Some people have expressed an interest in how the ratings for International Test Cricket are created so I have expanded the previously short paragraph with the full gory details. If you have any comments on how these ratings are calculated please send me feedback.
I would like to stress that when I create an official database for International Test Cricket I play several complete leagues. This involves every team playing every other team home and away in a 6 test series resulting in 84 tests being played by each country. If after that a country or individual players do far better/worse than anticipated the rating are adjusted accordingly and the whole process repeated. In the end though it is subjective and other people may use different ratings.
For each player in an International Test Cricket database the following details are stored. An important point to remember is often I only have a players averages to study and do not know specifically all the players abilities. Player styles are normally determined from the ACS yearbook. If just using averages you can come up with a better system please send me details.
Player Surname - normally the full name is displayed for players batting but for fielders (ct., b. etc) just the surname is displayed. If 2 players have the same surname e.g. Waugh then the initials are appended automatically.
Batting Skill - based on test batting average. A player with a test batting average of 40 would have a batting skill of 40. Normally if a player has played less than 20 tests this rating may be adjusted higher or lower accordingly.
Normal Batting Position - can be entered as a normal batting position or more normally as one of the following categories
When the computer selects its initial bating line up it first sorts on category and then batting skill within the category.
Batting Style - one of the following categories
For my new game Cricket Manager I am going to replace this factor with Scoring Rate (runs per 100 balls) which will allow more flexibility. This can not be determined from just averages though so will always be partly subjective. 1 Day averages often do list scoring rate so these will be used as a initial starting point and adjusted down to test match levels.
Bowling Skill - based on 70 - test bowling average. A player with a test bowling average of 25 would have a bowling skill of 45. Whilst this may seem initially strange it means a batsman can be directly compared against a bowler. Thus a batsman averaging 40 is considered as good as a bowler averaging 30. As with batsmen if a player has played less than 20 tests this rating may be adjusted higher or lower accordingly. Also if the player is an occasional bowler the rating is adjusted lower to stop freak high ratings being applied to a player who does not normally bowl.
For Cricket Manager the bowling skill is going to be split in to 2 factors - Strike Rate (balls per wicket) and Economy Rate (runs conceded per 100 balls). This will allow greater accuracy and both factors can be determined from the normal averages.
Bowling Type - one of the following categories
Currently there is no provision for storing whether a bowler is right or left arm. This will be addressed in Cricket Manager but for International Test Cricket it means Left Arm Orthodox and Right Arm Off Spin are simulated the same way.
Fielding Skill - Wicket keeping skill is based roughly on the average number of dismissals per match. A good rule of thumb is number of dismissals per match multiplied by 12. A wicket keeper typically has a skill of 30+
For non wicket keepers fielding skill based roughly on the average number of catches per match. In a similar way to wicket keeping skill they can be based upon the number of catches per match in which case a good rule of thumb is number of dismissals per match multiplied by 15. Alternatively a Fielding Type (see below) can be selected. Fielding skill is also used in a shot stopping capability. A good fielder will stop some singles, turns 2's into 1's, etc. Conversely a poor fielder will allow some 1's to turn into 2's, etc.
To maintain the realism of International Test Cricket there should not be too many slip/specialist fielders in a side. Fielders should be rated relative to the other players in their squad. This might mean a poor rated fielder in the South African squad would be considered a normal rated fielder if he played for another team that typically is not as good as fielding (not naming any teams here for fear of retribution). A good breakdown for a side is to have 3 poor fielders (typically your pace bowlers), 4 normal fielders and 3 slip/specialist fielders plus the wicket keeper. Note Wicket Keeping and Fielding skills must be a multiple of 3 for reasons not worth going in to.
Unfortunately just using averages fielding skill is the one factor that can be most inaccurate. Averages do not show run outs, do not show shot stopping capabilities, etc. There is not a lot I can do about that. My rationale for using a catch ratio is that over time the better fielders are normally placed in the areas where catches are most likely to happen.
Fielding Type - one of the following categories
The table for how a fielding type equates to skill is
Saturday, 16 March 2013.