The calculation for the minimum number of plate appearances is a hybrid formula I derived from the Major League Baseball (MLB) requirement. MLB states that “in order [for a player] to qualify for batting titles in averaged categories, a player must average at least 3.1 plate appearances for every game his team has played.” (Source: ESPN.com)
Figuring that we are playing softball, and are certainly not playing in a league similar in reputation or authority as MLB, I took the average number of games played by the team’s players, and multiplied that by 3.1 plate appearances per game, and then multiplied that value by 7/9, since we play 7 innings (compared to the 9 innings played in MLB). As a result, the minimum plate appearances requirement is a moving target, but allows for players to miss a game here and there, and still qualify with the minimum number of plate appearances. The formula looks like this: PAmin = Gavex(3.1)x(7/9) = Gavex(2.4).
The average number of games played, itself, is a tricky statistic. In general, the value can be calculated by totaling the player-games played for the entire season and then dividing by the number of players on the team, Gave = ΣG/N. However, this is only true when the team has been together for the entire season, i.e., players were not added or subtracted during the course of the season. In the case that a player is added or subtracted, the average number of games played should be weighted accordingly. For example, if a player was added halfway through the season, his games played should be multiplied by 2 for the calculation of Gave.
Additionally, Gave should be kept unrounded through the calculation. The final value of PAmin should be rounded to the nearest whole number (which may require some judgement). I am currently working on the theory that the cut-off point should be one-third and not the typical one-half.
I had considered using the median of games played in place of the average. An average is what I call a “dirty statistic,” and in general, the median will provide more “useful” and “accurate” information, but that is in the case of relatively large sets. The sets (in this case, the number of players) I am dealing with are in the range of 15 to 20, and the median loses its value with such small sets.
