_meta:
  id: 35
  name: "ORB Pullback V1"
  status: documented_failure
  description: >
    [DOCUMENTED FAILURE — NOT PROMOTED]

    Phase 1 attempt: V23's gainers_leader candidate pool + pullback continuation entry.

    Diagnostic result (200d):
      - Base pullback (no quality gates): 133 trades, WR 25.6%, return -10.14%
      - All quality filters (impulse_min, depth_max, vol_contraction, vwap_floor): negative selection
        Adding each filter either left WR unchanged or DECREASED it (min 13.3%)
      - V23 immediate-entry same pool: WR 61.5% (+36pp gap)

    Root cause: V23's candidate pool selects stocks that immediately continue after breakout.
    Waiting for a pullback negatively selects against V23's edge — catches the stocks
    that stall (typically failing breakouts). All 5 quality filters showed negative selection;
    this is NOT a tunable parameter problem but a structural incompatibility.

    Conclusion: orb_pullback_v1 on V23 candidates = negative-EV. Pivoting to vwap_reclaim_v1.

    Original Phase 1 multi-engine hypothesis:

    Engine: orb_pullback_v1 (independent engine_family, not a V23 variant).
    Same candidate universe and scoring as V23 (gainers_leader candidate pool).
    Different entry: instead of immediate ORB breakout, waits for:
      1. Post-breakout impulse peak within 9:40-9:55 ET window
      2. Pullback of 25-50% of impulse move (with volume contraction)
      3. VWAP floor check (pullback can't breach VWAP by >0.3%)
      4. Continuation bar: green + above pullback extreme

    Stop: pullback_low (structural) rather than pure ATR.

    Gates (standalone 200d):
      trades >= 50, WR >= 50%, total_return >= 0%, max_dd >= -20%
    Portfolio gates (combined with V23, 600d):
      trade_overlap <= 20%, daily_pnl_corr <= 0.30,
      combined_600d_dd improvement >= 5pp vs V23 standalone (-51.25%)

    Evaluation: not standalone — portfolio contribution to V23 is the target metric.
    Run via apps/intraday_bt/portfolio_report.py for combined analysis.

    Initial capital intentionally lower ($4000) for composite sleeve weighting (40%
    of a hypothetical $10k combined portfolio). For standalone comparison use $10000.

strategy_mode: orb

orb_strategy:
  engine_family: orb_pullback_v1
  live_readiness: research_only

  orb_minutes: 5
  sim_bar_minutes: 5

  entry_direction: long_only
  order_timeout_minutes: 45

  allow_doji_breakout: true
  allow_red_to_green_breakout: true

  # === Candidate filters identical to V23 (gainers_leader pool) ===
  min_price: 10.0
  min_avg_dollar_volume: 25000000
  min_atr_14: 0.50
  min_atr_pct: 0.04
  min_rvol: 1.5
  min_abs_gap_pct: 0.02
  min_premarket_dollar_vol: 1500000
  max_candidates: 20
  max_candidates_per_sector: 3
  min_candidates_to_trade: 1
  ticker_cooldown_days: 0
  max_gap_pct: 0.04

  min_candidate_breadth: 0.60
  market_regime_spy_threshold: 0.0015
  market_regime_ticker: QQQ
  rolling_loss_days: 7
  rolling_loss_threshold: -0.07
  max_simultaneous_entries: 3
  min_breakout_rel_vol: 1.2

  # === Scoring weights identical to V23 ===
  weight_rvol: 0.35
  weight_gap: 0.20
  weight_dollar_vol: 0.05
  weight_premarket_dollar_vol: 0.25
  weight_body_ratio: 0.0
  weight_momentum: 0.15

  # === Stop / exit parameters (base ATR same as V23) ===
  atr_stop_multiplier: 0.75
  breakeven_at_r: 1.0
  trailing_at_r: 1.0
  trailing_stop_atr_multiplier: 0.8
  trailing_tighten_at_r: 2.0
  trailing_stop_atr_multiplier_tight: 0.3

  partial_exit_at_r: 99.0
  partial_exit_pct: 0.50

  risk_per_trade_pct: 0.05
  max_position_pct: 0.70
  daily_max_loss_pct: 0.05
  max_stops_per_day: 5
  exit_minutes_before_close: 5

  slippage_bps: 5.0
  initial_capital: 10000

  compound_returns: false
  daily_budget_reset: true
  settlement_days: 1

  drawdown_governor_threshold: 0.025
  drawdown_governor_min_scale: 0.30

  streak_sizing_win_bonus: 0.70
  streak_sizing_max: 2.5

  # === Pullback entry — core engine feature ===
  pullback_entry: true
  pullback_max_bars: 8
  pullback_min_retracement_pct: 0.25
  pullback_stop_at_low: true

  # === Extended pullback controls (orb_pullback_v1) ===
  # Impulse peak must form by 9:55 ET (25 min from open)
  pullback_impulse_window_end_min: 25

  # Impulse must move at least 0.4× ATR above breakout level
  pullback_impulse_min_move_atr: 0.4

  # Pullback depth: 25% to 60% of impulse move
  pullback_depth_max_pct: 0.60

  # Pullback phase must have lower avg volume than impulse phase (70% threshold)
  pullback_volume_contraction_ratio: 0.70

  # Abort if pullback penetrates VWAP by more than 0.3%
  pullback_vwap_floor: true
  pullback_vwap_floor_tolerance_pct: 0.003

  # Stop: structural pullback low (not VWAP — cleaner for initial testing)
  pullback_stop_mode: pullback_low
  pullback_stop_vwap_buffer_pct: 0.002

  # Reclaim bar must have 1.2× average post-ORB bar volume
  pullback_reclaim_confirm_rel_vol: 1.2

universe:
  source: midlarge

backtest:
  start_date: null
  end_date: null
  lookback_trading_days: 200

cache:
  enabled: true
  dir: data/cache/intraday

output:
  dir: runs/intraday_orb
  verbose: false