How the scan actually works
The short version lives on the homepage. This is the long version — what each setup checks, and exactly how a signal earns its way onto your scan.
The setups
Range expansion
Price clears a defined consolidation range with volume confirmation well above average — not just a new high, but a new high that the market actually showed up for. This is the strongest setup in the lineup by real historical expectancy, and the reason it's checked first when more than one setup fires on the same bar.
Momentum reversal
A fresh RSI threshold cross — momentum tipping from oversold or overbought back through the midline — filtered so it only counts when the broader trend agrees. This isn't tracked as a separate setup type; it's folded directly into the breakout trigger, since a fresh cross is one of the conditions breakout actually checks.
Trend continuation entry
Price retraces to its 30-period moving average within an already-established trend, then reclaims it — a shallow dip bought, not a trend break. This setup has historically run a thinner margin than breakout, which is exactly why it's one of the two the expectancy gate actively watches.
Volatility regime shift
Both the bar's range and its ATR expand off a compressed base at the same time — a sign that whatever was compressing price action is letting go, often before the market has committed to a clear direction. An earlier warning than breakout, with a corresponding tradeoff in reliability.
Mid-trend flag
A brief, tight consolidation inside an already-active trend — a flag, in chart-pattern terms — entered on resumption past that small range. The rarest of the five triggers in practice, and treated with the most caution: it needs a longer track record before the gate will fully trust it.
Market context gate
Before any stock setup is even considered, SPY's own price is checked against its own 30-period moving average. If SPY is trending down, only short setups are allowed through on individual stocks; if it's trending up, only longs. This runs independently of which setup fired — it's a gate on direction, not on pattern. Crypto symbols bypass this filter entirely, since there's no equivalent broad-market reference for them.
Expectancy gating
Every setup type, and every direction (long and short), has a real historical performance record — the mean R-multiple it's actually produced, computed from real trade history, not a backtest assumption. A setup or direction only clears the gate once it has a meaningful sample size behind it (fewer trades than that, and it's treated as "not enough evidence yet," not blocked). Below that sample threshold, a setup is gated the moment its real performance goes negative.
This runs both ways: when a setup or direction's real performance turns positive again, it's un-gated automatically on the next scan. Nothing about this is manually curated — the gate reflects whatever the numbers say, updated on an ongoing basis.
Why gating matters more than pattern-matching
Most scanners will show you every stock that technically matches a shape. That's necessary but not sufficient — plenty of technically valid patterns have a historically weak or negative edge in practice. Scorance's scoring system (a 0-100 score across trend alignment, momentum, and volatility conditions) decides whether a setup is worth considering in the first place; the expectancy gate then decides whether that setup type or direction has actually been worth trading, based on real outcomes. A signal has to clear both before it reaches your scan.