I get what they were trying to do, but contribution is never gonna scale directly with salary, so comparing production per dollar isn't really all that meaningful (ex: Lebron isn't gonna do 20x as much as a vet min player, but he definitely deserves to be paid 20x as much)
I'd like to see a comparison of production compared to players with similar salaries (ex. if you get paid $7.5 million/year, where you rank compared to other players in the $7-8 million/year range would be a lot more meaningful)
This is just really biased toward the bottom end of the spectrum
as folks who do stats will often tell you, any and all choices are biased in that they are designed to value some variable more than another variable. (for example: "towels waived per game" COULD be a stat, but most of us are biased against it and dont include it in our analyses...unfortunately. 
)
so it is not a question of "bias" per se. it is a question of clearly stating what you value and why. this is why folks argue over the assumptions behind an analysis so much. it is the most important part.
then the challenge is to adhere to correctly to the methods chosen to measure the variable chosen. what ever outcome/results you achieve will/should reflect your own bias in what is important enough to measure.
ALL RESULTS will provide a bias towards the selected variable. obvious, right?
on the findings above, would a measurement system that biased it towards the UPPER end of the spectrum suit you better? maybe, and that could be valid GIVEN YOUR ASSUMPTIONS. there is not single, pure, pristine, universal, and "correct" measurement of nba players' value since the definition of that is subjective and open to interpretation.
all of this is crudely put, but hopefully makes a straightforward point.
I'm not saying that the person that created the stat is biased (though he very well may be), simply that the stat, by making an assumption about salaries being directly proportional to production, (statistically) biases the results toward people who are paid less
Essentially, it's ignoring that a player that is twice as valuable doesn't need to give twice the production, since you can only have five people on the court and every little edge helps
It makes me wonder if maybe another model would work better (like using the square root of the salary, etc.)
Actually, now I want to find all the Win shares info for all NBA players, compare it to salary, find a model that fits it, and use THAT to evaluate our players (possibly excluding rookie deals since they are pretty much all good deals)
I don't know when I'll have time to do it, but if anyone else wants to, feel free, just let me know what you come up with
Edit: I just started working on it, should be interesting
Topic: Which C's players are the best bargains? (Read 1706 times)