How to build 150 DFS lineups

Entering 150 lineups is not the same as entering one lineup 150 times. The point of a large portfolio is coverage: of different players, different stacks and different ways the slate could go. The risk is that software, left alone, builds 150 slight variations of the single highest-projected lineup.

1. Decide your exposure budget first

With 150 lineups, every player's exposure is a budget you are spending. Before building, decide the handful of players you want well above the field, the ones you want to fade, and a sensible ceiling for everyone else. Everything you do not decide should default to spread out rather than concentrated. See the guide on exposure for how to think about the numbers.

2. Build around correlation

Most large-field winning lineups tell a coherent story about a game — a quarterback with his receivers, often with a player from the other side who benefits if the game becomes a shootout. Decide what share of your lineups should stack, and around which games, rather than leaving it to chance.

3. Control overlap

Two lineups that share eight of nine players are close to the same bet. A maximum-overlap rule, or an optimizer that penalizes near-duplicates, keeps the portfolio from collapsing onto one core. On single-game slates the same six players with a different Captain is a legitimate variation; the same Captain and core repeated is not.

4. Review before you upload

5. Upload, then handle late news

Export in the site's upload format, or fill the contest's entries file so all 150 entries are replaced at once. If a player is ruled out after you build, rebuild only the affected lineups instead of starting over.

Doing it here