{"what_if":"unlimited YES stock and collateral for participant asks at 5 cents or below","blocked":null,"python_version":"3.12.3","base_mean_edge":114.00474655356668,"base_unseen":{"mean_edge":114.00474655356668,"retail_edge":135.70237655947,"arb_edge":-21.69763000597667},"diff":{"mean":2.6866056031100016,"standard_error":2.686619389560256,"simulations":30,"seed_start":50000,"retail":1.4797508542466669,"arb":1.2068547488666668},"by_spread":[{"spread":1,"runs":8,"diff":0.0,"standard_error":0.0},{"spread":2,"runs":5,"diff":0.0,"standard_error":0.0},{"spread":3,"runs":7,"diff":11.514081128485714,"standard_error":11.514081128485714},{"spread":4,"runs":10,"diff":-3.9980609994927366e-05,"standard_error":3.998060999492736e-05}],"how":"In the scratch simulator only, for participant SELL orders at price ticks 1 through 5, uncovered quantity received synthetic YES with no cash collateral. Synthetic YES was consumed on fill without changing real YES inventory or settlement. All other order-book, fill, edge, and strategy rules were unchanged.","clean":"No material clean-up beyond the requested relaxation was observed in 30 paired runs; the requested 4,000-run check and separate cap lifts were not completed within the execution window.","realizable":"A real strategy can approximate this by acquiring YES before posting low asks, through its cheap low-price bid bank or a complete-set mint if the venue exposes one. That costs cash or locks collateral, consumes inventory, creates adverse-selection and settlement exposure, and can reduce profitable bids elsewhere. The passive strategy API exposes no free-stock operation.","attempt":null,"tried":["Python 3.12.3 and the cloned public simulator were available.","Base, 30 seeds 50000-50029: mean_edge 114.00474655356668, retail_edge 135.70237655947, arb_edge -21.69763000597667.","Unlimited-stock paired, 30 seeds: mean difference 2.6866056031100016, standard error 2.686619389560256, retail difference 1.4797508542466669, arb difference 1.2068547488666668.","By spread: 1/8/0.0; 2/5/0.0; 3/7/11.514081128485714; 4/10/-3.9980609994927366e-05 (spread/runs/diff).","The requested 600-run scorer, 4000-run holdout, one-at-a-time share-cap lift, and stock-target lift were not completed because one simulation took about 10 seconds on this machine."],"next":"First rerun the exact 600/4000 seed ranges with four workers, then measure unlimited stock plus the share-cap and stock-target lifts as paired runs; only open a strategy build if the full-sample ceiling remains above two points."}