In this video (made on Sunday, NOT using contest sims), I review the strategy I used for Game 1 of the NBA Finals, where I chopped the Milly, and Game 2, where I won the followup ToC semis ticket. We start with geometric mean ownership filtering, discuss how and when it stands in as a proxy for duplicates, and then on to the importance of salary limits, to avoid unnecessarily overduped lineups. Our approximation of geometric mean ownership assumes independence between people in lineup, and if everybody jams the same guys to fill salary, that's not true. So I set a limit $200 below max, in addition to the geometric mean limit, and then set a min uniques of 3 or 4, depending on how I feel about the resulting pool (I did not do that step for NBA, but seems like generally good practice for a 150-maxer).

0 Comments Sort By