How we find setups — and what the data shows.
StockSetups is built on a testable premise: that clean chart setups do a little better than chance. In our sample, fresh breakouts show a small average edge — not a guarantee. Here's exactly how the scanner works, where the data comes from, and what the numbers do and don't say.
The breakout-edge study
The cleanest test of the premise: take every fresh 50-day-high breakout across our US-equity universe over roughly the last 252 sessions of stored history — no cherry-picking, winners and losers alike — and measure what happened next.
A ~50% hit rate is modest on its own; whether a trade pays also depends on its entry, stop and target, which this fixed-horizon measure ignores. In this sample the average edge is small, and past results don't guarantee future ones. The numbers refresh daily.
Hypothetical results from price history and published signals, not actual trades; they exclude commissions and slippage. Past performance doesn't guarantee future results.
This is a historical base rate. For the live, forward-tracked record of the actual signals we publish — by lane, by pattern, with sample sizes — see the Signal Report Card.
Market-wide base rates after big one-day moves, gaps and volume spikes: What happens after a stock jumps 10%.
How the scanner works
- 1. Scan the whole market, nightly. After the US close we pull daily bars for the full ~12,300-symbol US-equity universe (stocks + ETFs) and run the same chart- and candlestick-pattern detectors over every one — triangles, flags, channels, bases, head-and-shoulders, and the major candle signals.
- 2. Sort each setup into one of four lanes. A detected setup is classified by where it is in a breakout's life cycle — setting up, breaking out, broke out, or retesting the breakout. The lanes are the product's spine.
- 3. Score it 0–100 and write a trade plan. Each setup gets a quality score (A–D grade) from base tightness, reward:risk, distance from the 50-day, daily-range calm and pattern structure, plus a concrete plan: entry, stop, target, and reward:risk. The level you act on is explicit.
- 4. Enrich, then watch it live. We layer on fundamentals, short data, news, SEC filings, Congressional and insider activity and Reddit sentiment — then an always-on engine watches the tape intraday and fires 35+ alert types with a "why it's moving" catalyst.
Where the data comes from
Licensed market data, official filings and named public feeds — nothing synthetic.
| Price bars (daily + intraday) | Databento — the US-equity feed behind the scan and the live tape. |
| Fundamentals, SEC filings, insider trades | SEC EDGAR — financials, 8-K/13D/13G filings, and Form 4 insider activity. |
| Short interest / days-to-cover | FINRA's public short-data feed. |
| News | Public RSS headlines (e.g. Yahoo Finance), categorized into catalysts. |
| Congress + Reddit intel | House/Senate disclosures (and OGE executive 278-Ts) plus Reddit (WSB) mention data from ApeWisdom. |
What the numbers don't say
Chart and candlestick patterns frequently fail. A forward return measured at a fixed horizon ignores the entry, stop, and target a trader would actually use, so it understates good risk management and overstates a buy-and-hold of every signal. The breakout study is a base rate over a finite history, and our live record is deliberately small early and grows over time. None of this is financial advice — it's a tool to find and frame setups, not a promise of returns.
Questions
Is the breakout edge a backtest or a live record?
Two separate things. The breakout-edge study on this page is a historical base-rate measured over roughly the last 252 sessions of our stored price history — every fresh 50-day-high breakout, winners and losers. Our live, forward-tracked record of the actual signals we publish lives on the Signal Report Card, which starts the day we began logging and is not fit to history.
Where does StockSetups get its data?
Daily and intraday price bars come from Databento; fundamentals, SEC filings and insider Form 4s from SEC EDGAR; short interest from FINRA's public feed; news from public RSS; plus Congressional disclosures and Reddit mention data from ApeWisdom (apewisdom.io). There is no synthetic or fabricated data.
Why only long setups?
The scanner is long-only by design: it sorts every detected setup into one of four lanes that track a breakout's life cycle — setting up, breaking out, broken out, and retesting the breakout — rather than trying to be everything to everyone.
Does a high score mean I should buy?
No. The 0–100 score ranks setup quality; it is not advice and chart patterns frequently fail. Every signal ships with a trade plan (entry, stop, target, risk:reward) precisely because the level you act on matters more than the score.
Get daily signals & real-time alerts.
StockSetups scans ~12,300 US stocks & ETFs after every close and sorts every long setup into four ranked lanes — each with a trade plan — plus an always-on engine firing 35+ real-time intraday alerts. Free for 7 days, no credit card.
Start free — 7-day trial, no card →Free daily signals by email. Top 3 setups per lane after every close. No account needed.
Unsubscribe anytime · Privacy