Win Probability (WP): Win Probability is a statistic that measures a team’s chances of winning at any point in the game. At any state of the match, the win probability roughly (though not exactly) corresponds to the fraction of times teams in a similar state in previously played matches have won the game. We have developed a mathematical model that computes a team-neutral and player-neutral win probability for each possible state of a limited over cricket match.
Win Probability Added (WPA): Win Probability Added is a statistic which attempts to measure a player’s contribution to the win of his team by measuring for each specific play of the game, the factor by which a player has altered the outcome of a game. It is the difference in the Win Probability (or Win Expectancy) between two game states (innings, score, overs, wickets) as a result of a specific play. An average team, at any point in a game, has a certain likelihood of winning the game. With each change in the score, number of wickets or the number of overs, there is a change in the team’s probability of winning the game. This change is credited/debited to the batsman and the bowler involved in a specific play.
When it comes to assessing the contributions of individual players, WPA has its strengths and weaknesses. Its greatest strength is that it is the cleanest way to attribute team wins and losses to individual players. On the other hand, WPA is dependent on the game situation, so that a six in a close game is worth more than a six in a one-sided match. WPA is more of a descriptive statistic and not that great of a predictive statistic.
Runs Above Average (RAA): RAA measures the performance of a batsman/bowler in terms of how many more runs the player contributed over an average player in a given year/season. Unlike the WPA, the RAA is a context-neutral performance measure, meaning that the RAA score of a player does not depend on the game context of the player performance. The detailed method for computing the RAA statistic can be found in this article.
Wins: The Wins statistic is also a context-neutral measure of player performance that is derived from RAA. The batting Wins is calculated by dividing batting RAA by 10 times the overall batting average in a given season. The bowling Wins is calculated by dividing bowling RAA by 10 times the overall bowling average in the given season. More details can be found in this article.
EFscore: The eigenfactor score, a metric used for ranking teams in different formats of the game. This is an alternative way of ranking teams in international Cricket, by taking into account the relative strengths of the teams. A victory against a relatively stronger team will lead to a higher EFscore as compared to a victory against a relatively weaker team. The advantage of this method is that it is a non-parametric way of ranking teams, that is we do not give any additional parameters to the algorithm for computing the team rankings.