{"sources": [{"url": "https://api.imdnotdead.fun/artifacts/151e0b1b16f90a1384b823bc528d6d1407f7b4bbfcc7966c12119fe483bfd326", "quote": "{\n \"approach\": \"Started from the spread-4 source file (its changes are the most invasive and are all guarded by self.s == 4, including its own belief, jump-sum "}, {"url": "https://api.imdnotdead.fun/artifacts/41fa8829432f502a00c5302c0e2200b475c5c2e100a2a16b086aaab0ce62a395", "quote": "{\n  \"approach\": \"Graft without constant tuning onto crossing-cover artifact 40497298132dcc0ccc9f4c0672031c7eaae00ddcf8cf24983db989feeb3c5ca0. Spread 1: dmin=0.3"}, {"url": "https://api.imdnotdead.fun/artifacts/40497298132dcc0ccc9f4c0672031c7eaae00ddcf8cf24983db989feeb3c5ca0", "quote": "{\n  \"approach\": \"Uses tiny tick probes, a sub-tick probability filter, and marginal retail-versus-arbitrage sizing to quote one tick inside the competitor. Star"}, {"url": "https://api.imdnotdead.fun/artifacts/f7304b601a970104247074052cd75db749edc8d9e989fcce2a67c98bdcdc44bd", "quote": "{\n \"approach\": \"The Paradigm-priced spread-1 block (5x ask at tick 2 under 1 cent ahead of the rival, front asks kept), plus three pieces each built on the base"}, {"url": "https://api.imdnotdead.fun/artifacts/fbaba62304b28380cd79b2de69f597deba95429638ed18a2bf5391077d025558", "quote": "{\"approach\":\"The queue retention rules remain intact because all tested deletions lost paired edge on 125 choosing runs. Queue-first spread-2 combination: s2-qu"}, {"url": "https://api.imdnotdead.fun/artifacts/ad3b2ad91284e519d0c80ecfe18eac37e0853b90703d564341744a9639e2a761", "quote": "{\n  \"approach\": \"Spread 3 only: cap the routine ammunition inventory target at min(existing target, current desired main/front ask quantity qa). The ledger iden"}, {"url": "https://api.imdnotdead.fun/artifacts/46b20a51dbb98abf4c9f7832f362dc8320c2c3288dae986a8863a7455bd57402", "quote": "{\"approach\": \"Start from the priced spread-4 censored own-fill flow estimator and its existing arithmetic-preserving optimizations. Add the five supplied rules,"}, {"url": "https://api.imdnotdead.fun/artifacts/61d3ba0c42cd1e20117ad7e31d1b2c8aad12f63f5075d11cdd5af9d277213b66", "quote": "{\n  \"approach\": \"For spread 1 only, split each new ordinary SELL quote at ticks 3 through 98 equally: retain half at its original price and place half one tick "}, {"url": "https://api.imdnotdead.fun/artifacts/c7f4f274520f484181ba0e664f87875a63b1b410c8c8fbb66cd5d32448cfbabd", "quote": "{\"approach\": \"Spread 1 only: in the broad ammunition-purchase branch, replace the bid at ceil(100*phat) with the previous tick only when that tick is at least m"}, {"url": "https://api.imdnotdead.fun/artifacts/cabede3c4180c49f8c5b5d6746b51f47bff599a891a34d9777555c81ee7dfcb1", "quote": "{\n \"approach\": \"Spread 2 only (every change is under self.s == 2). Measured on the start file: on the step after the arbitrageur fully takes one of our main quo"}, {"url": "https://api.imdnotdead.fun/artifacts/1d3fdd0207683d66c09de9f06c83f6cb8ea9f61f5e1f1143b22a9e740e8cc2e3", "quote": "{\"approach\":\"In spread 2 only, the target-stock branch now buys only at tick 1 instead of paying up to tick 2. Keep tick-1 stocking, both crossing-stock branche"}, {"url": "https://api.imdnotdead.fun/artifacts/d8a185edb6be4d9a8805e417809ae476278d873633703ae3f9fe85f18379d3c4", "quote": "{\n  \"approach\": \"Spread 3 only: when probability estimate is below the existing low-price boundary and YES inventory already covers inferred retail-buy flow tim"}, {"url": "https://api.imdnotdead.fun/artifacts/92ec7378991a8902017d1081c92c85c479af1c145e0763f9d12f33bb124096e3", "quote": "{\n \"approach\": \"In spread-3 runs only (guard self.s != 3 added to one condition), drop the first ammunition branch: the bid for YES stock placed at ceil(p) when"}, {"url": "https://api.imdnotdead.fun/artifacts/be28481bad801e43bc8216b42bef60a70442b4365aa8252cb49d5208f253206a", "quote": "{\"approach\":\"For spread 4 only, remove the special low-price bid-risk discount by setting kLb and kLf to the existing kL during initialization. Separate crossin"}, {"url": "https://api.imdnotdead.fun/artifacts/f3dbc5ccc00ed338fd02259601dca57f385ca460999e4f3e66fab4f213945718", "quote": "{\n \"approach\": \"Spread 4 only. The start file locates p inside the gap of five or more ticks between its main quotes with 0.01-share probes around p's tick, rea"}, {"url": "https://api.imdnotdead.fun/artifacts/b3068290c77e8999273e639d0931ec3be3d337b4be9fe41655995b1ef43b0cd6", "quote": "{\n  \"approach\": \"Spread 4 only: cap the routine ceil-tick stock-buying target at min(the existing flow-scaled remaining-retail target, the current desired ask q"}, {"url": "https://raw.githubusercontent.com/danrobinson/prediction-market-challenge/b244530ee2a6247b5f930830eeca9dd950da0e7e/orderbook_pm_challenge/runner.py", "quote": "    config = sample_config(base_config, variance, seed=seed)\n    factory = load_strategy_factory(strategy_path)\n    engine = SimulationEngine(config, factory, seed=seed)\n    return asdict(engine.run())"}, {"url": "https://raw.githubusercontent.com/danrobinson/prediction-market-challenge/b244530ee2a6247b5f930830eeca9dd950da0e7e/orderbook_pm_challenge/config.py", "quote": "    n_steps: int = 2_000"}, {"url": "https://raw.githubusercontent.com/danrobinson/prediction-market-challenge/b244530ee2a6247b5f930830eeca9dd950da0e7e/orderbook_pm_challenge/results.py", "quote": "    def mean_edge(self) -> float:\n        successes = [result.total_edge for result in self.simulation_results if not result.failed]\n        return sum(successes) / len(successes) if successes else 0.0"}]}