Every candidate is walk-forwarded on each selected timeframe, then confirmed on Yahoo NQ=F (closest public tape to NQ1!). Synthetic discovery is not a validated badge. One cycle is one market — NQ survivors are not ES edges. Operator manual
Search mode Systematic Evolutionary Focused Literature
Random samples across selected families.
Contract Nasdaq-100
NQ MNQ
Highest vol. Opening drive, ETH handoff. Mini for history, MNQ to size.
S&P 500
ES MES
Same 09:30–16:00 ET session. More mean-reversion literature. Use MES to size.
Gold
GC MGC
COMEX core 08:20–13:30 ET — not the NYSE open. Asia and London matter.
Bitcoin
BTC MBT
CME futures, not a 24/7 perp. US 09:30–16:00 is liquidity overlap, not a cash open. Use MBT to size.
Timeframes 5m 15m 30m 1h 4h 6h 1D 2D
Each selected chart is walked independently. A 5m edge that dies on 15m is not promoted twice. Pine is locked to the timeframe it survived on. 4h/6h Yahoo confirmation resamples 1h (~2y); 1D/2D use Yahoo daily (~10y).
Candidates this cycle · 24 Families · 21 selected Core Unusual All None
Core is session structure, day-type, VWAP, overnight inventory, and documented trend. Unusual is failed auctions, sweeps, lunch, climax. All 38 families still have to clear the same gates.
Start on 5m · 15m · 30m · 1h · 4h · 6h · 1D · 2D