I have collected and continue to collect data on an informal sports league. The league has approximately 60 players. Prior to each match, two players are selected as captains and select players for their team from a limited pool of 10 assigned to play that day in a structure: (Team 1 picks 1, Team 2 picks 2, Team 1 picks 2, Team 2 picks 2, Team 1 picks 2, Team 2 picks 1). The levels of the players differ substantially, and therefore pick order differs depending upon who the 10 players are who are assigned to play that day. A player picked 1st one day might be picked 5th on a different day, depending upon who is assigned to play that day.

I have collected data on player pick order, team composition, and wins/losses of each team, as shown in the table below. I am interested in modeling player “value”, essentially seeing which players are over/under valued based on their pick order and the wins/losses recorded over a season. How would I go about doing this with the data below (sample data; anonymized)/is this possible at all?
Player Win/Loss Data

I have some experience modelling in R (and tidy data, which the example data is not), but I cannot think of an operation to calculate player “value”. I would appreciate any basic advice the community can provide on initial data structure and what formula/calculation would achieve the answer needed!