Volatility Breakout Scanner

One sweep across the top 100 pairs for the full breakout-thrust stack: BB breakout, band expansion, 4H EMA alignment, volume climax, ATR turning up, ADX rising above 20. Fires produce a complete trade plan — retest entry, 2.0×ATR stop, 3 / 4.5 / 7.5 ATR targets, 4h time stop.

Stack pips (left→right): BB breakout · BB expansion · HTF EMA · Volume biggest of 10 · ATR rising · ADX>20 rising. Signals evaluate closed bars only. FIRE = all six on the latest closed bar. Recent = full stack fired within the lookback window; the plan is frozen at the fire bar's ATR and band level. ▲/▼ marks setup direction (long/short); shorts are the mirrored stack — breakdown below the lower band with price under the HTF EMA. Retest: PENDING = limit at the band not yet filled, FILLED = price came back to the entry, EXPIRED = no retest within 12 bars, STOPPED = SL traded through after the fire. Backtesting showed the retest entry carries most of the edge — chasing market fills roughly halved the profit factor. TP1 Hit tracks the trade after a FILLED retest: TP1 ✓ means price reached the 3×ATR target before the stop (hover for how many bars it took), SL ✗ means the stop traded first, RUN means the position is still open, AMBIG means TP1 and SL both sat inside one candle so the order can't be known from this timeframe — check a lower TF. The 4h time stop is a discretionary rule and is not auto-evaluated here. Timestamps: FIRED badges show the signal bar's close time (the plan is anchored to that bar); Detected is the wall-clock moment this browser first saw the fire, so it measures your scan lag — if the scanner wasn't running at bar close, detection lags the signal. Detection times are stored locally and survive re-scans. Rating (1–10) is a heuristic quality score measured at the fire bar — thrust strength in ATRs, volume climax magnitude, ADX level, band-expansion rate and 7d mover status, minus a staleness penalty for older fires. It ranks setups against each other; it is not a probability of profit and hasn't been validated against outcomes yet — log it alongside results to find your own cutoff. Sizing: you set the account and the risk ceilings once at the top; every fire then sizes itself. Quantity comes from how much of the account you are willing to lose (base risk × rating multiplier × today's dial) divided by the distance to the stop — so a wide stop buys a small position and a tight stop buys a big one, and the loss on a stop-out is roughly the same either way. Leverage is an output of that, not an input: it is clamped so the estimated liquidation price still sits beyond your stop, which on a wide-stop pair can mean far less leverage than your ceiling allows. Raising your max leverage does not make a position bigger; it only lowers the margin you must post. Fuel (0–5) asks whether the crowd is positioned against the breakout and may be force-closed into it: open-interest build, funding sign, account and position crowding, taker flow, and an estimated liquidation cluster between entry and TP1. ⚡ marks the case worth caring about — open interest climbing while funding still runs against the move, meaning the other side is adding and paying to hold. Be clear about what this is: no exchange publishes where liquidations rest, so the cluster is modelled from open interest and assumed leverage tiers, and crowding can persist for weeks without resolving. Fuel boost (off by default) lets the score nudge size ±25%; leave it off until your own journal shows fuel actually separates winners from losers. Columns: the table shows the essentials by default. 24h %, SL, TP1 and ADX are hidden behind the Columns button — every one of them is still in the drawer and in the copied plan, and SL and TP1 in particular are not numbers you act on from a table cell. ADX is deliberately triple-counted otherwise: it is already a stack pip and a rating component. Data: Binance spot for signals, Binance USDT perps for positioning, all client-side. Educational tooling, not financial advice.