How to Read This The Numbers & Methodology

↓

The numbers

Opportunity Score — How good this buying opportunity looks right now. Multiplies three things: stock quality (is this a strong stock?), recovery odds (has it bounced back from pullbacks like this before?), and the pullback gate (is it actually in a dip right now?). Stocks near all-time highs score low regardless of quality. Stocks with quality below 50 are excluded entirely — backtesting showed this cuts extreme downside by 43% while keeping 83% hit rates. Higher = better opportunity.
Pullback — How washed-out this stock is right now. 0 = no pullback, 100 = extreme selloff. This is the gate: stocks need a meaningful pullback before the Opportunity Score gives them credit. The deeper the pullback, the more weight the score carries.
Value Depth — Shown in the rationale for top stocks. Measures undervaluation across four dimensions: position in the 52-week range, distance below the 200-day moving average, RSI oversold level, and drawdown depth. 0 = near highs, 100 = deep value.
1Y / 3Y / 5Y Prob — The chance this stock was higher after 1, 3, or 5 years when it was in a similar pullback. For example, "92%" means it was higher 92 out of 100 times.
Quality — Is this a strong stock having a temporary dip, or a weak stock that might keep falling? Looks at: Does it usually trend up over time? Has it recovered from past drops? Is the selling slowing down?
Typical 1Y — The most common 1-year return when this stock was in a similar situation before.
Downside — A bad-but-realistic scenario (worst 1 in 10 outcomes).
Past Cases — How many similar historical situations the stats are based on. More = more reliable.

How it works

Measure value depth — Score each stock on four dimensions: where it sits in its 52-week range, how far below its 200-day moving average, RSI oversold level, and drawdown from recent highs.
Strip out the market — Remove broad market moves so we only see the stock's own weakness, not just a bad day for everything.
Find matching history — Look for every past day this stock was in a similar position (similar drawdown, similar range position, similar technical setup).
Check quality — Filter for stocks with strong long-term trends and a history of bouncing back. Stocks with quality below 50 are gated out of the Opportunity Score entirely. This prevents "value traps" — cheap stocks that stay cheap. Backtested across 34 stocks over 33 years, this gate reduced the worst 1-in-20 outcomes from -42% to -20%.
Apply the pullback gate — Only give high scores to stocks that are actually in a pullback. A stock near all-time highs gets almost no score, no matter how good its quality or history.
Show the odds — Report what happened next across 1, 3, and 5 years in those matched past cases.

Important

Not financial advice. This shows what happened in the past, not what will happen next.
Not a trading system. No buy/sell signals, position sizing, or timing.
No guarantees. Stocks can keep falling — including to zero.

Main strategy 3 S&P 500 stocks every 6 months v5 PIT S&P 500: Gradient Boosted Trees confirmed by HuggingFace Chronos. 44% backtest CAGR vs SPY 12%. 10/10 walk-forward splits beat SPY.

~44% backtest CAGR →

Backtest — DCA Optimizer Historical Performance Simulation

↓

What if you invested $1,000 every month into the top stocks by Opportunity Score, instead of just buying SPY? Each month, the backtest ranks stocks by their point-in-time Opportunity Score (the score the scanner would have shown on that historical date — no look-ahead bias) and buys the top 1, 5, or 10. Each position is held for the chosen hold period, then sold. All proceeds stay as cash — no reinvestment. Stocks only — crypto (BTC & ETH) is excluded.

Click to load backtest data…

Finds stocks that have pulled back hard and shows the historical odds of recovery.