branch main commit 8a6f186 built last · fix(bot-status): total equity = cash + pos_capital (was showing cash only when I
CURRENT CHAMPION

V6 · V66 Cooldown(4,48)

+70,576% compound · +245% min α · 4/4 beats B&H · MDD 27.7%

GRU 2x128 ensemble · 5 seeds × 4 folds · 36 features · tb_vol_10_7 labeling

Last roundR130MTF filter · REJECT
Rounds since V66 beaten12paradigm exhaustion
Healthy bots5/5all running
Live capital (real)€110.19as of 2026-05-25 23:45 UTC · 3 invested · 2 cash
R&D signal: 12 rounds without beating V66 baseline. Refinement is exhausted — shift focus to NEW paradigms. R117 MACD_div (+179K% compound, decorrelated from V66) is the live lead. See pipeline for next experiment (R120 portfolio).

Production bots live capital · trades · state · refreshed at commit time (2026-05-25 23:45 UTC)

V2
GRU Ensemble (always-invested)
capital (live)€22.88 trades1 stateINVESTED backtest min α
V3
GRU + DailyRSI>80 + Trail 10%
capital (live)€23.00 trades0 stateCASH backtest min α
V4
GRU 3-Regime adaptive (V5.4 Robust-5)
capital (live)€23.31 trades0 stateINVESTED backtest min α+117%
V5
V115_cmp (GRU + peak_drop + ratchet + regime cooldown)
capital (live)€20.01 trades0 stateINVESTED backtest min α**+238%**
V6
V66 Cooldown(4,48) (GRU defensive uniform)
capital (live)€21.00 trades0 stateCASH backtest min α**+245%**

Latest experimental results most recent 3 rounds

R130 mean-reversion REJECT
2026-05-25

Multi-TA combination strategies — 10 standalone (gap from R117)

namedescriptionfold_0_pnlfold_1_pnlfold_2_pnlfold_3_pnlfold_0_alphafold_1_alphafold_2_alphafold_3_alphacompound_pnlmin_alphabeats_bh_foldstotal_tradesavg_wrverdict_flag
macd_div_soloR117 baseline (MACD bullish divergence alone)-71.3-55.6-11.3-31.6-55.1-116.8-222.9-50.5-92.3-222.9057946.4
macd_div_volumeMACD bullish div + volume > 1.5× 20-bar avg-59.1-24.8-25.1-22.4-42.9-86.0-236.7-41.3-82.1-236.7031648.1
macd_div_atrMACD bullish div + ATR > 60th pctile (expansion)-55.5-49.8-27.42.0-39.3-111.0-239.0-16.9-83.5-239.0036944.3
macd_cross_volumeMACD signal cross up + vol>1.2× + above zero-47.9-64.5-17.3-12.0-31.7-125.7-228.9-30.9-86.5-228.90117242.9
bb_squeeze_breakoutBB width squeeze then upward breakout + vol spike-70.0-72.3-37.8-24.0-53.8-133.5-249.4-42.9-96.1-249.40148843.0
rsi_oversold_volumeRSI<25 + 5-bar low + vol>1.5× (oversold rev)-13.34.318.414.92.9-56.9-193.1-4.023.0-193.1111457.4
donchian_breakout20-bar high break + ATR>50th pctile-45.1-44.7117.9-27.5-28.8-105.9-93.6-46.4-52.0-105.90158644.8
triple_confluenceMACD cross + RSI 30-50 + price>EMA2052.2-14.6-11.6-3.068.4-75.8-223.1-21.911.5-223.1162949.3
obv_divOBV higher low while price lower low (accumulation)-63.7-37.1-1.4-30.4-47.5-98.4-213.0-49.3-84.3-213.00276748.9
vwap_stretchPrice <-1.5 ATR below VWAP + bouncing-80.0-21.6-15.0-48.6-63.7-82.9-226.5-67.5-93.1-226.50244749.6
R128 other ITERATE
2026-05-24

Allocation policy sweep using R121 classifier (Pareto curve)

default_w_v66labelfold_0_pnlfold_1_pnlfold_2_pnlfold_3_pnlfold_0_alphafold_1_alphafold_2_alphafold_3_alphamin_alphacompound_pnlbeats_bh_foldsgate_passed
0.00/100 — pure MACD539.71351.0202.5274.7483.01253.0-31.9268.9-31.9105111.93False
0.110/90529.01236.0222.8273.4472.31138.0-11.6267.6-11.6101178.63False
0.2525/75 — moderate MACD512.91080.3255.9271.3456.2982.321.5265.521.595496.14False
0.550/50486.2860.1318.6268.0429.5762.184.2262.284.286580.24False
0.770/30 — V66-heavy464.7713.8376.7265.3408.0615.8142.3259.5142.379921.64False
0.8585/15448.7618.9425.4263.3392.0520.9191.0257.5191.075190.04False
1.0100/0 — pure V66 (= R128 only if all predicted recovery)432.6535.0479.2261.3375.9437.0244.8255.5244.870673.64True
R121 hybrid-sma REJECT
2026-05-23

Real-time regime classifier (4-class) — unlock R120 portfolio deployment

foldstartenddominant_regimeaccuracyacc_bearacc_lateralacc_bullacc_recoveryn_trainn_testtrain_time_slatency_ms_per_bar
12020-09-012022-07-01bull0.7461.00.1981.01.09756680.390.023
22022-07-012024-05-01bull0.9841.00.9841.00.85216436700.430.042
32024-05-012026-05-22bull0.9941.01.00.9881.023136930.480.018

Navigate

→ R&D History
All 40 rounds, sortable, filterable by paradigm and verdict
→ Production Strategies
V2-V6 deep dive: edge thesis, weaknesses, comparisons
→ Discarded Paths
Graveyard with failure mode + re-explore triggers
→ R&D Pipeline
Kanban: queued / in-flight / recently done
→ Ideas Inbox
Your idea capture — Claude analyzes on request