momentum devandClaude Opus 4.8
7e4925d778
Every RTH selection/entry lever has now been refuted (6 of them separate a population but don't convert through the entry). The one axis never touched is PREMARKET — the source trader's actual edge (his words: news drops pre/after-hours, NO circuit-breaker halts pre-market so moves run clean, and RTH after 9:30 is choppy/unpredictable — the exact window our bot trades). strategy/evaluation/premarket_setup.py::detect_premarket_setup — pure, operates on PREMARKET bars: gap vs prev_close + real premarket volume floor (the DXST-vs-CF tell: low pm volume = fail) + price floor + fresh premarket-high breakout + premarket-VWAP hold. Calibrated to the thin pm tape (cum pm volume p50 ~12k shares, so a 20k floor catches the genuine movers not 1%-gap noise). Real-tape validated + SELECTIVE: 6/30 fires JEM(245%)/CELZ(143%); 7/1 fires LHAI(166%)/SOC/SNDQ — 2-3 names/day (~8% of pm names), vs the liquid detector's 50%. 6 unit tests green. labeler pm1 track (build/report/validate-premarket): slides the detector over premarket bars, scores forward on PREMARKET bars via the same _forward_labels (the trader enters+exits pre-9:30), stratifies +1R by gap. Pre-registered bar locked 2026-07-01 (median pm upside >=+10%, adverse >-6%, +1R >50%, big-gap subset clears). SHADOW-ONLY. Real caveats it sets aside: thin premarket fills, and live premarket needs extended_hours execution the bot doesn't have. Full 26-day backtest building. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
54 lines
2.0 KiB
Python
54 lines
2.0 KiB
Python
"""Premarket setup detector — fires on a gapped, liquid, VWAP-holding premarket breakout;
|
|
silent on thin/no-gap/below-VWAP premarket noise."""
|
|
|
|
import pandas as pd
|
|
|
|
from strategy.evaluation.premarket_setup import detect_premarket_setup
|
|
|
|
|
|
def _pm_run(n=14, start=3.0, step=0.06, vol=20_000):
|
|
rows = []
|
|
px = start
|
|
for i in range(n):
|
|
o = px
|
|
px = px + step
|
|
rows.append((o, px + 0.01, o - 0.01, px, vol))
|
|
return pd.DataFrame(rows, columns=["open", "high", "low", "close", "volume"])
|
|
|
|
|
|
def test_fires_on_gapped_liquid_premarket_breakout():
|
|
sig = detect_premarket_setup(_pm_run(), prev_close=2.5) # gap to ~3.8 off 2.5 = ~+50%
|
|
assert sig.is_valid, sig.reason
|
|
assert sig.reason == "premarket_setup"
|
|
assert sig.entry_level and sig.stop_level
|
|
|
|
|
|
def test_no_gap_does_not_fire():
|
|
sig = detect_premarket_setup(_pm_run(), prev_close=100.0) # tiny gap off a high prev_close
|
|
assert not sig.is_valid and sig.reason == "gap below floor"
|
|
|
|
|
|
def test_thin_premarket_volume_does_not_fire():
|
|
sig = detect_premarket_setup(_pm_run(vol=500), prev_close=2.5) # gapped but no participation
|
|
assert not sig.is_valid and sig.reason == "thin premarket volume"
|
|
|
|
|
|
def test_below_price_floor_does_not_fire():
|
|
sig = detect_premarket_setup(_pm_run(start=0.5, step=0.01), prev_close=0.4,
|
|
min_pm_volume=0)
|
|
assert not sig.is_valid and sig.reason == "below price floor"
|
|
|
|
|
|
def test_knife_down_off_the_high_does_not_fire():
|
|
b = _pm_run()
|
|
b.loc[b.index[-1], ["close", "high", "low"]] = [3.1, 3.15, 3.05] # dumps off the pm high / VWAP
|
|
sig = detect_premarket_setup(b, prev_close=2.5)
|
|
# fails a continuation gate (velocity, breakout, or VWAP) — no longer a valid setup
|
|
assert not sig.is_valid
|
|
assert sig.reason in ("no premarket velocity", "not at premarket high", "below premarket VWAP")
|
|
|
|
|
|
def test_insufficient_bars_is_quiet():
|
|
sig = detect_premarket_setup(_pm_run(n=5), prev_close=2.5)
|
|
assert not sig.is_valid and "need >=" in (sig.reason or "")
|