I'm running tons of scatter plots on my leagues. Looking at H to H, All play, PPG.
Its beginning to shed a little bit of light on the amount of minimum luck that contributes to the overall success of season long head to head matchups.
The R2 value in a scatter plot (degree of correlation) was calculated in powerpoint for:
Season long All-Play win percentage AND average points per game (R=0.9605)
Season long All-Play win percentage AND forecasted Head to Head wins based on weekly rankings (R=0.9772)
Season long All-Play win percentage AND actual season long Head to Head win percentage (R=6568)
Forecasted wins based on weekly rankings was calculated by this. In a 12 team league, if you finished in the top 4 in scoring then you got a 75% of winning. Middle 4 then 50%, bottom 4 then you got a 25% chance of winning.
Essentially, your All-Play record is exceptionally correlated with points per game and forecasted win percentage (R values >0.96)
But when compared to your actual head to head season long record, the R value was 0.6568.
This sheds light on the amount of minimum luck in fantasy football in head to head leagues. Approximately 30% of your season long win total cannot be explained.