{
  "approach": "Retains the base's sub-tick belief filter, one-tick-inside main quotes, marginal quote sizing, resting-order management and cheap YES inventory. Selected change: permit small ammunition only at tick 1 when estimated p<1c; activate the large tick-1 bid only for 1c<p<2c; suppress tick-1 asks unless at least 500 YES are held or H<=100. Separately measured fixed tick-1 versus just-under-probability stock bids, max(spread,tick) sizing, horizons and caps, ask activation, and cash allocation. Direct moving-bid grafts lost on the full tuning set. Chose the highest complete 600-seed tuning mean available at freeze, before the final tuning rerun and single untouched holdout evaluation; no generalization gain is claimed without a paired holdout comparison.",
  "blocked": null,
  "python_version": "3.12.3",
  "base_mean_edge": 102.33197908899267,
  "score": {
    "mean_edge": 109.29650163975117,
    "retail_edge": 134.38229969436017,
    "arb_edge": -25.085798054621666,
    "simulations": 600,
    "seed_start": 1000,
    "failures": 0
  },
  "holdout": {
    "mean_edge": 97.4026568625055,
    "simulations": 600,
    "seed_start": 30000,
    "failures": 0
  },
  "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=1.0, f1=0.6, a1=100.0, nb1=0.0, gs=1.0, p1t=1.0, m1=1.0, zd=1.0, pfl=0.0005)\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\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        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            return 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) >= 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 = qbid(bid_t)\n        qa = qask(ask_t)\n\n        Y = state.yes_inventory\n        if ask_t == 1 and H > P['H1']:\n            qa = qa if Y >= 500.0 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        if amt > 1 or phat >= 0.01:\n            amt, qam = 0, 0.0\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 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                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": "715597fc3b8ffa6aa640d9b0013bd94140849a91a1bcb480664b4384b4baecb1",
  "tried": [
    "Official simulator commit b244530ee2a6247b5f930830eeca9dd950da0e7e; Python 3.12.3. Base reproduction seeds 0-599 exactly matches published 102.33197908899267, failures=0. Base tuning seeds 1000-1599: mean=107.2442092237515, retail=135.80165652288784, arb=-28.557447299150834, failures=0.",
    "All search used only seeds 1000-1599. Initial screens use 1000-1199; extensions use 1200-1599 and are joined in seed order for n=600. Final combined candidates were also scored directly on all 600 tuning seeds. Means use the official BatchResult sum/number-of-successes arithmetic on raw simulator JSON. Selection uses only trials complete at freeze; later results are excluded. No failed unsandboxed candidate accepted. No seed identification or replay.",
    "bulk bid floor(ptick-0.1), below 6c; target 0.6*min(H,2000)/(spread/100)-YES, cap 20000 shares; base residual cash rule scaled by bid tick: screen n=200, mean=102.9510347742, retail=132.656305694371, arb=-29.705270920226997, failures=0. Full n=600: mean=94.5928809697745, retail=129.548463846311, arb=-34.955582876552, paired delta=-12.651328253977, paired SE=2.0302029470275027, failures=0.",
    "placement swap plus denominator max(spread,bid tick)/100: screen n=200, mean=107.4034996509465, retail=135.7285580022525, arb=-28.325058351367, failures=0. Full n=600: mean=99.47212891518934, retail=132.09413891519867, arb=-32.62201000002683, paired delta=-7.772080308562168, paired SE=1.7370265741410444, failures=0.",
    "moving bid plus max denominator, below 4c, target coefficient 0.5 and horizon 500, cap 20000; retain other base behavior: screen n=200, mean=109.3539583393005, retail=134.518587549729, arb=-25.164629210497, failures=0. Full n=600: mean=101.6894662687635, retail=130.84805106232415, arb=-29.158584793577997, paired delta=-5.554742954988, paired SE=1.569060695313167, failures=0.",
    "base fixed tick-1 bulk bid; base parameter changes: g1=0.25: screen n=200, mean=104.421363518393, retail=132.907346252986, arb=-28.4859827346585, failures=0.",
    "base fixed tick-1 bulk bid; base parameter changes: ydf1=1.0, f1=1.0: screen n=200, mean=109.7736140460655, retail=136.59390055327, arb=-26.820286507280997, failures=0. Full n=600: mean=104.50581270420116, retail=136.84054953911615, arb=-32.334736834931334, paired delta=-2.7383965195503333, paired SE=0.9692292690701723, failures=0.",
    "moving floor(ptick-0.1), target 0.6*min(H,2000)/(max(spread,bid tick)/100), cap 20000; activated below 4c: screen n=200, mean=111.2776211977485, retail=136.725068953533, arb=-25.447447755853, failures=0. Full n=600: mean=104.56474062921733, retail=134.33259067534865, arb=-29.767850046149338, paired delta=-2.679468594534167, paired SE=1.233565611877738, failures=0.",
    "crosspriority mechanism; base parameter changes: ydf1=0.45: screen n=200, mean=109.788021371208, retail=135.0180710118585, arb=-25.230049640720498, failures=0. Full n=600: mean=104.95470326899083, retail=133.48730104367385, arb=-28.53259777469517, paired delta=-2.289505954760667, paired SE=0.6074535773576057, failures=0.",
    "crosspriority mechanism; base parameter changes: ydf1=0.9: screen n=200, mean=109.98397785481801, retail=136.50207025288302, arb=-26.5180923981355, failures=0. Full n=600: mean=105.48339944152583, retail=136.92618650580883, arb=-31.442787064296834, paired delta=-1.7608097822256672, paired SE=0.7944118967663617, failures=0.",
    "value2 with moving coefficient 0.6 and horizon 2000: screen n=200, mean=111.47354528998851, retail=136.84730978364, arb=-25.3737644937225, failures=0. Full n=600: mean=105.59517257777, retail=136.12184793950883, arb=-30.52667536175767, paired delta=-1.6490366459815007, paired SE=0.6490870993752581, failures=0.",
    "sizing with joint bid budget: screen n=200, mean=105.640701202275, retail=135.146966245419, arb=-29.5062650432075, failures=0.",
    "base fixed tick-1 bulk bid; base parameter changes: ydf1=0.8, f1=0.8: screen n=200, mean=111.1438781640575, retail=136.497016676114, arb=-25.353138512127497, failures=0. Full n=600: mean=106.42791116679383, retail=136.4982174456688, arb=-30.070306278889337, paired delta=-0.8162980569576676, paired SE=0.5037723838872746, failures=0.",
    "base tick-1 bulk bid plus separate floor(ptick-0.1) trap at ticks >=2 below 4c; trap target 0.5*min(H,500)/(max(spread,tick)/100)-YES, share cap 4000: screen n=200, mean=111.42906755693849, retail=136.48892454072052, arb=-25.05985698385, failures=0. Full n=600: mean=106.51192644726484, retail=136.0708488506775, arb=-29.558922403431165, paired delta=-0.7322827764866666, paired SE=0.4367624710449738, failures=0.",
    "crosspriority mechanism; base parameter changes: ydf1=0.75: screen n=200, mean=111.0954931413935, retail=136.4577405540875, arb=-25.362247412764, failures=0. Full n=600: mean=106.66145823136034, retail=136.53968384940435, arb=-29.878225618058, paired delta=-0.5827509923911672, paired SE=0.41837562098814374, failures=0.",
    "base fixed tick-1 bulk bid; base parameter changes: ydf1=0.3, f1=0.3: screen n=200, mean=106.677759550595, retail=132.55497101869, arb=-25.877211468158997, failures=0.",
    "base fixed tick-1 bulk bid; base parameter changes: ydf1=0.3: screen n=200, mean=106.70337364038551, retail=132.5153081035755, arb=-25.811934463253, failures=0.",
    "fixed tick-1 bid with target horizon 1000; other bulk-bid constants unchanged: screen n=200, mean=111.43636755635951, retail=136.19868824296748, arb=-24.76232068668, failures=0. Full n=600: mean=106.75270835155, retail=135.508679699329, arb=-28.755971347794336, paired delta=-0.4915008722014998, paired SE=0.30499769928547166, failures=0.",
    "crosspriority mechanism; base parameter changes: ydf1=0.3: screen n=200, mean=106.82139093490551, retail=133.112772249983, arb=-26.2913813151435, failures=0.",
    "base tick-1 bulk bid, switching to floor(ptick-0.1) at ticks >=2 below 4c only when ask quote is at least 2 ticks higher; moving target 0.5*min(H,500)/(max(spread,tick)/100)-YES, cap 20000: screen n=200, mean=111.7801112036055, retail=136.4781198301645, arb=-24.698008626632, failures=0. Full n=600: mean=106.85144966308717, retail=136.0226147017675, arb=-29.1711650387, paired delta=-0.39275956066433343, paired SE=0.3874528716018568, failures=0.",
    "base fixed tick-1 bulk bid for competitor spread <3, latewindow moving-bid rule for larger spreads: screen n=200, mean=111.5982858815095, retail=136.337816051397, arb=-24.739530169961498, failures=0. Full n=600: mean=107.00566699631082, retail=135.8760202383175, arb=-28.870353242024333, paired delta=-0.23854222744066667, paired SE=0.20357962785284903, failures=0.",
    "value2 with minimum ask-minus-trap gap of 3 ticks: screen n=200, mean=111.987947833935, retail=136.422525126857, arb=-24.434577292995, failures=0. Full n=600: mean=107.08834600654232, retail=135.92134429117783, arb=-28.8329982846515, paired delta=-0.15586321720916654, paired SE=0.24775217738569605, failures=0.",
    "bulk bid max(1,floor(ptick-0.0)), activated 1c<p<6c; coefficient 0.6, horizon 2000, max(spread,tick) denominator, no effective share cap: screen n=200, mean=107.1226806981215, retail=135.90564544541, arb=-28.7829647473505, failures=0.",
    "base fixed tick-1 bulk bid; base parameter changes: g1=0.5: screen n=200, mean=107.165870399822, retail=135.7410606758525, arb=-28.575190276091497, failures=0.",
    "unmodified published base: screen n=200, mean=111.43190714093, retail=136.1966940843945, arb=-24.7647869435365, failures=0. Full n=600: mean=107.2442092237515, retail=135.80165652288784, arb=-28.557447299150834, paired delta=0.0, paired SE=0.0, failures=0.",
    "base with ydf=0, removing the full-demand small near-crossing bids but retaining small target top-ups: screen n=200, mean=111.6019088562565, retail=136.0455334283205, arb=-24.443624572132, failures=0. Full n=600: mean=107.2623428966065, retail=135.56245882781585, arb=-28.30011593122367, paired delta=0.018133672855000126, paired SE=0.2321319735641127, failures=0.",
    "fixed tick-1 bid activated only for estimated p between 1c and 4c: screen n=200, mean=111.63892525294101, retail=136.155599438628, arb=-24.516674185759, failures=0. Full n=600: mean=107.29440705590551, retail=135.79193222084965, arb=-28.497525164959, paired delta=0.05019783215399995, paired SE=0.08228093873017668, failures=0.",
    "crosspriority mechanism; base parameter changes: g1=2.0: screen n=200, mean=111.662793978037, retail=136.18117814358152, arb=-24.518384165618, failures=0. Full n=600: mean=107.39195997056967, retail=136.008264590084, arb=-28.616304619530165, paired delta=0.14775074681816672, paired SE=0.10255171562504901, failures=0.",
    "base with g_hi=-1, disabling the ceiling-priced full-demand near-crossing branch: screen n=200, mean=111.61275790445299, retail=136.203072027457, arb=-24.590314123078002, failures=0. Full n=600: mean=107.43841593229716, retail=135.71633688004252, arb=-28.277920947760165, paired delta=0.1942067085456668, paired SE=0.0699760834032619, failures=0.",
    "allow the base small under-p crossing bid across the whole tick rather than within 0.15 tick: screen n=200, mean=107.49724164906, retail=134.84814351102798, arb=-27.350901862037, failures=0.",
    "fixed tick-1 bid activated only for estimated p between 1c and 2c: screen n=200, mean=111.6020804474905, retail=136.098622434334, arb=-24.4965419869155, failures=0. Full n=600: mean=107.74234256393616, retail=135.71370321859365, arb=-27.97136065467183, paired delta=0.4981333401846667, paired SE=0.31669495516323815, failures=0.",
    "moving bid below 6c, max denominator, coefficient 0.6, target horizon 1000: screen n=200, mean=107.924912516133, retail=135.9572577862145, arb=-28.032345270146003, failures=0.",
    "bulk bid max(1,floor(ptick-0.5)), activated 1c<p<6c; coefficient 0.6, horizon 2000, max(spread,tick) denominator, no effective share cap: screen n=200, mean=108.151323847711, retail=136.16459557477398, arb=-28.013271727124003, failures=0.",
    "base fixed tick-1 bulk bid; base parameter changes: ydf1=1.2: screen n=200, mean=108.2214622283045, retail=136.58030924748448, arb=-28.358847019252998, failures=0.",
    "original published late-game file, comparison only: screen n=200, mean=108.4749699209115, retail=134.289929695644, arb=-25.8149597748055, failures=0.",
    "moving bid below 6c, max denominator, coefficient 0.6, target horizon 500: screen n=200, mean=108.655773226656, retail=135.7421858004935, arb=-27.0864125739055, failures=0.",
    "bulk2c purchase guard and tick-1 activation plus a floor(ptick-0.1) trap at ticks >=2 below 4c only when ask-minus-trap is at least 3 ticks; moving target 0.5*min(H,500)/(max(spread,tick)/100)-YES, cap 20000: direct full n=600, mean=108.73323160493283, retail=134.560483663398, arb=-25.827252058478, paired delta=1.4890223811813335, paired SE=0.5768422107234872, failures=0.",
    "fixed tick-1 bulk bid, below 4c, coefficient 0.5 and horizon 500; retain base behavior elsewhere: screen n=200, mean=108.8205604565815, retail=134.3941649039115, arb=-25.573604447398, failures=0.",
    "fixedwindow with joint bid budget: screen n=200, mean=108.841553175945, retail=134.3937846864275, arb=-25.55223151055, failures=0.",
    "permit small planned ammunition only at tick 1 when estimated p<1c; retain the large tick-1 bulk bid only for 1c<p<3c and keep it active instead of a small near-crossing bid: screen n=200, mean=113.91346588565649, retail=134.818330135304, arb=-20.904864249712, failures=0. Full n=600: mean=108.96501073231084, retail=134.34723848251, arb=-25.38222775021333, paired delta=1.7208015085593327, paired SE=0.47969366960134063, failures=0.",
    "permit small planned ammunition only at tick 1 when estimated p<1c; preserve the large tick-1 bulk bid for 1c<p<6c and keep it active instead of a small near-crossing bid: screen n=200, mean=113.8600513670475, retail=135.02709110600202, arb=-21.1670397390175, failures=0. Full n=600: mean=108.9661471336045, retail=134.48518184589932, arb=-25.519034712307665, paired delta=1.7219379098529994, paired SE=0.46695347746793864, failures=0.",
    "latewindow with stock bid jointly budgeted with main bid and small ammunition orders, rather than 60% of residual budget: screen n=200, mean=109.184232210593, retail=134.474575972847, arb=-25.2903437623205, failures=0.",
    "permit small planned ammunition only at tick 1 when estimated p<1c; retain the large tick-1 bulk bid only for 1c<p<2c and keep it active instead of a small near-crossing bid: screen n=200, mean=113.850410228896, retail=134.79045266386848, arb=-20.9400424350365, failures=0. Full n=600: mean=109.26965303125802, retail=134.0755729804585, arb=-24.805919949213166, paired delta=2.0254438075064995, paired SE=0.5941758548350695, failures=0.",
    "whole-tick small crossing bid, cap raised from 400 to 1500 shares: screen n=200, mean=109.27323677608051, retail=135.72951462384302, arb=-26.456277847826, failures=0.",
    "bulk bid max(1,floor(ptick-1.0)), activated 1c<p<6c; coefficient 0.6, horizon 2000, max(spread,tick) denominator, no effective share cap: screen n=200, mean=109.7668720669355, retail=136.107585349287, arb=-26.340713282421998, failures=0.",
    "base fixed tick-1 bulk bid; base parameter changes: ydf1=0.45: screen n=200, mean=109.78263072608301, retail=135.0111026090065, arb=-25.228471882994498, failures=0.",
    "base fixed tick-1 bulk bid; base parameter changes: ydf1=0.45, f1=0.45: screen n=200, mean=109.7964519168, retail=135.006019402724, arb=-25.2095674859975, failures=0.",
    "base fixed tick-1 bulk bid; base parameter changes: ydf1=0.9: screen n=200, mean=109.9839778548175, retail=136.50207025288302, arb=-26.518092398136, failures=0.",
    "fixed tick-1 bid with target horizon 500; other bulk-bid constants unchanged: screen n=200, mean=110.10877079389701, retail=135.5113128875525, arb=-25.402542093727998, failures=0.",
    "base fixed tick-1 bulk bid with share cap 10000: screen n=200, mean=110.1554584703305, retail=135.194853806851, arb=-25.039395336590502, failures=0.",
    "bulk bid max(1,floor(ptick-2.0)), activated 1c<p<6c; coefficient 0.6, horizon 2000, max(spread,tick) denominator, no effective share cap: screen n=200, mean=110.45249405537601, retail=136.15507580789, arb=-25.702581752585502, failures=0.",
    "base fixed tick-1 bulk bid for competitor spread <2, latewindow moving-bid rule for larger spreads: screen n=200, mean=110.575690424689, retail=136.24331041444302, arb=-25.6676199898225, failures=0.",
    "base fixed tick-1 bulk bid with share cap 20000: screen n=200, mean=110.8046268620925, retail=136.1968238163355, arb=-25.3921969543155, failures=0.",
    "base tick-1 bulk bid plus separate floor(ptick-0.1) trap at ticks >=2 below 4c; trap target 0.5*min(H,500)/(max(spread,tick)/100)-YES, share cap 400: screen n=200, mean=111.000383212534, retail=136.095816109005, arb=-25.0954328965455, failures=0.",
    "base tick-1 bulk bid plus separate floor(ptick-0.1) trap at ticks >=2 below 4c; trap target 0.5*min(H,500)/(max(spread,tick)/100)-YES, share cap 1500: screen n=200, mean=111.0030847350635, retail=136.405084249355, arb=-25.401999514365997, failures=0.",
    "base fixed tick-1 bulk bid; base parameter changes: ydf1=0.75: screen n=200, mean=111.008750228803, retail=136.3463779651465, arb=-25.337627736413502, failures=0.",
    "base fixed tick-1 bulk bid until final 100 steps, then max4c moving-bid rule: screen n=200, mean=111.250510725849, retail=136.2039390236645, arb=-24.9534282978875, failures=0.",
    "base fixed tick-1 bulk bid; base parameter changes: g1=7.0: screen n=200, mean=111.27562140213749, retail=136.19656824867, arb=-24.920946846604, failures=0.",
    "base fixed tick-1 bulk bid until final 300 steps, then max4c moving-bid rule: screen n=200, mean=111.28093215634, retail=136.231046585942, arb=-24.950114429675, failures=0.",
    "base fixed tick-1 bulk bid with share cap 40000: screen n=200, mean=111.430104723542, retail=136.1966940843945, arb=-24.7665893609255, failures=0.",
    "suppress tick-1 asks until 100 YES held except final 100 steps: screen n=200, mean=111.43190714093, retail=136.1966940843945, arb=-24.7647869435365, failures=0.",
    "suppress tick-1 asks until 1000 YES held except final 100 steps: screen n=200, mean=111.43190714093, retail=136.1966940843945, arb=-24.7647869435365, failures=0.",
    "suppress tick-1 asks until 250 YES held except final 100 steps: screen n=200, mean=111.43190714093, retail=136.1966940843945, arb=-24.7647869435365, failures=0.",
    "suppress tick-1 asks until 500 YES held, except final 100 steps: screen n=200, mean=111.43190714093, retail=136.1966940843945, arb=-24.7647869435365, failures=0.",
    "keep large tick-1 bid active when a small crossing bid would replace it; remove the duplicate small bid at tick 1: screen n=200, mean=111.455347966152, retail=136.2224752889735, arb=-24.767127322893998, failures=0.",
    "original fixed tick-1 base with joint bid budget instead of 60% residual allocation: screen n=200, mean=111.4729975988355, retail=136.21640844934052, arb=-24.743410850577003, failures=0.",
    "base fixed tick-1 bulk bid for competitor spread <4, latewindow moving-bid rule for larger spreads: screen n=200, mean=111.55254621190198, retail=136.27381727424898, arb=-24.7212710624165, failures=0.",
    "base fixed tick-1 bulk bid until final 600 steps, then max4c moving-bid rule: screen n=200, mean=111.615198683806, retail=136.333039999812, arb=-24.717841316081, failures=0.",
    "crosspriority mechanism; base parameter changes: g1=1.0: screen n=200, mean=111.6255212727125, retail=136.124403638913, arb=-24.498882366273, failures=0.",
    "base fixed tick-1 bulk bid; base parameter changes: g1=2.0: screen n=200, mean=111.639353152815, retail=136.1553969390025, arb=-24.516043786260504, failures=0.",
    "crosspriority mechanism; base parameter changes: g1=3.0: screen n=200, mean=111.66236607816299, retail=136.181380643207, arb=-24.5190145651165, failures=0.",
    "base fixed tick-1 bulk bid; base parameter changes: g1=9.0: screen n=200, mean=111.774609705877, retail=136.30539516447098, arb=-24.5307854586605, failures=0.",
    "starter --sandbox, seeds 1000-1019: 20/20 failed before strategy loading because Python 3.12 pathlib imports ntpath and the sandbox rejects it. This is a runner limitation shared by the starter; sandbox execution could not be verified. The sandbox was not patched and no sandbox success is claimed.",
    "frozen --sandbox, seeds 1000-1019: 20/20 failed before strategy loading because Python 3.12 pathlib imports ntpath and the sandbox rejects it. This is a runner limitation shared by the starter; sandbox execution could not be verified. The sandbox was not patched and no sandbox success is claimed.",
    "Frozen winner: permit small ammunition only at tick 1 when estimated p<1c; activate the large tick-1 bid only for 1c<p<2c; suppress tick-1 asks unless at least 500 YES are held or H<=100. Standalone official CLI rerun on all 600 tuning seeds: mean=109.29650163975117, retail=134.38229969436017, arb=-25.085798054621666, paired gain over base=2.0522924159996663, paired SE=0.5968385231758765, failures=0. One untouched holdout evaluation, seeds 30000-30599: mean=97.4026568625055, retail=123.39099317733115, arb=-25.988336314822835, failures=0. Code frozen before either final evaluation; no selection or changes from holdout results. No same-seed base holdout comparator was run, so no held-out gain over base is claimed. Nothing fabricated."
  ],
  "next": "Test an inventory target driven by observed retail demand instead of a fixed coefficient times remaining steps, while keeping cheap stock purchases at tick 1 unless new paired evidence supports moving them. Use a fresh holdout because 30000-30599 is now reported; the tuning gain here is not evidence of a generalization gain."
}