ken6199 wrote:scrabbarista wrote:.
Curious - how did you pick those 34 players last year? You probably have mentioned it somewhere but I honestly can't recall.
Pretty sure I never mentioned it. Basically, I was just running down the top 10, 15, or 20 players (you could go that far because there was overlap) in each of the three metrics I was using: WS, VORP, and RAPM/possessions. That's why it changed from week to week. Obviously, the players selected could affect the results at every point in the list, but I figured that this methodology, if done blindly, which it was, was basically fair. As implied in my post this morning, the new formula won't have this issue.




















