diff --git a/scripts/analysis/battery_sizing_screen.py b/scripts/analysis/battery_sizing_screen.py
new file mode 100644
index 0000000..23eb50d
--- /dev/null
+++ b/scripts/analysis/battery_sizing_screen.py
@@ -0,0 +1,368 @@
+#!/usr/bin/env python3
+"""
+Ekonomický screening velikosti baterie (15min, jednobusová energie).
+
+Typicky BA81: fixní nákup + prodej spot, limity výkonu z baterie min(0,5C, střídač),
+export/import podle připojení. Načte OTE z Postgres (stejná DB jako EMS) nebo z CSV.
+
+Příklad:
+  python3 scripts/analysis/battery_sizing_screen.py \\
+    --db \\
+    --date-from 2024-04-01 --date-to 2026-04-01 \\
+    --battery-kwh 12.5 32 48 \\
+    --load-kw 1.2 \\
+    --pv-daily-kwh-summer 55 --pv-daily-kwh-winter 12 \\
+    --sell-margin-fixed -0.02 \\
+    --buy-vat-kwh 4.443 \\
+    --capex-per-kwh 9000
+
+Vyžaduje: pip install pulp (volitelně psycopg2 pro --db).
+
+Omezení modelu: syntetický denní tvar FVE (kalibruj --pv-daily-kwh-* podle měření);
+mikroinvertory / GEN nejsou; zelený bonus není v účelové funkci; nákup je jedna flat
+sazba vč. DPH (reálné NT/VT přes HDO přidej později). Výsledek = screening, ne nabídka.
+"""
+from __future__ import annotations
+
+import argparse
+import csv
+import math
+import os
+import sys
+from dataclasses import dataclass
+from datetime import date, datetime, timedelta
+from typing import Iterable, Sequence
+
+try:
+    import pulp
+except ImportError:
+    print("Instaluj PuLP: pip install pulp", file=sys.stderr)
+    raise
+
+DT_H = 0.25  # 15 min
+SLOTS_PER_DAY = 96
+
+
+@dataclass
+class SiteLimits:
+    max_export_w: float = 16_000.0
+    max_import_w: float = 17_000.0
+    inv_batt_max_w: float = 12_000.0  # strop střídače z baterie / nabíjení
+    c_rate: float = 0.5  # P_batt = min(c_rate * E_kWh * 1000, inv_batt_max_w)
+    eta_charge: float = 0.95
+    eta_discharge: float = 0.95
+    soc_min_frac: float = 0.10
+    soc_max_frac: float = 0.95
+
+
+def batt_power_cap_w(usable_kwh: float, site: SiteLimits) -> float:
+    return min(site.c_rate * usable_kwh * 1000.0, site.inv_batt_max_w)
+
+
+def summer_day(d: date) -> bool:
+    m = d.month
+    return m >= 4 and m <= 9
+
+
+def pv_shape_96() -> list[float]:
+    """Nenormalizovaný denní tvar (96 slotů), plocha = 1 po normalizaci."""
+    w = [0.0] * SLOTS_PER_DAY
+    for t in range(SLOTS_PER_DAY):
+        h = t / 4.0  # hodiny od půlnoci
+        if 5.5 <= h <= 20.5:
+            w[t] = max(0.0, math.sin(math.pi * (h - 5.5) / 15.0)) ** 1.2
+        else:
+            w[t] = 0.0
+    s = sum(w)
+    if s <= 0:
+        return [1.0 / SLOTS_PER_DAY] * SLOTS_PER_DAY
+    return [x / s for x in w]
+
+
+def daily_pv_wh(d: date, summer_kwh: float, winter_kwh: float, shape: Sequence[float]) -> list[float]:
+    base = summer_kwh if summer_day(d) else winter_kwh
+    return [base * 1000.0 * sh for sh in shape]
+
+
+def daily_load_wh(load_kw: float) -> list[float]:
+    e_per_slot = load_kw * 1000.0 * DT_H
+    return [e_per_slot] * SLOTS_PER_DAY
+
+
+def effective_sell_kc_kwh(raw_ote: float, margin_fixed: float, margin_pct: float) -> float:
+    return raw_ote + margin_fixed + (raw_ote * margin_pct / 100.0)
+
+
+def load_prices_csv(path: str) -> list[tuple[datetime, float]]:
+    out: list[tuple[datetime, float]] = []
+    with open(path, newline="", encoding="utf-8") as f:
+        r = csv.DictReader(f)
+        for row in r:
+            ts = datetime.fromisoformat(row["interval_start"].replace("Z", "+00:00"))
+            px = float(row["sell_raw_price_czk_kwh"])
+            out.append((ts, px))
+    out.sort(key=lambda x: x[0])
+    return out
+
+
+def load_prices_db(date_from: date, date_to: date) -> list[tuple[datetime, float]]:
+    from datetime import timezone
+
+    try:
+        import psycopg2
+    except ImportError as e:
+        raise SystemExit("Pro --db instaluj psycopg2-binary nebo použij --price-csv") from e
+    from zoneinfo import ZoneInfo
+
+    prg = ZoneInfo("Europe/Prague")
+    t0 = datetime.combine(date_from, datetime.min.time(), tzinfo=prg).astimezone(timezone.utc)
+    t1 = datetime.combine(date_to, datetime.min.time(), tzinfo=prg).astimezone(timezone.utc)
+
+    conn = psycopg2.connect(
+        host=os.environ.get("PGHOST", "127.0.0.1"),
+        port=int(os.environ.get("PGPORT", "5432")),
+        dbname=os.environ.get("PGDATABASE", "ems"),
+        user=os.environ.get("PGUSER", os.environ.get("DB_USER", "ems_user")),
+        password=os.environ.get("PGPASSWORD", os.environ.get("DB_PASSWORD", "")),
+    )
+    cur = conn.cursor()
+    cur.execute(
+        """
+        SELECT interval_start, sell_raw_price_czk_kwh::float
+        FROM ems.market_interval_price
+        WHERE market_source = 'OTE_CZ'
+          AND interval_start >= %s
+          AND interval_start < %s
+        ORDER BY interval_start
+        """,
+        (t0, t1),
+    )
+    rows = cur.fetchall()
+    conn.close()
+    return [(r[0], float(r[1])) for r in rows]
+
+
+def prices_by_calendar_day(
+    series: list[tuple[datetime, float]],
+) -> dict[date, list[float]]:
+    """96 hodnot Kč/kWh (raw OTE) na kalendářní den Europe/Prague."""
+    from zoneinfo import ZoneInfo
+
+    prg = ZoneInfo("Europe/Prague")
+    buckets: dict[date, dict[int, float]] = {}
+    for ts, px in series:
+        local = ts.astimezone(prg)
+        d = local.date()
+        slot = local.hour * 4 + local.minute // 15
+        buckets.setdefault(d, {})[slot] = px
+    out: dict[date, list[float]] = {}
+    for d, mp in buckets.items():
+        if len(mp) < SLOTS_PER_DAY:
+            continue
+        out[d] = [mp[i] for i in range(SLOTS_PER_DAY)]
+    return out
+
+
+def solve_one_day(
+    pv_wh: Sequence[float],
+    load_wh: Sequence[float],
+    p_sell: Sequence[float],
+    p_buy_flat: float,
+    e_usable_wh: float,
+    p_batt_w: float,
+    site: SiteLimits,
+    soc_start_wh: float,
+) -> tuple[float, float, float, float]:
+    """
+    Vrátí (cash_kc, soc_end_wh, curtailed_wh, discharged_wh_sum).
+    cash = příjem z exportu − nákup z DS (jen energie, Kč).
+    """
+    e_min = site.soc_min_frac * e_usable_wh
+    e_max = site.soc_max_frac * e_usable_wh
+    max_ch = p_batt_w * DT_H
+    max_dis = p_batt_w * DT_H
+    max_exp = site.max_export_w * DT_H
+    max_imp = site.max_import_w * DT_H
+
+    prob = pulp.LpProblem("ems_day", pulp.LpMaximize)
+    soc = pulp.LpVariable.dicts("soc", range(SLOTS_PER_DAY + 1), lowBound=e_min, upBound=e_max)
+    ch = pulp.LpVariable.dicts("ch", range(SLOTS_PER_DAY), lowBound=0)
+    dis = pulp.LpVariable.dicts("dis", range(SLOTS_PER_DAY), lowBound=0)
+    gexp = pulp.LpVariable.dicts("gexp", range(SLOTS_PER_DAY), lowBound=0)
+    gimp = pulp.LpVariable.dicts("gimp", range(SLOTS_PER_DAY), lowBound=0)
+    curt = pulp.LpVariable.dicts("curt", range(SLOTS_PER_DAY), lowBound=0)
+
+    prob += soc[0] == soc_start_wh
+
+    obj = []
+    for t in range(SLOTS_PER_DAY):
+        prob += ch[t] <= max_ch
+        prob += dis[t] <= max_dis
+        prob += gexp[t] <= max_exp
+        prob += gimp[t] <= max_imp
+        prob += curt[t] <= pv_wh[t]
+        prob += (
+            pv_wh[t] - curt[t] + dis[t] + gimp[t] == load_wh[t] + ch[t] + gexp[t]
+        ), f"balance_{t}"
+        prob += (
+            soc[t + 1]
+            == soc[t] + site.eta_charge * ch[t] - dis[t] / site.eta_discharge
+        ), f"socdyn_{t}"
+        obj.append(p_sell[t] * gexp[t] / 1000.0 - p_buy_flat * gimp[t] / 1000.0)
+
+    prob += pulp.lpSum(obj)
+
+    solver = pulp.PULP_CBC_CMD(msg=False, timeLimit=60)
+    prob.solve(solver)
+    if prob.status != pulp.LpStatusOptimal:
+        raise RuntimeError(f"LP status {pulp.LpStatus[prob.status]}")
+
+    cash = float(pulp.value(prob.objective))
+    soc_end = float(pulp.value(soc[SLOTS_PER_DAY]))
+    curt_total = sum(float(pulp.value(curt[t])) for t in range(SLOTS_PER_DAY))
+    dis_total = sum(float(pulp.value(dis[t])) for t in range(SLOTS_PER_DAY))
+    return cash, soc_end, curt_total, dis_total
+
+
+def simulate_year(
+    days: Iterable[date],
+    px_day: dict[date, list[float]],
+    usable_kwh: float,
+    site: SiteLimits,
+    sell_margin_fixed: float,
+    sell_margin_pct: float,
+    buy_vat_kwh: float,
+    summer_kwh: float,
+    winter_kwh: float,
+    load_kw: float,
+    shape: Sequence[float],
+) -> dict[str, float]:
+    e_wh = usable_kwh * 1000.0
+    p_batt = batt_power_cap_w(usable_kwh, site)
+    load_wh = daily_load_wh(load_kw)
+    cash_total = 0.0
+    curt_total = 0.0
+    dis_total = 0.0
+    soc_state = 0.5 * (site.soc_min_frac + site.soc_max_frac) * e_wh
+    n_days = 0
+    for d in days:
+        if d not in px_day:
+            continue
+        raw = px_day[d]
+        p_sell = [effective_sell_kc_kwh(x, sell_margin_fixed, sell_margin_pct) for x in raw]
+        pv_wh = daily_pv_wh(d, summer_kwh, winter_kwh, shape)
+        cash, soc_state, curt, dis = solve_one_day(
+            pv_wh, load_wh, p_sell, buy_vat_kwh, e_wh, p_batt, site, soc_state
+        )
+        cash_total += cash
+        curt_total += curt
+        dis_total += dis
+        n_days += 1
+    feq = (dis_total / e_wh / n_days) if n_days and e_wh > 0 else 0.0
+    return {
+        "cash_kc": cash_total,
+        "days": float(n_days),
+        "curt_wh": curt_total,
+        "dis_wh": dis_total,
+        "feq_cycles_per_day": feq,
+    }
+
+
+def main() -> None:
+    ap = argparse.ArgumentParser(description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter)
+    ap.add_argument("--db", action="store_true", help="Načti OTE z Postgres (env PG* / DB_*)")
+    ap.add_argument("--price-csv", type=str, default="", help="CSV: interval_start, sell_raw_price_czk_kwh")
+    ap.add_argument("--date-from", type=str, required=True)
+    ap.add_argument("--date-to", type=str, required=True)
+    ap.add_argument("--battery-kwh", type=float, nargs="+", required=True, help="Užitkové kWh (např. 12.5 32 48)")
+    ap.add_argument("--load-kw", type=float, default=1.0, help="Průměrný odběr (konstanta přes den)")
+    ap.add_argument("--pv-daily-kwh-summer", type=float, default=50.0)
+    ap.add_argument("--pv-daily-kwh-winter", type=float, default=10.0)
+    ap.add_argument("--sell-margin-fixed", type=float, default=-0.02)
+    ap.add_argument("--sell-margin-pct", type=float, default=0.0)
+    ap.add_argument("--buy-vat-kwh", type=float, default=4.443, help="Efektivní nákup Kč/kWh vč. DPH (flat screening)")
+    ap.add_argument("--max-export-w", type=float, default=16_000.0)
+    ap.add_argument("--max-import-w", type=float, default=17_000.0)
+    ap.add_argument("--inv-batt-max-w", type=float, default=12_000.0)
+    ap.add_argument("--c-rate", type=float, default=0.5)
+    ap.add_argument("--capex-per-kwh", type=float, default=0.0, help="CAPEX za 1 kWh rozšíření; vypíše jednoduchou návratnost vs. nejmenší baterie")
+    args = ap.parse_args()
+
+    d0 = date.fromisoformat(args.date_from)
+    d1 = date.fromisoformat(args.date_to)
+    if args.db:
+        series = load_prices_db(d0, d1)
+    elif args.price_csv:
+        series = load_prices_csv(args.price_csv)
+    else:
+        ap.error("Zadej --db nebo --price-csv")
+
+    px_day = prices_by_calendar_day(series)
+    shape = pv_shape_96()
+    site = SiteLimits(
+        max_export_w=args.max_export_w,
+        max_import_w=args.max_import_w,
+        inv_batt_max_w=args.inv_batt_max_w,
+        c_rate=args.c_rate,
+    )
+
+    day_list = [d0 + timedelta(days=i) for i in range((d1 - d0).days)]
+
+    results = []
+    for kwh in sorted(args.battery_kwh):
+        r = simulate_year(
+            day_list,
+            px_day,
+            kwh,
+            site,
+            args.sell_margin_fixed,
+            args.sell_margin_pct,
+            args.buy_vat_kwh,
+            args.pv_daily_kwh_summer,
+            args.pv_daily_kwh_winter,
+            args.load_kw,
+            shape,
+        )
+        results.append((kwh, r))
+
+    baseline_kwh = min(args.battery_kwh)
+    base = dict(results)[baseline_kwh]
+
+    print("Parametry: prodej = OTE + sell_margin_fixed (+ %), nákup = flat buy_vat_kwh")
+    print(f"  FVE tvar = syntetický den, léto {args.pv_daily_kwh_summer} kWh/d, zima {args.pv_daily_kwh_winter} kWh/d, load {args.load_kw} kW")
+    print(f"  Limity: export {args.max_export_w} W, import {args.max_import_w} W, P_batt = min({args.c_rate}*E_kWh, {args.inv_batt_max_w} W)")
+    print()
+
+    print(f"{'kWh':>8} {'P_batt_kW':>10} {'cash_kc/rok':>14} {'Δ vs min':>12} {'curt_MWh/y':>12} {'Feq/den':>8}")
+    for kwh, r in results:
+        pkw = batt_power_cap_w(kwh, site) / 1000.0
+        days = max(int(r["days"]), 1)
+        cash_y = r["cash_kc"] * (365.0 / days)
+        curt_mwh = r["curt_wh"] / 1e6 * (365.0 / days)
+        delta = cash_y - base["cash_kc"] * (365.0 / days) if kwh != baseline_kwh else 0.0
+        print(
+            f"{kwh:8.1f} {pkw:10.2f} {cash_y:14.0f} {delta:12.0f} {curt_mwh:12.2f} {r['feq_cycles_per_day']:8.2f}"
+        )
+
+    if args.capex_per_kwh > 0:
+        print()
+        base_cash = base["cash_kc"] * (365.0 / max(int(base["days"]), 1))
+        for kwh, r in results:
+            if kwh <= baseline_kwh:
+                continue
+            cash_y = r["cash_kc"] * (365.0 / max(int(r["days"]), 1))
+            delta = cash_y - base_cash
+            extra_kwh = kwh - baseline_kwh
+            capex = extra_kwh * args.capex_per_kwh
+            if delta > 0:
+                years = capex / delta
+                print(
+                    f"vs {baseline_kwh} kWh → +{extra_kwh:.0f} kWh CAPEX ~{capex:,.0f} Kč, "
+                    f"odhad +{delta:,.0f} Kč/rok → návratnost ~{years:.1f} r"
+                )
+            else:
+                print(f"vs {baseline_kwh} kWh → +{extra_kwh:.0f} kWh: model neukazuje vyšší roční cash ({delta:,.0f} Kč/rok)")
+
+
+if __name__ == "__main__":
+    main()
diff --git a/scripts/analysis/data/ote_2026-04-09_11_12.csv b/scripts/analysis/data/ote_2026-04-09_11_12.csv
new file mode 100644
index 0000000..f2cf5ef
--- /dev/null
+++ b/scripts/analysis/data/ote_2026-04-09_11_12.csv
@@ -0,0 +1,288 @@
+2026-04-09,00:00,2.6720
+2026-04-09,00:15,2.5505
+2026-04-09,00:30,2.4905
+2026-04-09,00:45,2.4618
+2026-04-09,01:00,2.4598
+2026-04-09,01:15,2.4830
+2026-04-09,01:30,2.4660
+2026-04-09,01:45,2.4603
+2026-04-09,02:00,2.5270
+2026-04-09,02:15,2.4995
+2026-04-09,02:30,2.5213
+2026-04-09,02:45,2.5305
+2026-04-09,03:00,2.5265
+2026-04-09,03:15,2.5483
+2026-04-09,03:30,2.5655
+2026-04-09,03:45,2.6203
+2026-04-09,04:00,2.5300
+2026-04-09,04:15,2.6015
+2026-04-09,04:30,2.6713
+2026-04-09,04:45,2.8440
+2026-04-09,05:00,2.6540
+2026-04-09,05:15,2.7750
+2026-04-09,05:30,3.0300
+2026-04-09,05:45,3.2593
+2026-04-09,06:00,2.9765
+2026-04-09,06:15,3.4128
+2026-04-09,06:30,3.7728
+2026-04-09,06:45,3.8188
+2026-04-09,07:00,3.9350
+2026-04-09,07:15,3.8950
+2026-04-09,07:30,3.6658
+2026-04-09,07:45,3.2233
+2026-04-09,08:00,3.9688
+2026-04-09,08:15,3.3950
+2026-04-09,08:30,2.9795
+2026-04-09,08:45,2.4095
+2026-04-09,09:00,3.1465
+2026-04-09,09:15,2.6005
+2026-04-09,09:30,2.1475
+2026-04-09,09:45,1.6428
+2026-04-09,10:00,2.0153
+2026-04-09,10:15,1.6153
+2026-04-09,10:30,0.9560
+2026-04-09,10:45,0.2468
+2026-04-09,11:00,0.1235
+2026-04-09,11:15,0.0003
+2026-04-09,11:30,0.0000
+2026-04-09,11:45,-0.0003
+2026-04-09,12:00,-0.0003
+2026-04-09,12:15,-0.0025
+2026-04-09,12:30,-0.0313
+2026-04-09,12:45,-0.0760
+2026-04-09,13:00,-0.0808
+2026-04-09,13:15,-0.1385
+2026-04-09,13:30,-0.2323
+2026-04-09,13:45,-0.3113
+2026-04-09,14:00,-0.2458
+2026-04-09,14:15,-0.2328
+2026-04-09,14:30,-0.1720
+2026-04-09,14:45,-0.1265
+2026-04-09,15:00,-0.0748
+2026-04-09,15:15,-0.0053
+2026-04-09,15:30,0.0035
+2026-04-09,15:45,0.0003
+2026-04-09,16:00,0.0003
+2026-04-09,16:15,0.1078
+2026-04-09,16:30,0.9075
+2026-04-09,16:45,1.7903
+2026-04-09,17:00,0.7188
+2026-04-09,17:15,1.9618
+2026-04-09,17:30,2.6005
+2026-04-09,17:45,3.1290
+2026-04-09,18:00,2.4015
+2026-04-09,18:15,2.8648
+2026-04-09,18:30,3.1983
+2026-04-09,18:45,3.4253
+2026-04-09,19:00,3.4435
+2026-04-09,19:15,3.6148
+2026-04-09,19:30,3.8523
+2026-04-09,19:45,3.8893
+2026-04-09,20:00,4.1890
+2026-04-09,20:15,3.8510
+2026-04-09,20:30,3.5613
+2026-04-09,20:45,3.1660
+2026-04-09,21:00,3.3128
+2026-04-09,21:15,3.0373
+2026-04-09,21:30,2.6313
+2026-04-09,21:45,2.2740
+2026-04-09,22:00,2.9655
+2026-04-09,22:15,2.7910
+2026-04-09,22:30,2.5323
+2026-04-09,22:45,2.2793
+2026-04-09,23:00,2.4410
+2026-04-09,23:15,2.2813
+2026-04-09,23:30,2.1535
+2026-04-09,23:45,1.9268
+2026-04-11,00:00,3.5698
+2026-04-11,00:15,3.2533
+2026-04-11,00:30,2.9323
+2026-04-11,00:45,2.7985
+2026-04-11,01:00,3.2713
+2026-04-11,01:15,3.1870
+2026-04-11,01:30,2.9530
+2026-04-11,01:45,2.9568
+2026-04-11,02:00,2.9888
+2026-04-11,02:15,2.9133
+2026-04-11,02:30,2.9605
+2026-04-11,02:45,2.9473
+2026-04-11,03:00,3.0010
+2026-04-11,03:15,2.9330
+2026-04-11,03:30,2.9580
+2026-04-11,03:45,2.8618
+2026-04-11,04:00,2.9760
+2026-04-11,04:15,2.9570
+2026-04-11,04:30,2.9235
+2026-04-11,04:45,2.8015
+2026-04-11,05:00,2.9408
+2026-04-11,05:15,2.8243
+2026-04-11,05:30,2.8233
+2026-04-11,05:45,2.7850
+2026-04-11,06:00,2.8100
+2026-04-11,06:15,2.8503
+2026-04-11,06:30,2.8383
+2026-04-11,06:45,2.6828
+2026-04-11,07:00,2.9005
+2026-04-11,07:15,2.8145
+2026-04-11,07:30,2.4048
+2026-04-11,07:45,1.7628
+2026-04-11,08:00,2.7298
+2026-04-11,08:15,2.3223
+2026-04-11,08:30,1.6730
+2026-04-11,08:45,0.5990
+2026-04-11,09:00,1.4445
+2026-04-11,09:15,0.5615
+2026-04-11,09:30,0.2000
+2026-04-11,09:45,0.0003
+2026-04-11,10:00,0.1933
+2026-04-11,10:15,0.0000
+2026-04-11,10:30,-0.0305
+2026-04-11,10:45,-0.2223
+2026-04-11,11:00,-0.0308
+2026-04-11,11:15,0.0000
+2026-04-11,11:30,0.0250
+2026-04-11,11:45,0.1450
+2026-04-11,12:00,-0.4685
+2026-04-11,12:15,-1.0418
+2026-04-11,12:30,-0.9168
+2026-04-11,12:45,-1.2465
+2026-04-11,13:00,-1.5560
+2026-04-11,13:15,-1.7465
+2026-04-11,13:30,-2.1328
+2026-04-11,13:45,-2.4545
+2026-04-11,14:00,-1.8465
+2026-04-11,14:15,-1.8458
+2026-04-11,14:30,-1.7578
+2026-04-11,14:45,-1.4893
+2026-04-11,15:00,-1.7045
+2026-04-11,15:15,-1.3535
+2026-04-11,15:30,-1.0890
+2026-04-11,15:45,-0.6923
+2026-04-11,16:00,-0.7728
+2026-04-11,16:15,-0.3718
+2026-04-11,16:30,-0.2075
+2026-04-11,16:45,-0.3675
+2026-04-11,17:00,0.3510
+2026-04-11,17:15,-0.1903
+2026-04-11,17:30,-0.0250
+2026-04-11,17:45,0.1198
+2026-04-11,18:00,-0.4573
+2026-04-11,18:15,1.8173
+2026-04-11,18:30,2.3750
+2026-04-11,18:45,2.9240
+2026-04-11,19:00,2.3338
+2026-04-11,19:15,2.6563
+2026-04-11,19:30,2.8090
+2026-04-11,19:45,3.0140
+2026-04-11,20:00,2.8350
+2026-04-11,20:15,2.4450
+2026-04-11,20:30,2.4565
+2026-04-11,20:45,2.2635
+2026-04-11,21:00,3.0835
+2026-04-11,21:15,2.1543
+2026-04-11,21:30,1.9325
+2026-04-11,21:45,1.8295
+2026-04-11,22:00,3.3630
+2026-04-11,22:15,2.5945
+2026-04-11,22:30,1.7628
+2026-04-11,22:45,1.4008
+2026-04-11,23:00,2.4143
+2026-04-11,23:15,1.7565
+2026-04-11,23:30,1.5728
+2026-04-11,23:45,1.5035
+2026-04-12,00:00,1.5390
+2026-04-12,00:15,1.5708
+2026-04-12,00:30,1.7003
+2026-04-12,00:45,1.7023
+2026-04-12,01:00,1.7770
+2026-04-12,01:15,1.7503
+2026-04-12,01:30,1.8833
+2026-04-12,01:45,2.0250
+2026-04-12,02:00,1.8593
+2026-04-12,02:15,1.9398
+2026-04-12,02:30,1.9483
+2026-04-12,02:45,2.0395
+2026-04-12,03:00,2.0418
+2026-04-12,03:15,2.1465
+2026-04-12,03:30,2.1558
+2026-04-12,03:45,2.2473
+2026-04-12,04:00,2.3520
+2026-04-12,04:15,2.4003
+2026-04-12,04:30,2.4923
+2026-04-12,04:45,2.5645
+2026-04-12,05:00,2.4895
+2026-04-12,05:15,2.5743
+2026-04-12,05:30,2.5745
+2026-04-12,05:45,2.5400
+2026-04-12,06:00,2.6265
+2026-04-12,06:15,2.6030
+2026-04-12,06:30,2.5383
+2026-04-12,06:45,2.4885
+2026-04-12,07:00,2.7490
+2026-04-12,07:15,2.5110
+2026-04-12,07:30,2.3325
+2026-04-12,07:45,2.2773
+2026-04-12,08:00,2.1050
+2026-04-12,08:15,1.8573
+2026-04-12,08:30,1.7498
+2026-04-12,08:45,1.3183
+2026-04-12,09:00,1.4575
+2026-04-12,09:15,1.3108
+2026-04-12,09:30,0.8415
+2026-04-12,09:45,0.6773
+2026-04-12,10:00,0.8375
+2026-04-12,10:15,0.4075
+2026-04-12,10:30,0.3775
+2026-04-12,10:45,0.3193
+2026-04-12,11:00,0.4165
+2026-04-12,11:15,0.3638
+2026-04-12,11:30,0.3085
+2026-04-12,11:45,0.2470
+2026-04-12,12:00,0.2023
+2026-04-12,12:15,0.1435
+2026-04-12,12:30,0.1298
+2026-04-12,12:45,0.0615
+2026-04-12,13:00,0.0473
+2026-04-12,13:15,0.0173
+2026-04-12,13:30,0.0015
+2026-04-12,13:45,0.0013
+2026-04-12,14:00,0.0000
+2026-04-12,14:15,0.0278
+2026-04-12,14:30,0.0688
+2026-04-12,14:45,0.0628
+2026-04-12,15:00,0.1483
+2026-04-12,15:15,0.2505
+2026-04-12,15:30,0.4290
+2026-04-12,15:45,0.7538
+2026-04-12,16:00,0.8148
+2026-04-12,16:15,1.4820
+2026-04-12,16:30,1.5620
+2026-04-12,16:45,2.4738
+2026-04-12,17:00,1.5428
+2026-04-12,17:15,2.2933
+2026-04-12,17:30,3.1040
+2026-04-12,17:45,3.4375
+2026-04-12,18:00,2.8968
+2026-04-12,18:15,3.1310
+2026-04-12,18:30,3.4105
+2026-04-12,18:45,3.6880
+2026-04-12,19:00,3.3903
+2026-04-12,19:15,3.3798
+2026-04-12,19:30,3.6125
+2026-04-12,19:45,3.9263
+2026-04-12,20:00,3.7030
+2026-04-12,20:15,3.6903
+2026-04-12,20:30,3.6473
+2026-04-12,20:45,3.5030
+2026-04-12,21:00,3.5230
+2026-04-12,21:15,3.3825
+2026-04-12,21:30,3.4325
+2026-04-12,21:45,3.3353
+2026-04-12,22:00,3.4898
+2026-04-12,22:15,3.3675
+2026-04-12,22:30,3.2180
+2026-04-12,22:45,2.9855
+2026-04-12,23:00,3.3763
+2026-04-12,23:15,2.9848
+2026-04-12,23:30,2.9855
+2026-04-12,23:45,2.7558
diff --git a/scripts/analysis/join_inverter_ote_snapshot.py b/scripts/analysis/join_inverter_ote_snapshot.py
new file mode 100644
index 0000000..910f303
--- /dev/null
+++ b/scripts/analysis/join_inverter_ote_snapshot.py
@@ -0,0 +1,129 @@
+#!/usr/bin/env python3
+"""Join Deye inverter export (wide xlsx) with OTE 15min sell prices for BA81-style analysis.
+
+OTE CSV: regenerate from EMS DB (MCP or psql), example:
+
+  SELECT string_agg(
+    to_char((interval_start AT TIME ZONE 'Europe/Prague')::date, 'YYYY-MM-DD') || ',' ||
+    to_char(interval_start AT TIME ZONE 'Europe/Prague', 'HH24:MI') || ',' ||
+    trim(to_char(sell_raw_price_czk_kwh, 'FM9999990.0000')),
+    chr(10) ORDER BY interval_start
+  )
+  FROM ems.market_interval_price
+  WHERE market_source = 'OTE_CZ'
+    AND (interval_start AT TIME ZONE 'Europe/Prague')::date IN (...);
+
+Convention in sample logs: negative Battery Power(W) ≈ charging, positive ≈ discharging.
+Total Grid Power(W): small positive ≈ little/no export (sign per site firmware).
+
+Requires: openpyxl. Use read_only=False (these exports report max_row=1 in read_only mode).
+"""
+from __future__ import annotations
+
+import argparse
+import statistics as st
+from collections import defaultdict
+from datetime import datetime
+from pathlib import Path
+
+import openpyxl
+
+COLS = [
+    "Time",
+    "Total Solar Power(W)",
+    "Total Inverter Output Power(W)",
+    "Total Grid Power(W)",
+    "Battery Power(W)",
+    "SoC(%)",
+]
+
+
+def load_ote_csv(path: Path) -> dict[tuple[str, str], float]:
+    ote: dict[tuple[str, str], float] = {}
+    for line in path.read_text().splitlines():
+        line = line.strip()
+        if not line:
+            continue
+        d, hm, s = line.split(",")
+        ote[(d, hm)] = float(s)
+    return ote
+
+
+def floor_15(dt: datetime) -> datetime:
+    m = (dt.minute // 15) * 15
+    return dt.replace(minute=m, second=0, microsecond=0)
+
+
+def slot_key(dt: datetime) -> tuple[str, str]:
+    f = floor_15(dt)
+    return f.strftime("%Y-%m-%d"), f.strftime("%H:%M")
+
+
+def load_inverter_rows(fp: Path) -> list[dict[str, object]]:
+    wb = openpyxl.load_workbook(fp, read_only=False, data_only=True)
+    ws = wb.active
+    it = ws.iter_rows(values_only=True)
+    header = next(it)
+    idx = {str(h).strip(): i for i, h in enumerate(header) if h}
+    rows: list[dict[str, object]] = []
+    for r in it:
+        if not r or r[idx["Time"]] is None:
+            continue
+        rows.append({c: r[idx[c]] for c in COLS})
+    wb.close()
+    return rows
+
+
+def main() -> None:
+    p = argparse.ArgumentParser(description=__doc__)
+    p.add_argument("--ote-csv", type=Path, required=True)
+    p.add_argument("xlsx", type=Path, nargs="+")
+    args = p.parse_args()
+    ote = load_ote_csv(args.ote_csv)
+
+    for fp in args.xlsx:
+        data = load_inverter_rows(fp)
+        neg: list[tuple[float, datetime, dict]] = []
+        for r in data:
+            t = r["Time"]
+            if isinstance(t, str):
+                t = datetime.strptime(t, "%Y/%m/%d %H:%M:%S")
+            dk, hm = slot_key(t)
+            sell = ote.get((dk, hm))
+            if sell is None or sell >= 0:
+                continue
+            neg.append((sell, t, r))
+
+        print(f"\n=== {fp.name}  rows={len(data)}  OTE sell<0 samples={len(neg)}")
+        if not neg:
+            continue
+        socs = [float(x[2]["SoC(%)"]) for x in neg]
+        grids = [float(x[2]["Total Grid Power(W)"]) for x in neg]
+        bats = [float(x[2]["Battery Power(W)"]) for x in neg]
+        sols = [float(x[2]["Total Solar Power(W)"]) for x in neg]
+        print(f"  SoC %:   mean={st.mean(socs):.1f}  min={min(socs):.0f}  max={max(socs):.0f}")
+        print(f"  Grid W:  mean={st.mean(grids):.0f}  med={st.median(grids):.0f}")
+        print(f"  Bat W:   mean={st.mean(bats):.0f}  med={st.median(bats):.0f}")
+        print(f"  Solar W: mean={st.mean(sols):.0f}  med={st.median(sols):.0f}")
+
+        buckets: dict[str, list] = defaultdict(list)
+        for sell, t, r in neg:
+            if t.hour < 9 or t.hour > 18:
+                continue
+            _, hm = slot_key(t)
+            buckets[hm].append((sell, r))
+        print("  15min buckets (OTE<0, 09-18h) medians:")
+        for hm in sorted(buckets.keys()):
+            b = buckets[hm]
+            sell = b[0][0]
+            socs_b = [float(x[1]["SoC(%)"]) for x in b]
+            print(
+                f"    {hm}  sell={sell:+.3f}  n={len(b):2d}  "
+                f"SoC_med={st.median(socs_b):.0f}%  "
+                f"Pgrid_med={st.median([float(x[1]['Total Grid Power(W)']) for x in b]):.0f}W  "
+                f"Psol_med={st.median([float(x[1]['Total Solar Power(W)']) for x in b]):.0f}W"
+            )
+
+
+if __name__ == "__main__":
+    main()