{"approach": "Preserves the base strategy on spreads 1-3 using per-instance copies of its parameter dictionaries, with overrides only after the unchanged self.s inference identifies spread 4. For spread 4, extends the immediate-cover horizon W from 50 to 200 steps; the separate cheap crossing-bank horizon Wf stays 2000. On spread-4 both sides, replaces the fixed two-share minimum for retaining a front-of-competitor queue position with a $2 threshold divided by the quote price for asks (or estimated fair price for bids), floored at $0.05. This replenishes thin low-price queues before the competitor absorbs the remainder of a retail order. Raising the share cap, cash allocations or cheap crossing-bank targets did not improve screened net edge; the retained changes concern immediate-cover planning and queue quantity. Aggressive replenishment of thin low-price queues under the one-step value rule exposed too much stock to rare jumps. All reported final scores use the public simulator without changes to its engine or configuration. The frozen file is evaluated on all 4000 tuning seeds and once on all 4000 untouched holdout seeds.", "blocked": null, "python_version": "3.12.3", "base_mean_edge": 109.27254289017199, "base_holdout_mean_edge": 98.28063388702982, "score": {"mean_edge": 98.84752384692902, "retail_edge": 124.52530289979967, "arb_edge": -25.6777790528859, "simulations": 4000, "seed_start": 4000, "failures": 0}, "holdout": {"mean_edge": 98.40769532192527, "retail_edge": 126.21554314516477, "arb_edge": -27.807847823237275, "simulations": 4000, "seed_start": 50000, "failures": 0}, "by_spread": [{"spread": 1, "runs": 1016, "base": 35.099921878704336, "yours": 35.099921878704336, "diff": 0.0, "standard_error": 0.0}, {"spread": 2, "runs": 998, "base": 83.51205221971222, "yours": 83.51205221971222, "diff": 0.0, "standard_error": 0.0}, {"spread": 3, "runs": 1017, "base": 122.97345583671554, "yours": 122.97345583671554, "diff": 0.0, "standard_error": 0.0}, {"spread": 4, "runs": 969, "base": 153.82041508580312, "yours": 154.34492049300826, "diff": 0.5245054072051594, "standard_error": 0.1582480941681672}], "constants": {"spread4.P.W": 200, "spread4.queue_notional": 2.0, "spread4.queue_price_floor": 0.05}, "code": "\"\"\"Stale-quote defence with tick probes, a sub-tick belief and cheap YES stock.\n\nEvery step the arbitrageur removes any order on the wrong side of p, so tiny\ntwo-sided probes around the estimate locate p to a tick; a grid filter narrows\nthat further.  Quotes sit one tick inside the competitor and are sized by the\nmarginal rule: one more share is worth quoting while the chance a retail order\nreaches it, times its edge, exceeds the expected arbitrage loss.  Near p=0 a\nretail buy pays almost its whole notional as edge, so YES is held to cover low\nasks.  That YES is bought when it is cheap: a large bid rests at tick 1 while p\nis above it, so the arbitrageur fills it on the crossing for about 0.1 tick a\nshare instead of nearly a full tick once p is near zero, and smaller bids sit on\nwhichever tick p is within 0.15 of.  Below tick 1 the estimate of p decays with\nthe time spent there, since no probe can see it.\n\"\"\"\nimport math\nfrom orderbook_pm_challenge.strategy import BaseStrategy\nfrom orderbook_pm_challenge.types import CancelOrder, PlaceOrder, Side\n\nP = dict(lam_r=0.25, m=5.0, sig_n=1.2, lam_j=0.0019, sig_j=0.35, sig=0.02,\n         qmax=3000.0, rmin=2, kA=0.7, kL=0.15, pL=0.05,\n         ammo=1.0, ydem=0.4, W=200.0, fy=0.6, ammo_q=60.0, ammo_pmax=0.05, ammo_maxpay=0.02,\n         fb=0.9, fu=0.7, xu=0.1, sticky=1.0, min_ahead=2.0, keep_hi=1.5, keep_lo=0.7, Hu=300.0, bins=8.0, filt=1.0, uni=0.01, jumpinfo=1.0, dmin=0.0, kLb=0.08, kLf=0.08, pinf=1.0, pre=0.0, H1=100.0,\n         ydf=0.6, Wf=2000.0, g_hi=0.15, g_lo=0.15, qf=400.0, pmaxf=0.05, pre_gap=1.0,\n         ydf1=0.6, g1=5.0, f1=0.6, a1=100.0, nb1=0.0, gs=1.0, p1t=1.0, m1=1.0, zd=1.0, pfl=0.0005)\nP.update({'W': 50, 'ammo_pmax': 0.02, 'fy': 0.7, 'fu': 0.25, 'Hu': 500})\nC = {'W': 50, 'ammo_pmax': 0.02, 'fy': 0.7, 'reserve': 200, 'scar': 0.0025, 'fu': 0.25, 'Hu': 500}\nSQ2 = math.sqrt(2.0)\nSQ2PI = math.sqrt(2.0 * math.pi)\ndef cdf(x): return 0.5 * (1.0 + math.erf(x / SQ2))\ndef pdf(x): return math.exp(-0.5 * x * x) / SQ2PI\ndef inv(p):\n    p = min(max(p, 1e-9), 1 - 1e-9)\n    t = math.sqrt(-2.0 * math.log(min(p, 1 - p)))\n    x = t - (2.515517 + 0.802853 * t + 0.010328 * t * t) / (1 + 1.432788 * t + 0.189269 * t * t + 0.001308 * t ** 3)\n    x = -x if p < 0.5 else x\n    for _ in range(2):\n        d = pdf(x)\n        if d < 1e-12:\n            break\n        x -= (cdf(x) - p) / d\n    return x\ndef lpm(k):\n    return pdf(k) + k * cdf(k)\ndef fl(q):\n    return math.floor(q * 100 + 1e-9) / 100\n\nNODES = [(-2.2, .06), (-1.3, .17), (-0.6, .27), (0.6, .27), (1.3, .17), (2.2, .06)]\nPROBE = 0.01\n\n\nclass Strategy(BaseStrategy):\n    def __init__(self):\n        self.n = 0\n        self.ahead = set()     # main orders placed strictly ahead of any competitor order at their tick\n        self.res = {}          # order id -> uncovered quantity at placement\n        self.bc = None         # belief: sub-tick grid centres (ticks)\n        self.bw = None         # belief: weights\n        self.st = 0.5          # last one-step sd in ticks\n        self.pcb = self.pca = None\n        self.pb = set()\n        self.pa = set()\n        self.tb = 0\n        self.s = None          # competitor spread parameter\n        self.params = P.copy()\n        self.controls = C.copy()\n\n    def _oid(self, tag):\n        self.n += 1\n        return tag + str(self.n)\n\n    def on_step(self, state):\n        H = max(1, state.steps_remaining)\n        cb = state.competitor_best_bid_ticks or 0\n        ca = state.competitor_best_ask_ticks or 100\n        if self.s is None:\n            self.s = max(1, (ca - cb + 1) // 2)\n            if self.s == 4:\n                self.params.update({'W': 200})\n                self.controls.update({'queue_notional': 2.0, 'queue_price_floor': 0.05})\n        P = self.params\n        C = self.controls\n        lo, hi = cb, ca\n        for o in state.own_orders:\n            if o.side is Side.BUY:\n                lo = max(lo, o.price_ticks)\n            else:\n                hi = min(hi, o.price_ticks)\n        if P['pinf'] > 0:\n            sb = {o.price_ticks for o in state.own_orders if o.side is Side.BUY and o.remaining_quantity <= PROBE + 1e-9}\n            sa = {o.price_ticks for o in state.own_orders if o.side is Side.SELL and o.remaining_quantity <= PROBE + 1e-9}\n            ea = self.pa - sa\n            eb = self.pb - sb\n            if sa and ea:\n                lo = max(lo, min(max(ea), hi))\n            if eb and (sb or min(eb) <= P['p1t']):\n                hi = min(hi, max(min(eb), lo))\n        if P['jumpinfo'] > 0 and self.pcb is not None:\n            if ca > self.pca + 1:\n                lo = max(lo, ca - 1)\n            if cb < self.pcb - 1:\n                hi = min(hi, cb + 1)\n        self.pcb, self.pca = cb, ca\n        if lo > hi:\n            lo, hi = hi, lo\n        # sub-tick belief over p_(t-1): predict with one-step noise, truncate to [lo, hi]\n        wid = max(hi - lo, 1e-6)\n        nb = max(2, min(48, int(round(wid * P['bins']))))\n        cs = [lo + (i + 0.5) * wid / nb for i in range(nb)]\n        if self.bc is None or P['filt'] <= 0:\n            ws = [1.0 / nb] * nb\n        else:\n            sg = max(self.st, 0.6 * wid / nb)\n            k = -0.5 / (sg * sg)\n            ws = []\n            uni = P['uni'] * sg\n            for c in cs:\n                t = uni\n                for ci, wi in zip(self.bc, self.bw):\n                    d = c - ci\n                    d2 = d * d * k\n                    if d2 > -12.5:\n                        t += wi * math.exp(d2)\n                ws.append(t)\n            tot = sum(ws)\n            ws = [w / tot for w in ws]\n        self.bc, self.bw = cs, ws\n        mean_t = sum(c * w for c, w in zip(cs, ws))\n        var_t = max(sum(c * c * w for c, w in zip(cs, ws)) - mean_t * mean_t, 0.0)\n        phat = min(max(mean_t / 100.0, 0.002), 0.998)\n        var = P['sig'] ** 2 + P['lam_j'] * P['sig_j'] ** 2\n        sd = math.sqrt(H * var)\n        self.tb = self.tb + 1 if hi <= 1 else 0\n        if self.tb > 0 and P['zd'] > 0:\n            # below the last tick there is no probe: the score keeps walking away from it\n            phat = min(phat, max(P['pfl'], cdf(inv(0.01) - P['zd'] * 0.02 * math.sqrt(self.tb) / sd)))\n        z = inv(phat)\n        Z = z * sd\n        sp = pdf(z) / sd * P['sig']\n        self.st = sp * 100.0\n        s = math.sqrt(sp * sp + var_t / 1e4 + 1e-8)\n        sd1 = math.sqrt(max(1.0, H - 1.0) * var)\n        jp = [(cdf((Z + n * P['sig_j']) / sd1), w) for n, w in NODES]\n        lj = P['lam_j']\n\n        def AB(x):\n            k = (x - phat) / s\n            a = s * lpm(k)\n            b = a - (x - phat)\n            ja = sum(w * max(0.0, x - p) for p, w in jp)\n            jb = sum(w * max(0.0, p - x) for p, w in jp)\n            return (1 - lj) * a + lj * ja, (1 - lj) * b + lj * jb\n\n        if H == 1:\n            def AB(x):\n                return (1 - phat) * x, phat * (1 - x)\n        rho = P['lam_r'] * 0.5\n        mu = math.log(P['m']) - 0.5 * P['sig_n'] ** 2\n        kA = P['kL'] if phat < P['pL'] else P['kA']\n\n        def qsize(loss, gain, px):\n            loss *= kA\n            if gain <= 0:\n                return 0.0\n            tail = loss / (rho * gain)\n            if tail >= 1.0:\n                return 0.0\n            return min(P['qmax'], math.exp(mu + P['sig_n'] * inv(1 - max(tail, 1e-6))) / px)\n\n        ptk = phat * 100\n        def qbid(t):\n            if not 1 <= t <= 99:\n                return 0.0\n            a, b = AB(t / 100.0)\n            if phat * 100 - t < P['dmin']:\n                return 0.0\n            if P['nb1'] > 0 and self.s == 1 and t == 1 and ptk < 1.0:\n                return 0.0\n            if phat < P['pL']:\n                ytg = P['ydem'] * min(H, P['W']) / max(ca - 1, ptk + 1) * 100.0\n                a *= (P['kLb'] if state.yes_inventory < ytg else P['kLf']) / P['kL']\n            return qsize(a, b, max(phat, 0.05))\n\n        def qask(t):\n            if not 1 <= t <= 99:\n                return 0.0\n            a, b = AB(t / 100.0)\n            if t - phat * 100 < P['dmin']:\n                return 0.0\n            if phat < C.get('scar_p', .1):\n                a = max(0.0, a - C.get('scar', 0.0) * (1 - min(1.0, state.yes_inventory / C.get('stock', 3000.0))))\n            return min(C.get('askcap', P['qmax']), qsize(b, a, t / 100.0))\n\n        bids = [o for o in state.own_orders if o.side is Side.BUY and o.remaining_quantity > PROBE + 1e-9]\n        asks = [o for o in state.own_orders if o.side is Side.SELL and o.remaining_quantity > PROBE + 1e-9]\n        sticky = P['sticky'] > 0\n        def top(orders, comp, step):\n            if sticky and sum(o.remaining_quantity for o in orders if o.price_ticks == comp and o.order_id in self.ahead) >= (C['queue_notional'] / max(comp / 100.0 if step < 0 else phat, C['queue_price_floor']) if self.s == 4 else P['min_ahead']):\n                return comp\n            return comp + step\n        bid_t = top(bids, cb, 1)\n        ask_t = top(asks, ca, -1)\n        qb = min(C.get('bidcap', 3000.0), qbid(bid_t))\n        qa = qask(ask_t)\n\n        Y = state.yes_inventory\n        if ask_t == 1 and H > P['H1'] and P['pre'] > 0:\n            qa = qa if Y >= P['a1'] > 0 else 0.0          # let the competitor's 2-tick ask erode instead of fronting it for nothing\n        # ammunition: hold YES so that low asks are covered\n        ptick = phat * 100\n        amt = 0\n        qam = 0.0\n        if P['ammo'] > 0 and ask_t >= 2 and phat < max(P['ammo_pmax'], P['pmaxf']):\n            wealth = state.cash + Y * phat\n            x = max(ask_t, ptick + 1) / 100.0\n            capw = P['fy'] * wealth / max(phat, 0.005)\n            target = min(P['ydem'] * min(H, P['W']) / x, capw)\n            full = min(P['ydf'] * min(H, P['Wf']) / x, capw) if phat < P['pmaxf'] else 0.0\n            t = max(1, min(99, int(math.ceil(ptick + 1e-9))))\n            g = t - ptick\n            if Y < full and g <= P['g_hi'] and bid_t <= t < ask_t:\n                amt, qam = t, min(P['qf'], full - Y)     # a tick p has just crossed: nearly free\n            elif Y < target and phat < P['ammo_pmax'] and bid_t <= t < ask_t and t / 100.0 - phat < P['ammo_maxpay']:\n                amt, qam = t, min(P['ammo_q'], target - Y)\n            elif Y < full and t >= 2 and 1.0 - g <= P['g_lo']:\n                amt, qam = t - 1, min(P['qf'], full - Y)  # rest just under p: filled only on the crossing, or by retail\n\n        if P['pre'] > 0 and ca == 2 and phat < 0.01 and H > P['H1'] and Y < P['pre'] and not qam and 1.0 - ptick < P['pre_gap']:\n            amt, qam = 1, min(P['ammo_q'], P['pre'] - Y)\n        q1 = 0.0\n        if P['ydf1'] > 0 and P['m1'] < ptick < P['g1'] + 1.0 and amt != 1:\n            q1 = max(0.0, P['ydf1'] * min(H, P['Wf']) / (self.s / 100.0) - Y)\n        # order management\n        actions = []\n        free = state.free_cash - 2.0\n        keep_ask = 0.0\n        new_b = []\n        new_a = []\n        def cancel(o):\n            nonlocal free\n            actions.append(CancelOrder(o.order_id))\n            x = o.price_ticks / 100.0\n            if o.side is Side.BUY:\n                free += x * o.remaining_quantity\n            else:\n                free += (1 - x) * min(self.res.get(o.order_id, 0.0), o.remaining_quantity)\n            self.ahead.discard(o.order_id)\n            self.res.pop(o.order_id, None)\n        for o in state.own_orders:\n            if o.remaining_quantity <= PROBE + 1e-9:\n                cancel(o)\n        kept_ids = set()\n        def manage(orders, t, q, comp):\n            kept = 0.0\n            for o in orders:\n                if not (C.get('recover', 0) and o.side is Side.SELL and t < comp and self.res.get(o.order_id, 0.0) > 1 and state.yes_inventory > C.get('recover_yes', 1) and state.free_cash < C.get('recover_cash', 200)) and o.price_ticks == t and q > 0 and kept + o.remaining_quantity <= max(q * P['keep_hi'], 1.0):\n                    kept += o.remaining_quantity\n                    kept_ids.add(o.order_id)\n                else:\n                    cancel(o)\n            return kept\n        kb = manage(bids, bid_t, qb, cb)\n        ka = manage(asks, ask_t, qa, ca)\n        add_b = (qb - kb) if (kb < qb * P['keep_lo'] and bid_t != cb) else 0.0\n        add_a = (qa - ka) if (ka < qa * P['keep_lo'] and ask_t != ca) else 0.0\n\n        budget = P['fb'] * max(free, 0.0)\n        cost = add_b * bid_t / 100.0 + qam * amt / 100.0\n        if cost > budget:\n            f = budget / cost\n            add_b *= f\n            qam *= f\n            cost = budget\n        out = []\n        q1 = min(q1, P['f1'] * max(budget - cost, 0.0) * 100.0)\n        if q1 >= 0.01:\n            out.append(PlaceOrder(side=Side.BUY, price_ticks=1, quantity=fl(q1)))\n            free -= fl(q1) / 100.0 + 1e-6\n        if amt and qam >= 0.01:\n            q = fl(qam)\n            out.append(PlaceOrder(side=Side.BUY, price_ticks=amt, quantity=q))\n            free -= q * amt / 100.0 + 1e-6\n        if bid_t >= 1 and add_b >= 0.01:\n            q = fl(add_b)\n            oid = self._oid('b')\n            out.append(PlaceOrder(side=Side.BUY, price_ticks=bid_t, quantity=q, client_order_id=oid))\n            free -= q * bid_t / 100.0 + 1e-6\n            if bid_t > cb:\n                self.ahead.add(oid)\n        if ask_t <= 99 and add_a >= 0.01:\n            x = ask_t / 100.0\n            covk = sum(o.remaining_quantity - min(self.res.get(o.order_id, 0.0), o.remaining_quantity)\n                       for o in asks if o.order_id in kept_ids)\n            cov = max(0.0, Y - covk - 0.1)\n            unc = max(0.0, add_a - cov)\n            if unc > 0:\n                ub = max(0.0, free - (1 - P['fu']) * max(state.cash, 0.0)) if x >= P['xu'] else 0.0\n                if H < P['Hu'] and phat < 0.05:\n                    ub = max(0.0, free * (1 - H / P['Hu']))\n                ub = min(ub, max(0.0, free - C.get('reserve', 0.0) * min(1.0, H / C.get('release', 50.0))))\n                unc = min(unc, ub / (1 - x))\n            q = fl(min(add_a, cov + unc))\n            if q >= 0.01:\n                oid = self._oid('a')\n                out.append(PlaceOrder(side=Side.SELL, price_ticks=ask_t, quantity=q, client_order_id=oid))\n                r = max(0.0, q - cov)\n                self.res[oid] = r\n                free -= r * (1 - x) + 1e-6\n                if ask_t < ca:\n                    self.ahead.add(oid)\n\n        # probes: tiny orders that the arbitrageur removes when p crosses them\n        f = int(math.floor(ptick + 1e-9))\n        r = max(int(P['rmin']), int(math.ceil(2.0 * s * 100)))\n        self.pb = set()\n        self.pa = set()\n        if free > 1.0:\n            for t in range(f, f + r + 1):\n                if cb < t < ca and 1 <= t <= 99:\n                    out.append(PlaceOrder(side=Side.BUY, price_ticks=t, quantity=PROBE))\n                    self.pb.add(t)\n            for t in range(f + 1 - r, f + 2):\n                if cb < t < ca and 1 <= t <= 99:\n                    out.append(PlaceOrder(side=Side.SELL, price_ticks=t, quantity=PROBE))\n                    self.pa.add(t)\n        return actions + out\n", "code_sha256": "db82772cbc00f1ac695975bdd31d6e9b7a8e57f54b157fae9dd2c2266cd13479", "tried": ["Scorer: unmodified public simulator commit b244530ee2a6247b5f930830eeca9dd950da0e7e; standard 2000-step configuration, unsandboxed, four workers. Base seeds 1000-1599 mean_edge=109.27254289017199; zero failures. Identity refactor matched every field on all 600 runs.", "Selection uses spread-4 runs only inside seeds 4000-7999. A screens are the first 120 spread-4 seeds in 4000-5999; B screens are the first 120 in 6000-7999; C screens are the first 120 in 7000-7999, disjoint from the short B screens. The final 50000-53999 candidate evaluation is performed only after freezing. The baseline holdout was computed blind and not inspected during selection.", "Diagnostics, 40 spread-4 runs: low-p callbacks=8254; fraction with desired ask at 3000 shares=0.3922946450205961; fraction whose ask addition was policy-clipped=0.11630724497213472; fraction with cash below 200=0.042161376302398836. Quote clipping is not a measurement of missed fills. Increasing qmax to 6000 and f1 to 0.8 did not improve retail in the screens.", "Fill attribution, first 80 A runs; instrumentation preserves base total_edge exactly: {'main_ask_arb': -12.470469349931957, 'main_ask_retail': 147.40093032332518, 'main_bid_arb': -8.907488822931501, 'main_bid_retail': 40.870082365741155, 'probe_arb': -0.7583985253431836, 'probe_retail': 0.032264529048339, 'stock1_arb': -2.8101268342288885, 'stock_other_arb': -1.0521424736682086, 'stock_other_retail': 0.3287100553653449}.", "A ahead100 {'params.min_ahead': 100}: n=120, mean_edge=162.36071982677, retail_edge=188.9429755524325, arb_edge=-26.582255725544165; paired changes edge=-3.122842559897499, retail=-2.611644652899167, arb=-0.5111979069666669, edge SE=0.42800854972776164; failures=0.", "A noammo {'params.ammo': 0}: n=120, mean_edge=162.78787782772332, retail_edge=187.93239629227668, arb_edge=-25.144518464477503; paired changes edge=-2.6956845589441674, retail=-3.622223913055001, arb=0.9265393540999998, edge SE=1.4598249673510508; failures=0.", "A W0 {'params.W': 0}: n=120, mean_edge=162.79050955447082, retail_edge=188.02012183635833, arb_edge=-25.229612281806666; paired changes edge=-2.6930528321966674, retail=-3.534498368973333, arb=0.8414455367708334, edge SE=1.1451529126309534; failures=0.", "A stocklean {'params.ammo': 0, 'params.kLf': 0.3}: n=120, mean_edge=163.6443064767367, retail_edge=187.20219882880002, arb_edge=-23.557892351973333; paired changes edge=-1.8392559099308319, retail=-4.352421376531666, arb=2.5131654666041663, edge SE=1.7010092943044797; failures=0.", "A gwide {'params.g_hi': 0.3, 'params.g_lo': 0.3}: n=120, mean_edge=163.666692548615, retail_edge=191.7789784077667, arb_edge=-28.1122858590625; paired changes edge=-1.8168698380524972, retail=0.22435820243500035, arb=-2.0412280404849996, edge SE=0.7029218954357574; failures=0.", "A ydf103 {'params.ydf1': 0.3}: n=120, mean_edge=163.74048013201585, retail_edge=190.69761940866417, arb_edge=-26.957139276560834; paired changes edge=-1.7430822546516642, retail=-0.8570007966674993, arb=-0.8860814579833333, edge SE=1.1405938083595693; failures=0.", "A riskstock {'params.kA': 0.9, 'params.ammo': 0, 'params.kLf': 0.3}: n=120, mean_edge=163.78296837992832, retail_edge=185.07392342533916, arb_edge=-21.290955045395833; paired changes edge=-1.700594006739165, retail=-6.4806967799925, arb=4.7801027731816665, edge SE=1.6323462203481136; failures=0.", "A kA12 {'params.kA': 1.2}: n=120, mean_edge=163.93272458123664, retail_edge=185.766736762855, arb_edge=-21.8340121815175; paired changes edge=-1.5508378054308318, retail=-5.787883442476667, arb=4.2370456370600005, edge SE=0.6634930566151384; failures=0.", "A bidrisk07 {'controls.bid_risk': 0.7}: n=120, mean_edge=164.43312549680417, retail_edge=193.23973365983917, arb_edge=-28.806608162944165; paired changes edge=-1.0504368898633318, retail=1.6851134545075022, arb=-2.735550344366667, edge SE=0.39094444233980385; failures=0.", "A sigj045 {'params.sig_j': 0.45}: n=120, mean_edge=164.4589702703, retail_edge=191.07074942962333, arb_edge=-26.611779159268334; paired changes edge=-1.0245921163674994, retail=-0.48387077570833287, arb=-0.5407213406908338, edge SE=0.5229444284197826; failures=0.", "A askrisk07 {'controls.ask_risk': 0.7}: n=120, mean_edge=164.5105310361625, retail_edge=192.6866364331175, arb_edge=-28.176105396879166; paired changes edge=-0.9730313505049986, retail=1.1320162277858352, arb=-2.105047578301667, edge SE=0.3144701222690074; failures=0.", "A askrisk2 {'controls.ask_risk': 2.0}: n=120, mean_edge=164.52167955673667, retail_edge=187.58502368138, arb_edge=-23.063344124544997; paired changes edge=-0.9618828299308322, retail=-3.9695965239516653, arb=3.0077136940324993, edge SE=0.59170797931735; failures=0.", "A lamj003 {'params.lam_j': 0.003}: n=120, mean_edge=164.63766308299833, retail_edge=190.00000033605417, arb_edge=-25.362337252981668; paired changes edge=-0.8458993036691667, retail=-1.5546198692774997, arb=0.7087205655958331, edge SE=0.49683322631590443; failures=0.", "A qmax1500 {'params.qmax': 1500}: n=120, mean_edge=164.66572899604168, retail_edge=190.19982852201002, arb_edge=-25.53409952587417; paired changes edge=-0.8178333906258316, retail=-1.354791683321667, arb=0.5369582927033331, edge SE=0.504827971613447; failures=0.", "A kA05 {'params.kA': 0.5}: n=120, mean_edge=164.833377836235, retail_edge=194.7413539044217, arb_edge=-29.907976068059167; paired changes edge=-0.650184550432497, retail=3.1867336990900013, arb=-3.8369182494816676, edge SE=0.40815124128883706; failures=0.", "A asklow05 {'controls.ask_risk_low': 0.5}: n=120, mean_edge=164.84639515168084, retail_edge=192.5641012847217, arb_edge=-27.717706132954167; paired changes edge=-0.6371672349866667, retail=1.00948107939, arb=-1.6466483143766666, edge SE=0.8182283536488981; failures=0.", "A pmaxf10 {'params.pmaxf': 0.1}: n=120, mean_edge=164.94318840128665, retail_edge=191.7260271710625, arb_edge=-26.78283876969167; paired changes edge=-0.5403739853808329, retail=0.17140696573083408, arb=-0.7117809511141668, edge SE=0.18460692129146186; failures=0.", "A reserve0 {'controls.reserve': 0}: n=120, mean_edge=165.12466956879, retail_edge=192.09711911790166, arb_edge=-26.972449549035833; paired changes edge=-0.35889281787750005, retail=0.5424989125699989, arb=-0.9013917304583334, edge SE=0.29582817269396866; failures=0.", "A reserve400 {'controls.reserve': 400}: n=120, mean_edge=165.1340974794175, retail_edge=191.231006999285, arb_edge=-26.096909519781665; paired changes edge=-0.34946490724999996, retail=-0.3236132060466668, arb=-0.02585170120416663, edge SE=0.12192352211179541; failures=0.", "A bidrisk15 {'controls.bid_risk': 1.5}: n=120, mean_edge=165.13579458220332, retail_edge=188.9923107822175, arb_edge=-23.856516199959167; paired changes edge=-0.34776780446416455, retail=-2.5623094231141663, arb=2.214541618618333, edge SE=0.4345472166277323; failures=0.", "A ydf10 {'params.ydf': 1.0}: n=120, mean_edge=165.14786492012834, retail_edge=191.55503913785418, arb_edge=-26.407174217641668; paired changes edge=-0.33569746653916727, retail=0.0004189325225003889, arb=-0.3361163990641669, edge SE=0.09693848351840162; failures=0.", "A askrisk15 {'controls.ask_risk': 1.5}: n=120, mean_edge=165.18733344645, retail_edge=189.3477903410625, arb_edge=-24.160456894534164; paired changes edge=-0.29622894021749874, retail=-2.2068298642691664, arb=1.910600924043333, edge SE=0.4187558446781333; failures=0.", "A stock02 {'params.ydf1': 0.9, 'params.f1': 0.8, 'params.g1': 10}: n=120, mean_edge=165.18850479617416, retail_edge=192.1840993997592, arb_edge=-26.995594603507502; paired changes edge=-0.29505759049333186, retail=0.6294791944274993, arb=-0.9245367849300002, edge SE=1.0300292937921323; failures=0.", "A keepfresh {'params.keep_hi': 1.2, 'params.keep_lo': 0.9}: n=120, mean_edge=165.2592159576925, retail_edge=191.67317884486167, arb_edge=-26.413962887095; paired changes edge=-0.2243464289749985, retail=0.11855863952999991, arb=-0.3429050685175003, edge SE=0.5954948491465095; failures=0.", "A ydf109 {'params.ydf1': 0.9}: n=120, mean_edge=165.27362082428752, retail_edge=192.47047872692335, arb_edge=-27.196857902552498; paired changes edge=-0.20994156237999825, retail=0.9158585215916664, arb=-1.1258000839750004, edge SE=0.9364073424441767; failures=0.", "A fu05 {'params.fu': 0.5}: n=120, mean_edge=165.278165942875, retail_edge=191.5091150439225, arb_edge=-26.230949100960835; paired changes edge=-0.20539644379249894, retail=-0.04550516140916727, arb=-0.1598912823833333, edge SE=0.18450316112270235; failures=0.", "A asklow2 {'controls.ask_risk_low': 2.0}: n=120, mean_edge=165.32488346543, retail_edge=189.90188405672248, arb_edge=-24.577000591213334; paired changes edge=-0.15867892123749938, retail=-1.6527361486091678, arb=1.4940572273641666, edge SE=0.7661689582038639; failures=0.", "A ahead30 {'params.min_ahead': 30}: n=120, mean_edge=165.3378267860925, retail_edge=191.65070096338917, arb_edge=-26.3128741771975; paired changes edge=-0.14573560057499838, retail=0.0960807580574977, arb=-0.2418163586200004, edge SE=0.27774046282179765; failures=0.", "A kA09 {'params.kA': 0.9}: n=120, mean_edge=165.34173646741502, retail_edge=189.0156119278225, arb_edge=-23.673875460385; paired changes edge=-0.14182591925250002, retail=-2.539008277509166, arb=2.3971823581924996, edge SE=0.3621851962888301; failures=0.", "A zd15 {'params.zd': 1.5}: n=120, mean_edge=165.34769362240084, retail_edge=191.5798630702575, arb_edge=-26.232169447770833; paired changes edge=-0.13586876426666528, retail=0.02524286492583343, arb=-0.16111162919333352, edge SE=0.16054535096008282; failures=0.", "A scar0 {'controls.scar': 0}: n=120, mean_edge=165.39878449169, retail_edge=191.57988138832667, arb_edge=-26.18109689656; paired changes edge=-0.08477789497749816, retail=0.025261182995001144, arb=-0.11003907798250008, edge SE=0.05903366666323391; failures=0.", "A askcap6000 {'params.qmax': 6000}: n=120, mean_edge=165.4536169287175, retail_edge=191.52642615218917, arb_edge=-26.072809223387498; paired changes edge=-0.02994545794999984, retail=-0.028194053142498395, arb=-0.0017514048099998311, edge SE=0.04209241254352171; failures=0.", "A qmax6000 {'params.qmax': 6000, 'controls.bidcap': 6000}: n=120, mean_edge=165.45528687720082, retail_edge=191.52642615218917, arb_edge=-26.071139274904166; paired changes edge=-0.028275509466665957, retail=-0.028194053142498395, arb=-8.145632666645142e-05, edge SE=0.04206319998059078; failures=0.", "A floor001 {'params.pfl': 0.001}: n=120, mean_edge=165.47524988919167, retail_edge=191.5463068745242, arb_edge=-26.071056985245832; paired changes edge=-0.008312497475833425, retail=-0.00831333080750009, arb=8.333316666645866e-07, edge SE=0.008313337852050972; failures=0.", "A fu01 {'params.fu': 0.1}: n=120, mean_edge=165.47567055706836, retail_edge=191.49372923333667, arb_edge=-26.018058676183333; paired changes edge=-0.007891829599165411, retail=-0.060890971994998704, arb=0.052999142394166504, edge SE=0.12970478175536088; failures=0.", "A zd07 {'params.zd': 0.7}: n=120, mean_edge=165.48134097426833, retail_edge=191.5533802928875, arb_edge=-26.0720393185325; paired changes edge=-0.002221412399165897, retail=-0.0012399124441666723, arb=-0.000981499955000024, edge SE=0.0014466798367216043; failures=0.", "A fy09 {'params.fy': 0.9}: n=120, mean_edge=165.4835623866675, retail_edge=191.55462020533164, arb_edge=-26.0710578185775; paired changes edge=0.0, retail=0.0, arb=0.0, edge SE=0.0; failures=0.", "A base {}: n=120, mean_edge=165.4835623866675, retail_edge=191.55462020533164, arb_edge=-26.0710578185775; paired changes edge=0.0, retail=0.0, arb=0.0, edge SE=0.0; failures=0.", "A f108 {'params.f1': 0.8}: n=120, mean_edge=165.48413833217248, retail_edge=191.55462020533164, arb_edge=-26.070481873073334; paired changes edge=0.000575945505000656, retail=0.0, arb=0.0005759455041666565, edge SE=0.000575945505000656; failures=0.", "A f104 {'params.f1': 0.4}: n=120, mean_edge=165.48878576854335, retail_edge=191.55462020533164, arb_edge=-26.065834436702502; paired changes edge=0.0052233818758338655, retail=0.0, arb=0.005223381874999985, edge SE=0.0052233818758338655; failures=0.", "A queue_bid {}: n=120, mean_edge=165.4949690448575, retail_edge=191.6066852396325, arb_edge=-26.111716194691667; paired changes edge=0.01140665819000081, retail=0.05206503430083389, arb=-0.040658376114166804, edge SE=0.04267430863770374; failures=0.", "A floor0001 {'params.pfl': 0.0001}: n=120, mean_edge=165.60160451968335, retail_edge=191.6726623383475, arb_edge=-26.0710578185775; paired changes edge=0.11804213301583294, retail=0.11804213301583294, arb=0.0, edge SE=0.11561840052432927; failures=0.", "A queue0p5 {'controls.queue_notional': 0.5}: n=120, mean_edge=165.60937587627333, retail_edge=191.62550968720248, arb_edge=-26.0161338108275; paired changes edge=0.12581348960583316, retail=0.07088948187083274, arb=0.05492400774999971, edge SE=0.11899718139548549; failures=0.", "A value_safe {'controls.queue_factor': 1, 'controls.low_queue_notional': 0}: n=120, mean_edge=165.61586586561668, retail_edge=191.49601608923666, arb_edge=-25.8801502235375; paired changes edge=0.13230347894916747, retail=-0.05860411609500004, arb=0.19090759503999974, edge SE=0.20407056635139761; failures=0.", "A valueq1p2 {'controls.queue_factor': 1.2}: n=120, mean_edge=165.6235512830325, retail_edge=191.92994777777582, arb_edge=-26.30639649466083; paired changes edge=0.13998889636500217, retail=0.3753275724441657, arb=-0.23533867608333353, edge SE=0.47683985059057693; failures=0.", "A queue1 {'controls.queue_notional': 1}: n=120, mean_edge=165.66087456360583, retail_edge=191.74675657769833, arb_edge=-26.085882014000834; paired changes edge=0.1773121769383339, retail=0.19213637236666592, arb=-0.014824195423333601, edge SE=0.10454852539988033; failures=0.", "A ahead10 {'params.min_ahead': 10}: n=120, mean_edge=165.7251196938367, retail_edge=191.87615578780665, arb_edge=-26.15103609389333; paired changes edge=0.24155730716916762, retail=0.3215355824750001, arb=-0.07997827531583382, edge SE=0.12487946555211131; failures=0.", "A W200 {'params.W': 200}: n=120, mean_edge=165.81682969723667, retail_edge=192.28699240274167, arb_edge=-26.47016270542333; paired changes edge=0.3332673105691674, retail=0.7323721974100006, arb=-0.39910488684583323, edge SE=0.342594490441723; failures=0.", "A Hu1000 {'params.Hu': 1000}: n=120, mean_edge=165.87393422461835, retail_edge=191.90695831510916, arb_edge=-26.03302409040417; paired changes edge=0.3903718379508342, retail=0.35233810977750024, arb=0.038033728173333316, edge SE=0.1549677712202171; failures=0.", "A lamj001 {'params.lam_j': 0.001}: n=120, mean_edge=165.92490854384167, retail_edge=192.38459261766832, arb_edge=-26.459684073825; paired changes edge=0.4413461571741679, retail=0.8299724123366687, arb=-0.38862625524750055, edge SE=0.7217645327305289; failures=0.", "A queue_ask {}: n=120, mean_edge=165.94525965107331, retail_edge=192.03997796176, arb_edge=-26.0947183105975; paired changes edge=0.46169726440583464, retail=0.48535775642833257, arb=-0.023660492020000156, edge SE=0.1662217105019298; failures=0.", "A ahead_notional {}: n=120, mean_edge=165.95659413711834, retail_edge=192.09203962438, arb_edge=-26.135445487173335; paired changes edge=0.4730317504508356, retail=0.5374194190483333, arb=-0.06438766859583359, edge SE=0.18386729825298292; failures=0.", "A sigj025 {'params.sig_j': 0.25}: n=120, mean_edge=165.99783276134835, retail_edge=192.14669212781, arb_edge=-26.148859366443332; paired changes edge=0.514270374680836, retail=0.5920719224783322, arb=-0.07780154786583354, edge SE=0.7077179138959421; failures=0.", "A value_mix {'controls.queue_factor': 1, 'controls.low_queue_notional': 2}: n=120, mean_edge=166.07042225503, retail_edge=191.96472463032165, arb_edge=-25.894302375204166; paired changes edge=0.5868598683625011, retail=0.4101044249899996, arb=0.17675544337333313, edge SE=0.2497502798954426; failures=0.", "A queue8 {'controls.queue_notional': 8}: n=120, mean_edge=166.19399216117083, retail_edge=192.43701742175418, arb_edge=-26.243025260506666; paired changes edge=0.7104297745033344, retail=0.8823972164224997, arb=-0.17196744192916683, edge SE=0.26498037824419046; failures=0.", "A valueq0p8 {'controls.queue_factor': 0.8}: n=120, mean_edge=166.292790313835, retail_edge=192.26646571604, arb_edge=-25.97367540212417; paired changes edge=0.8092279271675037, retail=0.711845510708334, arb=0.0973824164533331, edge SE=0.27276916007666724; failures=0.", "A queue4 {'controls.queue_notional': 4}: n=120, mean_edge=166.322098503955, retail_edge=192.45132849918082, arb_edge=-26.129229995144165; paired changes edge=0.8385361172875019, retail=0.8967082938491655, arb=-0.058172176566666914, edge SE=0.22697603569957392; failures=0.", "A queue_ask_W200 {'params.W': 200}: n=120, mean_edge=166.32560626344917, retail_edge=192.83367173864585, arb_edge=-26.508065475110833; paired changes edge=0.8420438767816688, retail=1.279051533314166, arb=-0.4370076565333336, edge SE=0.42444310689872167; failures=0.", "A queue2_W200 {'params.W': 200, 'controls.queue_notional': 2}: n=120, mean_edge=166.33380240487335, retail_edge=192.881569753325, arb_edge=-26.547767348367497; paired changes edge=0.8502400182058353, retail=1.326949547993334, arb=-0.47670952979000025, edge SE=0.4485877346615937; failures=0.", "A valueq1 {'controls.queue_factor': 1}: n=120, mean_edge=167.36448112351417, retail_edge=193.12715286855666, arb_edge=-25.762671744961665; paired changes edge=1.880918736846667, retail=1.572532663225, arb=0.30838607361583303, edge SE=0.4647344952265947; failures=0.", "B Hu1000: n=120, mean_edge=169.6635299203225, retail_edge=202.14855590011666, arb_edge=-32.48502597986083; paired changes edge=-0.12214603308833209, retail=0.05825631182833166, arb=-0.18040234491583348, edge SE=0.24215346870164972; failures=0.", "B W200: n=120, mean_edge=170.4751979909775, retail_edge=202.88357918015998, arb_edge=-32.40838118924833; paired changes edge=0.6895220375666662, retail=0.7932795918716671, arb=-0.1037575543033333, edge SE=0.38360467296813244; failures=0.", "B ahead10: n=120, mean_edge=170.47423499517498, retail_edge=202.82830436903416, arb_edge=-32.354069373916666; paired changes edge=0.6885590417641683, retail=0.7380047807458325, arb=-0.049445738971666736, edge SE=0.46819451004351853; failures=0.", "B ahead_notional: n=120, mean_edge=170.719837473115, retail_edge=203.03324980727334, arb_edge=-32.31341233422; paired changes edge=0.9341615197041678, retail=0.9429502189849971, arb=-0.008788699275000207, edge SE=0.47730387689926773; failures=0.", "B queue4: n=120, mean_edge=169.78756634857834, retail_edge=203.33238868473165, arb_edge=-33.544822336213336; paired changes edge=0.0018903951675001344, retail=1.2420890964433322, arb=-1.240198701268333, edge SE=1.8416681047108365; failures=0.", "B queue_ask: n=120, mean_edge=170.23816091446415, retail_edge=202.54068052274667, arb_edge=-32.302519608350835; paired changes edge=0.45248496105333363, retail=0.450380934458332, arb=0.0021040265941665785, edge SE=0.1458961227455199; failures=0.", "B queue_ask_W200: n=120, mean_edge=170.859868980715, retail_edge=203.27217305183748, arb_edge=-32.4123040711875; paired changes edge=1.0741930273041669, retail=1.1818734635491661, arb=-0.10768043624250015, edge SE=0.4006655096114375; failures=0.", "B queue_bid: n=120, mean_edge=170.33991032831085, retail_edge=202.63332376599166, arb_edge=-32.29341343774583; paired changes edge=0.5542343749000013, retail=0.5430241777033336, arb=0.01121019719916665, edge SE=0.47134812043326196; failures=0.", "B sigj025: n=120, mean_edge=168.19094493796416, retail_edge=202.58778231059918, arb_edge=-34.39683737272166; paired changes edge=-1.5947310154466654, retail=0.49748272231083396, arb=-2.092213737776666, edge SE=1.4245402111551388; failures=0.", "B valueq1: n=120, mean_edge=163.94224924100834, retail_edge=203.68898757056584, arb_edge=-39.74673832959; paired changes edge=-5.8434267124025006, retail=1.5986879822774982, arb=-7.4421146946449985, edge SE=4.4065818497427; failures=0.", "C value_mix: n=120, mean_edge=119.46596492778333, retail_edge=140.50041214332916, arb_edge=-21.034447215483333; paired changes edge=0.25808599820250006, retail=0.364426572344167, arb=-0.10634057416999977, edge SE=0.2413484217139791; failures=0.", "C value_safe: n=120, mean_edge=118.92668529319083, retail_edge=139.94783730338915, arb_edge=-21.02115201014; paired changes edge=-0.2811936363899994, retail=-0.18814826759583217, arb=-0.09304536882666653, edge SE=0.26948845693964857; failures=0.", "Full spread-4 tuning W200_4000_975: n=975, mean_edge=158.2883045600591, retail_edge=186.38812083290452, arb_edge=-28.099816272804617; paired changes edge=0.15794131070010245, retail=0.31167577331066665, arb=-0.1537344626105641, edge SE=0.09160670658239412; failures=0.", "W200_4000_975 half 4000: {'total_edge': 0.15572855141217026, 'retail_edge': 0.40844205440831605, 'arb_edge': -0.2527135029955375}, n=493, edge SE=0.1429076550896608.", "W200_4000_975 half 6000: {'total_edge': 0.1602045686439833, 'retail_edge': 0.21270113310083028, 'arb_edge': -0.05249656445746888}, n=482, edge SE=0.11404945889782464.", "Full spread-4 tuning ahead_notional_4000_975: n=975, mean_edge=158.47116586712073, retail_edge=186.63567192215345, arb_edge=-28.164506054990667; paired changes edge=0.3408026177617436, retail=0.5592268625595892, arb=-0.21842424479661535, edge SE=0.2201748823662512; failures=0.", "ahead_notional_4000_975 half 4000: {'total_edge': 0.22901002071298104, 'retail_edge': 0.6050412365521293, 'arb_edge': -0.37603121583975657}, n=493, edge SE=0.412019623293073.", "ahead_notional_4000_975 half 6000: {'total_edge': 0.4551464981456439, 'retail_edge': 0.5123669323141075, 'arb_edge': -0.057220434165352635}, n=482, edge SE=0.14459597972862415.", "Full spread-4 tuning queue2_W200_4000_975: n=975, mean_edge=158.5549118597431, retail_edge=186.8790933412288, arb_edge=-28.324181481443386; paired changes edge=0.4245486103841025, retail=0.802648281634974, arb=-0.3780996712493333, edge SE=0.23521208269118485; failures=0.", "queue2_W200_4000_975 half 4000: {'total_edge': 0.3096770928529409, 'retail_edge': 0.9594046467395535, 'arb_edge': -0.6497275538860041}, n=493, edge SE=0.43603885442530704.", "queue2_W200_4000_975 half 6000: {'total_edge': 0.5420416770705396, 'retail_edge': 0.6423144891109955, 'arb_edge': -0.10027281203796676}, n=482, edge SE=0.1662620051794778.", "Full spread-4 tuning queue_ask_full: n=975, mean_edge=158.4031986151071, retail_edge=186.5474125241438, arb_edge=-28.144213908996; paired changes edge=0.2728353657481024, retail=0.47096746454994814, arb=-0.19813209880194863, edge SE=0.2124117750885256; failures=0.", "queue_ask_full half 4000: {'total_edge': 0.21043581566064823, 'retail_edge': 0.5577144980772805, 'arb_edge': -0.3472786824182555}, n=493, edge SE=0.41098847515059944.", "queue_ask_full half 6000: {'total_edge': 0.33665897195788436, 'retail_edge': 0.38224072693796723, 'arb_edge': -0.04558175497863066}, n=482, edge SE=0.08987350911788615.", "Rejected the unrestricted one-step queue-value rule after B edge change -5.8434267124025: low-price asks became much larger before rare jumps. The guarded value_safe variant failed C. The mixed low-price queue variant passed short A/C screens but lacked a full advantage over the simpler rule. Stopped its optional full comparison and the remaining broad sweep; incomplete comparisons are not reported as scored results. No seed-specific branches or simulator modifications appear in the delivered strategy.", "Incremental W=200 with the queue change already enabled: n=975, paired changes {'total_edge': 0.08374599262235892, 'retail_edge': 0.24342141907538487, 'arb_edge': -0.15967542645271796}; net changes by tuning half {4000: 0.08066707213995988, 6000: 0.0868951789248957}. Both component additions gain on both halves.", "Incremental $2 queue threshold with W=200 already enabled: n=975, paired changes {'total_edge': 0.2666072996840001, 'retail_edge': 0.49097250832430744, 'arb_edge': -0.22436520863876921}; net changes by tuning half {4000: 0.15394854144077066, 6000: 0.3818371084265562}. Both component additions gain on both halves.", "Final isolation proof: all SimulationResult fields match the base exactly for all 456 non-4 runs among seeds 1000-1599; also exact on every non-4 holdout run. Zero failures in both final 4000-run ranges and both baseline checks.", "Best selected before holdout: queue2_W200; frozen SHA256 db82772cbc00f1ac695975bdd31d6e9b7a8e57f54b157fae9dd2c2266cd13479; whole tuning mean_edge=98.84752384692902, retail_edge=124.52530289979967, arb_edge=-25.6777790528859; unseen mean_edge=98.40769532192527, retail_edge=126.21554314516477, arb_edge=-27.807847823237275; paired unseen whole-run edge change=0.12706143489544985, SE=0.0384850215053414. Zero failures. Code and selection were not changed after seeing holdout. Parameterization matched 975 cached tuning runs exactly."], "next": "The unseen whole-run change is 0.12706143489544985 with paired SE 0.0384850215053414. Combine this spread-4 override only after the other items pass their independent spread-isolation and holdout checks. Do not infer a large constants-only opportunity from the short screens: capacity increases failed, and aggressive queue replenishment magnified jump losses. A wider effort should optimize inventory-aware queue exposure and stock replenishment under a calibrated jump mixture, with fresh untouched runs."}