S&P 500 screener

Rank the index by estimated probability

Set a target, a horizon and a stop, and the model ranks all 500 constituents by its estimated chance of reaching that target before that stop. It does not tell you what to buy and it does not size anything.

What the number means. “Est. chance” is a trained model's calibrated estimate that the target is reached before the stop, learned from several hundred thousand labelled outcomes. On held-out data it ranks well — far better than the heuristic it replaced — but ranking well is not the same as making money. Candidates hit their stop-loss too, the return edge is not statistically significant, and none of it has been validated on live forward data. Method carries the measured figures, pulled live, including the settings that lost money.

Track record

Past candidates scored against what actually happened

Every scan is logged and, once its time horizon has passed, checked against real prices. No evaluated outcomes yet — check back in a few days.

Reconstructed from history, not from live trading. Not live-validated.

Scan parameters

A 3–5 day horizon is where the model measured best
%
3 days
%
Stock volatility

Top candidates

Ranked by estimated chance of reaching target
# Ticker Trend Price Target Stop-loss Est. chance Volatility
Run a scan to see candidates.