Best player in a team sport. (theoreticall discussion)

[jstnrgrs]

Theoretically, would the best way to determine the best player in a team sport be to divide the players into new random teams after each game, and then see which player has the best record after many games?

For example, suppose you had 18 baseball players, and you wanted to see who was the best overall player. I think the best way to determine this be to divide these players into 2 random teams, and play a game, then divide them into two new random teams and have them play again. After doing this many times, the player who was on the winning team the most should be ligitmately called the best player.

Please do not consider issuse of logistics, or team chemistry in this discussion.

[LionelHutz00]

How many times would you redivide the teams? How many games would each division of teams play? What would you do about players playing the same positions? Like if 2 of your 18 players are pitchers, and they get put on the same team, the team with no pitchers doesn't have much of a chance, but that wouldn't mean that everyone on the no-pitchers team sucks.

[jstnrgrs]

[ QUOTE ]
How many times would you redivide the teams? How many games would each division of teams play? What would you do about players playing the same positions? Like if 2 of your 18 players are pitchers, and they get put on the same team, the team with no pitchers doesn't have much of a chance, but that wouldn't mean that everyone on the no-pitchers team sucks.

[/ QUOTE ]

Well, in my specific example, we would be looking for the best overall player (and pitching would be considered a part of that). Maybe a better example would be to divide up the teams so that they get on player for each position. In this case, though, we would only be looking for the best player at each position, not the best player overall.

The "season" would consits of 1000000000000000000000000000 games with a redrawing of teams after each one.