A Bayesian stochastic model for batting performance evaluation in one-day cricket

Abstract

We consider the modeling of individual batting performance in one-day international (ODI) cricket by using a batsman-specific hidden Markov model (HMM). The batsman-specific number of hidden states allows us to account for the heterogeneous dynamics found in batting performance. Parallel sampling is used to choose the optimal number of hidden states. Using the batsman-specific HMM, we then introduce measures of performance to assess individual players via reliability analysis. By classifying states as either up or down, we compute the availability, reliability, failure rate and mean time to failure for each batsman. By choosing an appropriate classification of states, an overall prediction of batting performance of a batsman can be made. The classification of states can also be modified according to the type of game under consideration. One advantage of this batsman-specific HMM is that it does not require the consideration of unforeseen factors. This is important since cricket has gone through several rule changes in recent years that have further induced unforeseen dynamic factors to the game. We showcase the approach using data from 20 different batsmen having different underlying dynamics and representing different countries.

Bayesian survival analysis of batsmen in Test cricket

“…On the other hand, Damodaran (2006) provides a method which does allow for within-innings comparisons, but lacks a natural cricketing interpretation. Various other performance metrics have been proposed, however have been in relation to limited overs cricket (Lemmer, 2004, 2011, Damodaran, 2006, Koulis et al, 2014.…”

Naive Bayes approach to predict the winner of an ODI cricket game

“…According to their findings, they accurately predict the number of runs scored by a batsmen with an accuracy of 90.74% and the number of wickets taken by a bowler with an accuracy of 92.25%. Using batsmen-specific hidden Markov chain approach, Koulis and Muthukumarana (2014) model the individual batting performance of an ODI game. In another study to predict the team's score, Nimmagadda et al (2018) use random forest together with both multiple linear regression and logistic regression.…”

“…Typical game-prediction algorithms use numerous inputs representing the player, the team, the cricket ground, the weather, and other off filed statistics. According to the literature, some of the most frequently used performance indicators in ODI game-prediction are home-field advantage (Bailey and Clearke, 2006;Paul and Stephen, 2002), the result of the coin toss (De Silva and Swartz, 1997;Dawson et al, 2009), day/night effect (De Silva and Swartz, 1997), the effect of bowling (Lemmer, 2008) and batting (Kimber and Hansford, 1993;Koulis et al, 2014;Lemmer, 2008;Lewis, 2005;Scarf et al, 2011;Tan and Zhang, 2001;Wickramasinghe, 2015). In addition to the incorporation of large volume of variables and factors used in game-prediction, the dynamic nature of the game makes the prediction process a daunting task.…”

Modelling Career Trajectories of Cricket Players Using Gaussian Processes

“…Within a Bayesian framework, [13] employed the use of a hidden Markov model to determine whether a batsman is in or out of form. The model estimates a number, K, of 'underlying batting states' for each player, including the expected number of runs to be scored when in each of the K states.…”