▲ B/B1/B2 long entry layers▼ S/S1/S2 short entry layersTRL{n} trailingSL{n} stop-lossMAX{n} max-holdsolid arrow = A-rated · faded = B-rated
COO × STC (1H)
Strategy Stats · BGUN
Risk Profile
Equity Curve ($10,000 → simulated)
Long / Short Breakdown
Yearly Returns
Year
Trades
PnL
Positive Mo.
Trade Log · BGUN
#
Side
Src
Layers
Entry
Exit
Entry Px
Exit Px
Hold(h)
PnL
Result
BGUN-BTC Simulator — Guide
What this is: A single-symbol (BTC) backtest display for the BGUN (BOOM A/B) strategy. All data comes from the local signals.db backtest engine — not live trading.
Strategy logic (Phase190B): BOOM A/B = 1H box breakout (N=18 bars, box width < nech 30th percentile) with check_sig SF A/B rating (relaxed). Head position opens on A/B signal; same-direction BOOM signals on later bars add layers (max 4). Graded TRL exit: A act 3% / B act 1% · dist 0.1% · SL 10% · MAX 120h (both grades).
Position sizing: $10,000 starting equity. Fixed base notional $3,850 (E0×3.85%/10% — risk budget 3.85%) per head and per layer; single-coin fence E×1.54 (4-layer cap); MC70% single-pool (margin = base/40x BTC real lev). Leverage scales effective notional & PnL linearly, margin stays constant.
TRL{n} / SL{n} / MAX{n} = exit markers, {n} = layers settled at once
No exit circle on still-open positions (data-end HOLD) — entry arrow kept, no fake MAX marker
Solid arrow = A-rated · faded arrow = B-rated
Trade log HOLD rows: Positions still open at data end (backtest truncated, not yet settled) are pinned to the top of the trade log — Entry = first-layer (L1) time, Entry Px = L1 price, Exit = OPEN, Hold = live elapsed time (updates in real time), Result = HOLD. Layers of the same position are aggregated into one row (layer badge shows count, e.g. 2L); settled multi-layer exits (TRL2/MAX3…) keep the original aggregated rows.
ZR·G BTC mode: The ZR-Genesis BTC variant — ZR signals only (1H COO zero-line cross, check_sig A/B, Relaxed ON — Phase142 baseline, corrected 2026-08-07), pyramid layers (L1 $20 / L2+ $30 ×50 scale), Trail Act 2% / Tr 0.3% / SL 6% / MAX Gated+36h (production v1.8), FLIP ON. Pool switch: ONLY-A (A-rated only, higher WR) vs A/B (adds B-rated, higher PnL). Leverage switch (segLev): 1x / 3x / 5x / 10x — effective notional = $2,000 base × leverage, margin stays $2,000, PnL scales with leverage. Notional fixed at $2,000 base. Equity $10,000.
Disclaimer: Past backtest performance does not guarantee future results. For research/display only.
F4 notional regression fix: v2_notional() was hardcoded $120 base instead of cfg notional $2,000 (Phase174X fix lost in v4.0 rewrite — same bug class as v3.1). Actual position sizes were $120-225 while UI declared $2,000 → PnL understated ~17x. Fixed: base = cfg notional. v4.0.1 lev1x: full +$17,468 (ROI 174.7%) / opt +$18,201 (ROI 182.0%), WR 71.5%/72.3%.
v4.0 · 2026-08-11
F4 v2 weighted deployment (Phase174Z-R5): layer/spd gate dimensions removed — segLayer/segSpd switches deleted. Notional weighting (OBV ×1.25 + spd model ×1.5 on head AND layers, no gates) replaces the ATR speed gate. Data files now f4_sim_f4_n2000_p{full,opt}_lev{1,3,5,10}.json.gz (8 files). v2-full: +$34,438 (+39% vs base), exp 4.07, MaxDD 0.98% (~flat); OOS exp 3.22 highest of all configs.
v3.2 · 2026-08-11
Layer/spd switches (Phase174Z): F4 mode adds segLayer (ON/OFF — ZR add-on layers vs L1-only) and segSpd (off/med/p75 — ATR speed gate on layers). Matrix-validated: layer ON adds ~60% PnL (+$14,053 vs +$8,772) at higher MaxDD; spd med = prod default (WR 72.9% / PF 3.08 / MaxDD 6.2%), p75 tightens quality (WR 73.9% / MaxDD 5.3%), off maximizes PnL. Data files now f4_sim_*_ly{on,off}_spd{off,med,p75}_lev{1,3,5,10}.json.gz (2×2×3×4 = 48 combos); legacy on/med paths kept for back-compat. Superseded by v4.0 (gates removed).
v3.1 · 2026-08-11
F4 notional fix (Phase174X): live_sim_layered open/add-layer sizes were hardcoded $150/$120 instead of using cfg notional $2,000 (OBV-aligned 1.25× → $2,500). Before the fix, actual position sizes were only $120-150 while the UI declared $2,000 (total PnL $773.96, ROI 7.7%); after: $12,899, ROI 129%, risk/trade $200 = 2% of $10k now matches actual SL losses.
HOLD rows + position aggregation (Phase168L/M): positions still open at data end no longer draw a fake MAX exit circle (entry arrow kept) — DS and ZR-G identical. Trade log pins open positions to the top: Exit=OPEN, Hold=live elapsed time (updates in real time), Result=HOLD pill. ZR-G layers of the same position aggregated into one row by pos_id (Entry = L1 time, Entry Px = L1 price, layer badge shows count); settled multi-layer exits unchanged. Also: Relaxed baseline corrected to ON (Phase142, was mislabeled NoRelaxed).
v2.6.2 · 2026-08-06
Subchart sync v3: switch to logical-range sync (main → COO). Time-range sync collapsed the subchart's rightOffset whenever the main chart moved — subchart got pulled back to the price axis. Indicator series now null-padded to align 1:1 with klines, so logical indices map to identical times; the subchart keeps the same 8-bar right gap as the main chart when scrolled to the edge.
v2.6.1 · 2026-08-06
Subchart time-axis fix: main→COO sync now pads the rightOffset gap (8 bars) — the COO/STC subchart was still glued to the right price axis after v2.6, causing misalignment with the main chart. Time-range sync kept (indicator series lags the latest forming bar by 1).
v2.6 · 2026-08-06
Chart time-axis fix: removed COO→main reverse sync (setVisibleRange was overriding rightOffset 8 → latest bar glued to right price axis, chart undraggable). Main chart is now the sole time-axis master; subchart follows one-way.
v2.5 · 2026-08-06
Simplified ZR-G switches: removed Tr0.05 (0.05% retrace is below taker fees — not executable live, backtest illusion) and Gated+48h MAX mode (research-only variant). Remaining: Pool ONLY-A/A/B · Tr0.2/Tr0.3 (default/production) · MAX120h/Gated+36h (production) · Lev 1x/3x/5x/10x.
v2.4 · 2026-08-06
Leverage multiplier paradigm (Phase168I): effective notional = $2,000 base × leverage (3x → $6,000 / 10x → $20,000) — PnL now visibly scales with leverage (3x earns 3×); margin constant $2,000; notional fixed at $2,000 (segNotional removed); header shows Capital $10,000 · Notional $2,000 · Updated; Risk panel shows effective notional + margin; data regenerated n2000 only.
v2.3 · 2026-08-06
Leverage mechanism (Phase168H): segLev switch 1x/3x/5x/10x in both modes (DS default 1x, ZR-G default 3x — production-aligned); engine leverage param (margin = notional/lev, MC70% budget × lev); liquidation logic (LIQ exit at 1/lev adverse move; SL hits first at ≤10x so LIQ ≈ 0, safety-net role); Risk panel shows Leverage + margin + LIQ rate.
v2.2 · 2026-08-06
DS↔ZR-G alignment: ZR-G stats Sharpe → Frequency (same format as DS); DS entry markers now actual-filled trades view (combo9 solid / ALL35 faded, 678→175 arrows); confirmed header KPIs shared-simplified and zero Chinese in UI.
v2.1 · 2026-08-06
ZR-G markers now actual trade-layer view: entry layers B/B1/B2 (S/S1/S2), exits aggregated per settlement with layer count TRL3/MAX4/SL2/FLIP1 · trade log layer badge without × · MAX buttons English (MAX120h/Gated+36h/Gated+48h) · removed Position/Exit header KPIs (user picks via seg controls) · time-axis no longer resets to latest bar on mode switch (rightOffset 8 + idempotent setData).
v2.0 · 2026-08-06
Added ZR·G BTC strategy mode: strategy switch in header, ONLY-A / A/B pool switch, Tr0.05/0.2/0.3 trail-distance switch, TRAIL exit markers, layer/rating columns in trade log. DS mode unchanged.