L lock at FLEXC force as CaptainX excludeBase projection and salary cells are editable.
Export uses this order.
Set your rules and press Optimize lineups. The optimizer enumerates every legal Captain + 5 FLEX combination under the $50,000 cap with branch-and-bound pruning.
Your hand-picked lineups for this slate. Press ☆ Save on any card in the Lineups tab, then export just these for DraftKings upload. Saved lineups survive rebuilds and slate switches, and are saved to your account automatically.
Saved-set exposure
#
Captain
Stack
Proj
Sim mean
P50
P90
P99
P(>T)
Win share
Score distribution
Mean score by game script
Player outcomes
Player
Pos
Tm
Sim mean
P10
P90
Top-scorer %
Portfolio exp
CPT exp
Pearson correlations measured from simulated outcomes. QB ↔ pass-catcher links should read strongly positive; DST ↔ opposing offense negative. Adjust Corr. strength and re-run to see the effect.
Run a simulation to measure correlations.
How it works: load a past slate (pregame salaries and projections), press Optimize lineups, then load that contest's results. Every lineup is scored with the real fantasy points and ranked against the actual field. To get the file on DraftKings: My Contests → Completed → open the contest → Export Lineups to CSV (the contest-standings file).
No contest results loaded. Pick PIT @ CLE · TNF Wk 4 · Final (backtest) in the slate menu to try it with saved results, or load a contest-standings CSV for the slate you have loaded.
DraftKings upload format: a CPT column and five FLEX columns, each filled with the player's DraftKings ID only. On DraftKings, go to Lineups → Upload Lineups, then upload the downloaded file (or paste into the DK upload template and save as CSV).
Lineup key
Roster rules
1 Captain + 5 FLEX from a single game, $50,000 salary cap. The Captain costs 1.5× salary and scores 1.5× fantasy points. Every lineup must include at least one player from each team.
Optimizer
For each eligible Captain, a depth-first search walks FLEX candidates sorted by objective value, pruning any branch whose best-case finish can't beat the current top-K pool or breaks the cap. Lineups that pass every stacking rule enter a heap; the final set is drawn greedily with randomized projections, minimum-unique and exposure limits.
Game script
Vegas total and spread set each team's implied points; offensive projections scale with implied total (elasticity 0.8), DSTs scale inversely with the opponent's implied total. The selected script shifts latent factors, which shifts each player's expected score through his factor loadings.
Live refresh
Refresh live data pulls the current DraftKings game line, player prop lines (passing, rushing and receiving yards, receptions, passing TDs, interceptions, kicking points) and the injury report from ESPN's public feeds. On the PIT @ CLE slate, each projection shifts by how far its lines moved; on imported CSVs, projections are rebuilt from line midpoints. Projections you typed in by hand are never overwritten. Salaries don't change after release — re-import the DK CSV only for a new slate.
Correlation model
Each simulation draws five latent factors: game pace, team A and B scoring, and team A and B pass rate. Every player loads on game pace, own-team scoring, opponent scoring and own-team pass rate according to position, plus an idiosyncratic shock. That makes QB–WR/TE positively correlated, RBs negatively tied to pass rate, and DSTs negatively tied to the opposing offense.
Distribution
Outcomes are log-normal (right-skewed, like real fantasy scoring) with position volatility σ shown above, scaled so the mean equals the script projection. Each simulation samples a game script from your mix weights.