{"approach":{"mechanism":"Lognormal marginal-fill sizing, with side-specific retail size conversion.","audit":"Pinned retail.py samples Poisson arrivals, independent Bernoulli sides, and lognormal notionals with sigma=1.2. engine.py executes arbitrage first, then retail; retail buys spend notional walking asks, sells convert notional to shares at max(true probability,0.05). market.py matches price then FIFO. Improving competitor prices avoids a hidden queue estimate.","derivation":"For one arriving order of share size Q and a first-in-price quote q, expected retail fill is E[min(Q,q)]; its derivative is Pr(Q>q). With estimated edge e, side arrival rate lambda/2 and adverse loss A per quoted share, marginal profit is (lambda/2)*e*Pr(Q>q)-A. S1 inverts the lognormal survival function at RISK*A/((lambda/2)*e). S0 uses mean share flow times reward/(reward+A). SCALE is fixed by S0 tuning and shared.","assumptions":"Approximation: uses midpoint probability, midpoint parameter bands (arrival 0.253 and mean notional 4.488), Gaussian volatility proxy and jump-loss allowance; not an exact optimizer. Multiple arrivals, uncertain probability, tick rounding and collateral can invalidate the single-order marginal model. Sell-order conversion uses estimated rather than hidden true probability. No reference strategy code or other strategies were read.","sources":"See sources.json: pinned simulator engine, retail, market, config, runner, types, process, loader, README and rules; numeric reference used only fixture.H.","selection":"Sequential tuning: four S0 scales on all 500 allowed tuning seeds, then four S1 adverse-loss multipliers at selected S0 scale on the same tuning set. Highest mean total edge wins each stage. All other strategy code is fixed. Frozen before holdout; S0 and S1 full sources differ only in TAIL boolean.","simulator":"Unmodified pinned Python modules; run_batch(factory, base_config=ChallengeConfig(), variance=ParameterVariance(), n_simulations=1, seed_start=seed, workers=1), default 2000 steps; no sandbox. Four OS processes use disjoint seed ranges."},"tried":[{"strategy":"S0","scale":0.5,"risk":1,"n":500,"mean":-5.6546202150496,"paired_vs_selected_s0":0.0,"paired_se":0.0,"failures":[]},{"strategy":"S0","scale":1,"risk":1,"n":500,"mean":-9.5569895226518,"paired_vs_selected_s0":-3.9023693076022004,"paired_se":1.2935731021164638,"failures":[]},{"strategy":"S0","scale":2,"risk":1,"n":500,"mean":-10.2005455171648,"paired_vs_selected_s0":-4.5459253021152,"paired_se":2.278200762414949,"failures":[]},{"strategy":"S0","scale":4,"risk":1,"n":500,"mean":-16.604359961358,"paired_vs_selected_s0":-10.949739746308401,"paired_se":3.2928333923867594,"failures":[]},{"strategy":"S1","scale":0.5,"risk":0.5,"n":500,"mean":-7.8709595371922,"paired_vs_selected_s0":-2.2163393221426,"paired_se":2.254474307286516,"failures":[]},{"strategy":"S1","scale":0.5,"risk":1,"n":500,"mean":-7.4997212581584,"paired_vs_selected_s0":-1.8451010431088,"paired_se":2.063734476703402,"failures":[]},{"strategy":"S1","scale":0.5,"risk":2,"n":500,"mean":-7.8426173539552,"paired_vs_selected_s0":-2.1879971389056,"paired_se":0.9707263422722111,"failures":[]},{"strategy":"S1","scale":0.5,"risk":4,"n":500,"mean":-3.9346555349728,"paired_vs_selected_s0":1.7199646800768,"paired_se":0.19596254114218725,"failures":[]}],"next":"The tail rule clears a two-SE gain against S0 on this holdout; test it as a sizing-only graft using the host probability and adverse-loss model. This does not establish transfer to a stronger strategy. Do not promote the standalone S1 over B.","code":"import math\nfrom statistics import NormalDist\nfrom orderbook_pm_challenge.types import CancelAll, PlaceOrder, Side\n\nclass Strategy:\n    # Only the sizing expression selected by TAIL differs between S0 and S1.\n    TAIL = True\n    SCALE = 0.5\n    RISK = 4\n\n    def on_step(self, state):\n        actions = [CancelAll()]\n        bid = state.competitor_best_bid_ticks\n        ask = state.competitor_best_ask_ticks\n        lo = bid if bid is not None else 0\n        hi = ask if ask is not None else 100\n        p = max(0.005, min(0.995, (lo + hi) / 200.0))\n        # Midpoint proxy; no latent-state or sampled-parameter access.\n        sigma = 0.007 * math.sqrt(2000 / max(10, state.steps_remaining))\n        sigma *= 4 * p * (1 - p)\n        sigma = max(0.001, sigma)\n        cash = max(0.0, state.cash - 0.01)\n        for side, tick in ((Side.BUY, math.floor(p * 100 - 1.5)),\n                           (Side.SELL, math.ceil(p * 100 + 1.5))):\n            if not 1 <= tick <= 99:\n                continue\n            # Only quote ahead of competitor liquidity: no hidden queue estimate.\n            if side == Side.BUY and tick <= lo:\n                continue\n            if side == Side.SELL and tick >= hi:\n                continue\n            price = tick / 100.0\n            edge = p - price if side == Side.BUY else price - p\n            z = edge / sigma\n            # Gaussian adverse-selection loss plus a conservative jump allowance.\n            adverse = sigma * math.exp(-z*z/2) / math.sqrt(2*math.pi)\n            adverse -= edge * 0.5 * math.erfc(z / math.sqrt(2))\n            adverse += 0.00015\n            arrival = 0.1265\n            mean = 4.488\n            denominator = max(p, 0.05) if side == Side.BUY else price\n            reward = arrival * edge\n            size = self.SCALE * mean / denominator * reward / (reward + adverse)\n            if self.TAIL:\n                # Marginal retail edge = arrival*edge*Pr(order size > q).\n                # Solve against linear per-share adverse loss using sigma_size=1.2.\n                tail = min(0.999, max(0.001, self.RISK * adverse / reward))\n                if self.RISK * adverse >= reward:\n                    size = 0.0\n                else:\n                    quantile = NormalDist().inv_cdf(1 - tail)\n                    size = self.SCALE * mean / denominator * math.exp(1.2*quantile - 0.72)\n            collateral = price if side == Side.BUY else 1 - price\n            covered = state.yes_inventory if side == Side.SELL else 0.0\n            size = min(size, covered + cash / collateral, 500.0)\n            size = math.floor(max(0.0, size) * 100) / 100\n            if size >= 0.01:\n                actions.append(PlaceOrder(side, tick, size))\n                cash -= max(0.0, size - covered) * collateral\n        return actions\n","code_sha256":"ab25358da54a116b52a14e3ff7b0e8f7cb493caa0d423c1d38170c3c8b280400","code_s0":"import math\nfrom statistics import NormalDist\nfrom orderbook_pm_challenge.types import CancelAll, PlaceOrder, Side\n\nclass Strategy:\n    # Only the sizing expression selected by TAIL differs between S0 and S1.\n    TAIL = False\n    SCALE = 0.5\n    RISK = 4\n\n    def on_step(self, state):\n        actions = [CancelAll()]\n        bid = state.competitor_best_bid_ticks\n        ask = state.competitor_best_ask_ticks\n        lo = bid if bid is not None else 0\n        hi = ask if ask is not None else 100\n        p = max(0.005, min(0.995, (lo + hi) / 200.0))\n        # Midpoint proxy; no latent-state or sampled-parameter access.\n        sigma = 0.007 * math.sqrt(2000 / max(10, state.steps_remaining))\n        sigma *= 4 * p * (1 - p)\n        sigma = max(0.001, sigma)\n        cash = max(0.0, state.cash - 0.01)\n        for side, tick in ((Side.BUY, math.floor(p * 100 - 1.5)),\n                           (Side.SELL, math.ceil(p * 100 + 1.5))):\n            if not 1 <= tick <= 99:\n                continue\n            # Only quote ahead of competitor liquidity: no hidden queue estimate.\n            if side == Side.BUY and tick <= lo:\n                continue\n            if side == Side.SELL and tick >= hi:\n                continue\n            price = tick / 100.0\n            edge = p - price if side == Side.BUY else price - p\n            z = edge / sigma\n            # Gaussian adverse-selection loss plus a conservative jump allowance.\n            adverse = sigma * math.exp(-z*z/2) / math.sqrt(2*math.pi)\n            adverse -= edge * 0.5 * math.erfc(z / math.sqrt(2))\n            adverse += 0.00015\n            arrival = 0.1265\n            mean = 4.488\n            denominator = max(p, 0.05) if side == Side.BUY else price\n            reward = arrival * edge\n            size = self.SCALE * mean / denominator * reward / (reward + adverse)\n            if self.TAIL:\n                # Marginal retail edge = arrival*edge*Pr(order size > q).\n                # Solve against linear per-share adverse loss using sigma_size=1.2.\n                tail = min(0.999, max(0.001, self.RISK * adverse / reward))\n                if self.RISK * adverse >= reward:\n                    size = 0.0\n                else:\n                    quantile = NormalDist().inv_cdf(1 - tail)\n                    size = self.SCALE * mean / denominator * math.exp(1.2*quantile - 0.72)\n            collateral = price if side == Side.BUY else 1 - price\n            covered = state.yes_inventory if side == Side.SELL else 0.0\n            size = min(size, covered + cash / collateral, 500.0)\n            size = math.floor(max(0.0, size) * 100) / 100\n            if size >= 0.01:\n                actions.append(PlaceOrder(side, tick, size))\n                cash -= max(0.0, size - covered) * collateral\n        return actions\n","tuning_score":{"seeds":"115000..115499","n":500,"s1_mean":-3.9346555349728,"s0_mean":-5.6546202150496},"paired":{"seeds":"200000..200999","n":1000,"complete":true,"mechanism_diff":1.8732341278019,"mechanism_se":0.14374148418939942,"mechanism_retail_diff":-2.5557101156283,"mechanism_arb_diff":4.4289442434324,"reference_diff":-105.1284176733217,"reference_se":6.103915320570419,"s1_mean":-5.4325395209394,"s0_mean":-7.3057736487413,"b_mean":99.6958781523823,"failures":[]},"graft":"For a quote ahead of competitor liquidity, replace mean-flow size by the lognormal share-size quantile with survival A/(side_arrival_rate*edge). Use notional/ask for retail buys and notional/max(p_est,0.05) for retail sells. Estimate A from the host model; retain its quotes, collateral caps, and probability estimate.","seconds_per_simulation":0.22457299653324297,"timing":{"benchmark_seeds":"115000..115019","n":20,"benchmark_strategy":"S0 scale=1; repeated during tuning, not additional seeds","all_run_mean_wall_seconds":0.49640423757317087,"note":"Per-simulation wall times under four-process contention are not aggregate throughput."},"scoreboard":{"s0_code_sha256":"8c7ec9ae41b1cb0fbb2a9af9cd88ec6f65b38f6a23b40b9a1cb1c9fc883da0d0","s1_code_sha256":"ab25358da54a116b52a14e3ff7b0e8f7cb493caa0d423c1d38170c3c8b280400","reference_scores_sha256":"bde43171c0191751c603d429d3c46ce55bae11086c2826bbb386c9e7ad58c470","reference_code_sha256":null,"note":"B code was intentionally not read; only reference total-edge numbers and their integrity metadata were used. Candidate tuning paired comparisons are in tried; holdout comparison is paired."},"per_seed":{"start":200000,"n":1000,"columns":["s0_total","s1_total","mechanism_retail_diff","mechanism_arb_diff","b_total"],"encoding":"base64(zlib(little-endian float64)), row-major; row i has seed 200000+i. Decode with base64.b64decode, zlib.decompress, struct.unpack(\"<\"+\"d\"*5000,raw). b_total is fixture.H value i from the cited numeric source.","sha256":"8a8657a531e9033bdcdf3b72e88521cb3be8862ba9ef2ee249ab67016cae1cd3","data":"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"}}