such inconsistent outcomes cannot all be predicted correctly. The challenge was to find a procedure which could compensate for inconsistent results. The consistency adjusted measure of the success of prediction is defined and shown to give a fair assessment of prediction results. For the first model the success rate of 56.8% is increased to 76.4% and for the second model from 52.7% to 70.9%. The same method can be used in any sports series where teams play against each other more than once.

Performance based market valuation of cricketers in IPL

Purpose – The purpose of this study is to identify the cricketers who were able to justify their salary by their on‐field performance in the Indian Premier League (IPL).Design/methodology/approach – A measure is developed to quantify the batting, bowling and wicket keeping performance of a cricketer into one single index called as the performance index. Based on the performance index of cricketers, from first three seasons of IPL, and using the binomial option pricing model, the neutral present values of the cricketers are determined. The distributional pattern of the present values of cricketers is identified and cricketers are classified based on the level of present values. Similarly, the distributional pattern of the bid prices of cricketers is identified and the cricketers are classified on the level of their bid prices. The cross tabulation of the classified bid prices and classified present values, can be used to identify the cricketers who were able to justify their salaries.Findings – The study c...