Principal Component Analysis.
Firstly, the principal component analysis converts observations of possibly correlated variables into a set of values of linearly uncorrelated variables called principal components.
Secondly, This transformation is defined so that the first principal component (RC 1) has the largest possible variance (that is, accounts for as much of the variability in the data as possible).
Finally, each succeeding component (RC 2 and RC 3, etc.) has the highest variance possible under the constraint that it is not highly related to the preceding components.
PCA allows looking into a group of variables to determine:
- 1) How Many Non-Related Components Exist
- 2) What variables are associated with that group
- 3) The amount of association of all component variables ( 0 to 1) (low to high association)
We now have a way in our Fantasy Football Analysis to find the PCA variable for each position. I can then eliminate metrics that do not add to our coherence of the data.
I discuss each position through the PCA lens to slice through the haze. This video will land you upon the key metrics for judgment.