Stocks that today trigger an edge we’ve backtested over 5 years and validated out-of-sample, each measured against its own segment’s benchmark. Every pick shows its Train vs Test edge — it’s only here because it held in both. Want the actual trades and equity curve behind a setup? See the evidence →
Not financial advice. Edges are historical averages, not guarantees; position-size and use stops.
What does each column mean? (key)
| Column | What it is | Why it matters | How it’s calculated |
|---|---|---|---|
| Symbol / Name | The stock (ticker and company). | What you'd trade. | For FTSE (.L) / DAX (.D) names the broker ticker is shown. |
| Index | NASDAQ 100, S&P 500, US small/mid cap, FTSE 100 or DAX 40. | Market & cap tier; it sets which segment's edges apply and the currency. | Membership against each index's constituents; the rest of US-listed common stock = small/mid. |
| Price | Latest closing price. | Rough entry level. | Most recent daily close, from our 5-year store. |
| Move | The recent price move that TRIGGERED the setup. | The setup itself; it's the exact condition that was backtested. | Cumulative % change over the setup's lookback window. |
| RSI | 14-day Relative Strength Index (0–100). | How stretched the move is; <30 = very oversold. | Standard 14-day RSI from daily closes. |
| Setup | The exact validated rule the stock triggers now (e.g. “Fell ≥20% over 10d”). | A specific, replicable condition with a measured, validated edge. | Matched live against the edge book; hover for the full plain-English reasoning. |
| Type | Dip (buy the drop), Breakout (buy the surge) or Short (sell the setup). | Direction of the validated edge for this stock's segment. | From the signal + the sign of its measured edge. |
| Edge | On average, how much MORE this setup made than just owning the market, over the hold period. | The real expected gain vs simply buying the index — not just how far the stock moved. | Average return minus the matched market (S&P 500 for US large caps, Russell 2000 for small, FTSE/DAX abroad), over 5 years. See full proof on /evidence. |
| Discovery / Proof | The same edge in the older years we FOUND it on (Discovery) vs the recent ~18 months we held back (Proof). | THE confidence check: the pick only exists because it worked in BOTH. Proof ≈ Discovery means it's real, not curve-fit to the past. | History split at a fixed date; both periods must agree in direction and size. |
| Win % | How often the setup beat the benchmark over the hold. | Reliability; ~55–60% beats a coin flip meaningfully at scale. | Share of historical instances with positive market-relative return. |
| N | Number of historical instances the edge was measured on. | Bigger N = more statistically trustworthy. | Count of non-overlapping signal occurrences over 5 years in that segment. |
| Hold | The holding period the edge was measured over. | The edge is specific to this horizon. | The forward window (trading days) used in the backtest. |
| Why it moved | AI read on the news/fundamentals behind the move. | Context: overreaction (buyable) or broken story (avoid)? | From the daily AI analysis; NASDAQ-100 names only. |
Type:Hold:
Win%:Index:Setup:
Loading…