Inconclusive · Instruments & watchlist · before 2026-09-26 (committed 2026-09-30)

Early full-universe sweeps (1h? and daily, full history)

What this means

These files show every stock as profitable, but the scraper dropped the minus sign: all 185 and 135 rows with a profit factor below 1 show a positive profit, and no row in either file is negative. So the profit figures are absolute values and the '500/500 profitable' claim is false. The lists picked from these files are unaffected, because every pick had a profit factor above 1. No profit figure from these two files should be cited.

Explanation checked against the data.

By timeframe

Timeframe	Verdict	n	Result
15-minute	Not tested: files are corrupted (minus signs dropped); nothing to extend
1-hour	Not tested: files are corrupted (minus signs dropped); nothing to extend
4-hour	Not tested: files are corrupted (minus signs dropped); nothing to extend
Daily	Not tested: files are corrupted (minus signs dropped); nothing to extend

Hypothesis: Most instruments in the universe are profitable with the strategy.
Method: TradingView headline scrapes over 500 symbols (run_tradingview_scraper.py, timeframe not recorded; run_1day_sweep.py, daily). Stop, filters and period not recorded.
Result: Files show 500/500 and 495/495 instruments profitable (median PnL +25.5% / +22.3%). But every row with PF < 1 (185 and 135 rows) has a positive PnL: the scraper dropped the minus sign, so these PnL figures are absolute values.
How much to trust it: Low: PnL sign bug; settings and period unrecorded
Did it change the live rule?: the daily top-40 list and the 80-ticker HarmonicList were picked from these files (their picks all had PF > 1, so the bug did not change them)
Notes: Do not cite any PnL from these two files.

Scripts and data for this study will be linked here when the research code is published on GitHub.

← All experiments

Education and research only. Not financial advice. Backtests use today's index members (survivorship bias).