Quote Originally Posted by Enscheff View Post
That is not trivial since most guys in that scenario aren't ranked by all 4 outlets. Makes it non-trivial (though certainly not impossible) to loop through and exclude values when the data-set for each player isn't a consistent number of data points.

As a sanity check, we can look at some of the obvious outliers though:

Judge is definitely one that jumps off the page with rankings of 63, 22, 90, and 143. His average is 79.5, worth $21.32. Throwing out the 22 and 143, his average moves to 76.5, for a value of $21.54.

Guys like Giolioto (10, 3, 25, 14), Kaprielian (58, 110, 87, 62), Allard (67, 61, 37, 26) and Acuna (31, 110, 67, 68) end up working out similarly. They amount to insignificant changes of value.

Someone like Albies would benefit the most since he has one bad ranking, and 3 good ones (35, 12, 11, 10). Removing his best and worst increases his value from $62.96 to $67.75. That is a $5 jump, but we are dealing with total values in the $300 range, so the overall effect is almost insignificant (less than 2%).
Judge and Albies were two that I noticed were affected by outliers. I think you are correct though that it won't affect the big picture.