{"approach":{"name":"Full-sweep marginal retail-tail exposure cap","blindness":"Built from the pinned simulator and its README/rules only; the supplied reference was accessed only as fixture.H numeric data. No other strategy or earlier network result was read.","engine_audit":{"event_order":"Participant actions, latent advance, arbitrage, retail. The callback cannot react between the move and the sweep. Source: https://raw.githubusercontent.com/danrobinson/prediction-market-challenge/b244530ee2a6247b5f930830eeca9dd950da0e7e/orderbook_pm_challenge/engine.py","orders_prices_quantities":"First sweep asks from lowest price, FIFO at ties, while ask < probability - 1e-9. Then sweep bids from highest price, FIFO at ties, while bid > probability + 1e-9. Participant and competitor orders are both swept. Each stale order is completely filled at its own resting price; no arbitrage quantity/notional budget. Better-priced competitor depth cannot shield a stale order. Source: https://raw.githubusercontent.com/danrobinson/prediction-market-challenge/b244530ee2a6247b5f930830eeca9dd950da0e7e/orderbook_pm_challenge/market.py","edge":"Arbitrage edge is nonpositive: q*(p-x) on bids and q*(x-p) on asks. Retail follows the sweep and has finite, lognormal order sizes. Sources: https://raw.githubusercontent.com/danrobinson/prediction-market-challenge/b244530ee2a6247b5f930830eeca9dd950da0e7e/docs/orderbook_prediction_market_challenge.md and https://raw.githubusercontent.com/danrobinson/prediction-market-challenge/b244530ee2a6247b5f930830eeca9dd950da0e7e/orderbook_pm_challenge/retail.py"},"baseline":"S0 estimates probability as the competitor best-quote midpoint (missing bid=0, ask=100), quotes symmetrically at a tuned minimum width or 1.2 estimated diffusion standard deviations, and sizes each side as a tuned multiple of mean retail notional converted to shares, capped at 120 shares and available collateral. It cancels/requotes each step. The full source is under 150 lines.","exposure_rule":"S1 changes ONLY RISK_CAP from zero to the tuned value. For quote margin m and quantity q, full-sweep marginal loss is L=E[(adverse move-m)+]. Approximate marginal retail benefit is 0.1265*m*Pr(V>q*c), where V is lognormal retail notional and c converts shares to notional. Cap q at the tail quantile solving Pr(V>q*c)=RISK_CAP*L/(0.1265*m); retain S0 prices and all other logic.","model_assumptions":"L uses a zero-mean normal mixture: diffusion sd=0.02*phi(Phi^-1(p))/sqrt(H*(0.02^2+0.0019*0.4^2)); jump sd=0.4 times the same derivative, capped at 0.45. Jump weight=0.0019. These are approximate range-midpoint priors, not inferred private parameters. Retail mean notional=4.488, arrival/side rate=0.1265, log-size sd=1.2. Source for published ranges/defaults: https://raw.githubusercontent.com/danrobinson/prediction-market-challenge/b244530ee2a6247b5f930830eeca9dd950da0e7e/orderbook_pm_challenge/config.py","limitations":"The midpoint estimator is crude, and the cap approximates the probability law and retail access, ignores volume ahead and multiple retail arrivals, and does not reoptimize prices. Positive mechanism gain alone does not establish competitiveness with B.","protocol":"Four unique S0 variants on all 500 tuning seeds, choose highest mean, freeze S0; four S1 risk multipliers with that S0, choose highest mean, freeze S1. Timing repeated 20 tuning seeds of the existing width=1,size=4 S0 variant, not a fifth variant; those duplicate timing results are not counted. Holdout: all 1000 requested seeds, unchanged pinned simulator, default ChallengeConfig and ParameterVariance, 2000 steps, run_batch(...n_simulations=1,seed_start=seed,workers=1), no sandbox. Four OS processes covered disjoint contiguous seed ranges. No other simulation seeds were used.","checks":"All 19 downloaded source-file Git blob hashes matched the pinned commit tree. 1,920 boundary/collateral cases passed for the frozen pair. Deterministic market unit checks verified ask-before-bid sweep order, resting execution prices, complete fills, equality exclusion, and absence of a sweep quantity budget. Exact source hashes, seed coverage, pair matching, failure status and reference payload hash checked locally."},"tried":[{"variant":"s0-w1-q2","n":500,"seeds":[115000,115499],"total_edge_mean":-48.7697246454486,"retail_edge_mean":52.394719115725,"arb_edge_mean":-101.1644437611648,"code_sha256":"cfe79db8e06f781a7a9f317cc25e63478e541aa8a06f2b633e41156e30432e98","paired_diff":-27.9127080797474,"paired_se":1.9740865845674207,"against":"s0-w2-q2"},{"variant":"s0-w1-q4","n":500,"seeds":[115000,115499],"total_edge_mean":-76.5508439902668,"retail_edge_mean":55.009967773453,"arb_edge_mean":-131.5608117637166,"code_sha256":"63f281b3909854f47d998b5d0988e7618f8b3daa86865b1e4f8d46380177cdd1","paired_diff":-55.6938274245656,"paired_se":3.236514013802993,"against":"s0-w2-q2"},{"variant":"s0-w2-q2","n":500,"seeds":[115000,115499],"total_edge_mean":-20.8570165657012,"retail_edge_mean":6.7169700597565996,"arb_edge_mean":-27.5739866254574,"code_sha256":"8cf0b258027e9fe64c40e7426ec692600145118017f8b4c3691156f1e103d42b","width":2.0,"size":2.0,"paired_diff":0.0,"paired_se":0.0,"against":"s0-w2-q2"},{"variant":"s0-w2-q4","n":500,"seeds":[115000,115499],"total_edge_mean":-43.004902740425,"retail_edge_mean":7.828133905229,"arb_edge_mean":-50.833036645654,"code_sha256":"21b622ff2ffbe1a12db14fd66425e13b12aae538a647afb85b756edc9663c757","paired_diff":-22.1478861747238,"paired_se":0.9641566992320005,"against":"s0-w2-q2"},{"variant":"s1-r0.5","n":500,"seeds":[115000,115499],"total_edge_mean":-19.8604689818742,"retail_edge_mean":6.714413357631,"arb_edge_mean":-26.5748823395048,"code_sha256":"f8063537f20faa3b9f5e728f8bc985d5122b112395bead1d97c1178976ad15aa","paired_diff":0.996547583827,"paired_se":0.08774552280781629,"against":"s0-w2-q2"},{"variant":"s1-r1","n":500,"seeds":[115000,115499],"total_edge_mean":-17.1703933732484,"retail_edge_mean":6.6952418761476,"arb_edge_mean":-23.8656352493964,"code_sha256":"36e634d335b58005d9c569a609eb075f640f496b9bb776db2f379688f85de518","paired_diff":3.6866231924528,"paired_se":0.2959421867754136,"against":"s0-w2-q2"},{"variant":"s1-r2","n":500,"seeds":[115000,115499],"total_edge_mean":-13.8106435064632,"retail_edge_mean":6.6413782187366,"arb_edge_mean":-20.4520217252004,"code_sha256":"4a20eb2c0e20d3e9c36428492630b0c21f4712b7a5b941d9a9abc4b0b5a3a123","paired_diff":7.046373059238,"paired_se":0.5109795144049871,"against":"s0-w2-q2"},{"variant":"s1-r4","n":500,"seeds":[115000,115499],"total_edge_mean":-9.2649755502234,"retail_edge_mean":6.4903656394346,"arb_edge_mean":-15.7553411896576,"code_sha256":"3555e077a3946c7baee5ecc011341153f8a74ee836be301b88a94b5e32d236ec","paired_diff":11.5920410154778,"paired_se":0.7272790644152065,"against":"s0-w2-q2","width":2.0,"size":2.0,"risk":4.0}],"next":"The cap improves this independent S0 by more than two paired standard errors. Test it as a graft on a stronger probability estimator using new authorized tuning/holdout seeds; retain it only if paired gains replicate. S1 trails supplied reference B, so do not submit S1 as a replacement for B. Finalist seeds remain untouched.","code":"import math\nfrom statistics import NormalDist\nfrom orderbook_pm_challenge.types import CancelAll, PlaceOrder, Side\n\nNORMAL = NormalDist()\nWIDTH = 2.0\nSIZE = 2.0\nRISK_CAP = 4.0\n\n\nclass Strategy:\n    def on_step(self, state):\n        b = state.competitor_best_bid_ticks\n        a = state.competitor_best_ask_ticks\n        if b is None:\n            b = 0\n        if a is None:\n            a = 100\n        p = max(0.005, min(0.995, (a + b) / 200.0))\n        # A diffusion-plus-jump normal approximation, using only public time.\n        horizon = max(1, state.steps_remaining)\n        scale = math.sqrt(horizon * (0.02 ** 2 + 0.0019 * 0.4 ** 2))\n        z = NORMAL.inv_cdf(p)\n        density = math.exp(-0.5 * z * z) / math.sqrt(2 * math.pi)\n        sigma = max(0.0001, 0.02 * density / scale)\n        jump_sigma = min(0.45, 0.4 * density / scale)\n        width = max(WIDTH / 100.0, 1.2 * sigma)\n        bid = math.floor((p - width) * 100 + 1e-9)\n        ask = math.ceil((p + width) * 100 - 1e-9)\n        actions = [CancelAll()]\n        cash = max(0.0, state.cash - 0.01)\n        for side, tick in ((Side.BUY, bid), (Side.SELL, ask)):\n            if not 1 <= tick <= 99:\n                continue\n            price = tick / 100.0\n            conversion = max(p, 0.05) if side is Side.BUY else price\n            # Several expected retail order sizes, with a finite collateral cap.\n            quantity = min(120.0, SIZE * 4.488 / conversion)\n            if RISK_CAP:\n                margin = p - price if side is Side.BUY else price - p\n                def shortfall(sd):\n                    t = margin / sd\n                    return sd * math.exp(-0.5*t*t)/math.sqrt(2*math.pi) - margin * (1-NORMAL.cdf(t))\n                # Arbitrage sweeps ALL stale quantity before retail. Each extra\n                # share bears full expected shortfall; retail reaches it only\n                # in the lognormal size tail. Cap at marginal break-even.\n                loss = 0.9981 * shortfall(sigma) + 0.0019 * shortfall(math.hypot(sigma, jump_sigma))\n                tail = RISK_CAP * loss / max(1e-12, 0.1265 * margin)\n                if tail >= 1:\n                    continue\n                tail = max(1e-9, tail)\n                notional = math.exp(math.log(4.488) - 0.5 * 1.2**2 + 1.2 * NORMAL.inv_cdf(1-tail))\n                quantity = min(quantity, notional / conversion)\n            cost = price if side is Side.BUY else 1-price\n            covered = 0.0 if side is Side.BUY else state.yes_inventory\n            affordable = cash / cost + covered\n            quantity = math.floor(min(quantity, affordable) * 100) / 100\n            if quantity < 0.01:\n                continue\n            cash -= cost * max(0.0, quantity-covered)\n            actions.append(PlaceOrder(side, tick, quantity))\n        return actions\n","code_sha256":"3555e077a3946c7baee5ecc011341153f8a74ee836be301b88a94b5e32d236ec","code_s0":"import math\nfrom statistics import NormalDist\nfrom orderbook_pm_challenge.types import CancelAll, PlaceOrder, Side\n\nNORMAL = NormalDist()\nWIDTH = 2.0\nSIZE = 2.0\nRISK_CAP = 0.0\n\n\nclass Strategy:\n    def on_step(self, state):\n        b = state.competitor_best_bid_ticks\n        a = state.competitor_best_ask_ticks\n        if b is None:\n            b = 0\n        if a is None:\n            a = 100\n        p = max(0.005, min(0.995, (a + b) / 200.0))\n        # A diffusion-plus-jump normal approximation, using only public time.\n        horizon = max(1, state.steps_remaining)\n        scale = math.sqrt(horizon * (0.02 ** 2 + 0.0019 * 0.4 ** 2))\n        z = NORMAL.inv_cdf(p)\n        density = math.exp(-0.5 * z * z) / math.sqrt(2 * math.pi)\n        sigma = max(0.0001, 0.02 * density / scale)\n        jump_sigma = min(0.45, 0.4 * density / scale)\n        width = max(WIDTH / 100.0, 1.2 * sigma)\n        bid = math.floor((p - width) * 100 + 1e-9)\n        ask = math.ceil((p + width) * 100 - 1e-9)\n        actions = [CancelAll()]\n        cash = max(0.0, state.cash - 0.01)\n        for side, tick in ((Side.BUY, bid), (Side.SELL, ask)):\n            if not 1 <= tick <= 99:\n                continue\n            price = tick / 100.0\n            conversion = max(p, 0.05) if side is Side.BUY else price\n            # Several expected retail order sizes, with a finite collateral cap.\n            quantity = min(120.0, SIZE * 4.488 / conversion)\n            if RISK_CAP:\n                margin = p - price if side is Side.BUY else price - p\n                def shortfall(sd):\n                    t = margin / sd\n                    return sd * math.exp(-0.5*t*t)/math.sqrt(2*math.pi) - margin * (1-NORMAL.cdf(t))\n                # Arbitrage sweeps ALL stale quantity before retail. Each extra\n                # share bears full expected shortfall; retail reaches it only\n                # in the lognormal size tail. Cap at marginal break-even.\n                loss = 0.9981 * shortfall(sigma) + 0.0019 * shortfall(math.hypot(sigma, jump_sigma))\n                tail = RISK_CAP * loss / max(1e-12, 0.1265 * margin)\n                if tail >= 1:\n                    continue\n                tail = max(1e-9, tail)\n                notional = math.exp(math.log(4.488) - 0.5 * 1.2**2 + 1.2 * NORMAL.inv_cdf(1-tail))\n                quantity = min(quantity, notional / conversion)\n            cost = price if side is Side.BUY else 1-price\n            covered = 0.0 if side is Side.BUY else state.yes_inventory\n            affordable = cash / cost + covered\n            quantity = math.floor(min(quantity, affordable) * 100) / 100\n            if quantity < 0.01:\n                continue\n            cash -= cost * max(0.0, quantity-covered)\n            actions.append(PlaceOrder(side, tick, quantity))\n        return actions\n","tuning_score":{"seeds":[115000,115499],"n":500,"s1_mean":-9.2649755502234,"s0_mean":-20.8570165657012,"selected_s0":"s0-w2-q2","selected_s1":"s1-r4","mechanism_diff":11.5920410154778,"mechanism_se":0.7272790644152065,"selection":"Highest total-edge mean on tuning only; no holdout-based tuning."},"paired":{"seeds":[200000,200999],"n":1000,"complete":true,"mechanism_diff":10.1517497548782,"mechanism_se":0.47492885312875416,"mechanism_retail_diff":-0.2243609679304,"mechanism_arb_diff":10.3761107228079,"reference_diff":-108.6786106362596,"reference_se":5.505568976759615,"s1_mean":-8.9827324838773,"s0_mean":-19.1344822387555,"b_mean":99.6958781523823,"failures":[],"mechanism_retail_se":0.020430662867478046,"mechanism_arb_se":0.4800509787156711,"s1_retail_mean":6.3459377951502995,"s0_retail_mean":6.5702987630807,"s1_arb_mean":-15.3286702790259,"s0_arb_mean":-25.7047810018338,"se_definition":"Sample standard deviation of per-seed paired differences (ddof=1) divided by sqrt(n). Every seed retained; no trimming.","reference":{"url":"https://api.imd.fun/artifacts/04e37cd4e1a420b91cd7a012f416e6ebf1002c5c56867981741a3a77c72b9c48","fixture":"H","scores_sha256":"bde43171c0191751c603d429d3c46ce55bae11086c2826bbb386c9e7ad58c470","used_indices":[0,999],"decoding":"raw=zlib.decompress(base64.b64decode(scores_f64le_zlib_b64)); assert sha256(raw).hexdigest()==scores_sha256 and len(raw)==8*n; values=struct.unpack('<'+'d'*n,raw); values[i] is total_edge for seed start+i"},"per_seed":{"encoding":"base64(zlib(row-major little-endian float64)); five columns, seed=200000+row_index","columns":["s1_total_edge","s0_total_edge","s1_minus_s0_retail_edge","s1_minus_s0_arb_edge","b_total_edge"],"n":1000,"sha256":"05c36b634e7b85eafd945dee2fda0e145d13dd6328f4219effbeb523ffac1f3a","data":"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"}},"graft":"Keep prices unchanged. Estimate per-share expected loss L from the next probability move past each quote. Cap size at the retail-notional tail quantile where arrival_rate/2 * quote_margin * P(retail_notional > size*conversion) equals risk_multiplier*L. Full sweeps make every extra share bear L; queue depth offers no protection.","seconds_per_simulation":0.3596387133002281,"scoreboard":{"s0_code_sha256":"8cf0b258027e9fe64c40e7426ec692600145118017f8b4c3691156f1e103d42b","s1_code_sha256":"3555e077a3946c7baee5ecc011341153f8a74ee836be301b88a94b5e32d236ec","tuning_candidates":"tried contains all eight variants with code hashes and paired differences against frozen S0.","finalist_seeds":"900000..903999 untouched. No promotion claimed; this is one blind mechanism experiment."},"runtime":{"timing_seeds":[115000,115019],"timing_n":20,"holdout_mean_seconds_per_simulation":0.3802470751140499}}