📡

Crypto 15M Signal Scanner

SMC / Confluence · Top-100 USDT Perpetuals · 4H → 1H → 15M

… last scan — next — price feed —
Analysis tool, not advice. This system generates rule-based analytical signals from public market data. It does not execute trades and does not guarantee profitability. The confluence score is a quality ranking, not a probability.

1 · System Status

Last data update: —
System——
Last Scan——
Next Expected—every 15 min
Scan Duration——
Universe——
API——
Data Freshness——
Active Signals——

2 · Market Overview (4H regime breadth across the universe)

4H Regime Breadth

Top Liquidity (24h quote volume)

    3 · Universe Screener (top-100 snapshot from the last scan)

    # Symbol Price 1D 4H 1H Regime RSI ADX ATR% RelVol Funding%/8h

    4 · Active Signals

    Why were there no signals? — last scan rejection funnel

    —

    Gate (in evaluation order)RejectedSurvivors

    Each gate only sees what survived the one above it. The gate that drops the most is the one actually constraining the system — tune that one, not the thresholds further down.

    5 · Recent Signals (latest 10, any status)

    TimeSymbolDirQualityTriggerTPSLRRStatus

    6 · Performance

    Total Signals——
    Resolved Trades——
    Win Rate——
    Profit Factor—ΣR won ÷ |ΣR lost|
    Average RR——
    Streaks (W / L)——

    Cumulative resolved trades (R)

    Win / loss distribution by quality

    LONG vs SHORT

    Outcome mix

    Expectancy—mean R per resolved trade
    Max Drawdown—peak-to-trough, R (all-time)
    R Sharpe—mean R ÷ std R (per-trade)
    Avg Slippage—trigger vs fill, gap opens
    Monte Carlo (50th)—bootstrap of past R · not a prediction
    Max DD (95th pct)—bootstrap distribution of drawdown

    Score calibration — historical win rate by quality tier (Wilson 95% CI)

    TierSignalsResolvedWinsLossesWin rate95% CI (Wilson)

    Descriptive statistics of PAST outcomes per tier. Small samples are statistically meaningless — the CI shows the honest uncertainty. This is NOT a probability of future wins.

    Win rate = wins ÷ (wins + losses). WAITING / EXPIRED / AMBIGUOUS / CANCELLED signals are excluded from the denominator. Sample sizes are shown — small samples are not statistically meaningful. No probability claims are made.

    7 · Signal History

    TimeSymbolDirScoreQuality TriggerEntry ZoneTPSLRR StatusOutcomeDuration

    8 · Methodology

    Timeframe model

    4H — macro regime. EMA 20/50/200 stack + ADX classify each pair as TRENDING_UP / TRENDING_DOWN / RANGING / MIXED. RANGING markets are hard-rejected. 1H — directional confirmation. The same stack must agree with the trade direction. 15M — signal generation. All entries, stops and targets are derived from 15M structure. 5M is never used as a primary timeframe.

    Market structure (swing → BOS / CHoCH)

    Swing highs/lows are 2-bar fractals, only usable once confirmed two bars later (strictly no lookahead). A BOS is a close beyond the most recent confirmed swing in the trend direction (continuation). A CHoCH is a close against the established trend (character change). Wick breaks never count. HH/HL/LH/LL and equal-high/low pools (ATR tolerance) map the structure sequence.

    Liquidity sweep

    A candle wicks beyond a prior confirmed swing high/low (excursion ≤ 1× ATR) but closes back on the original side — a stop-hunt rejection proxy. Real breakout-and-acceptance is not classified as a sweep.

    Displacement · Fair Value Gap · Order Block

    Displacement: candle body ≥ 1.5× ATR (institutional imbalance proxy). FVG (3-candle model): candle-3 low leaves a gap above candle-1 high (bullish) or below its low (bearish); gaps smaller than 0.1× ATR are ignored. Order Block: the last opposite-colour candle before a displacement that broke structure; zone = that candle's full range. These are price-action proxies — the system does not claim to see real institutional orders.

    Entry, stop, target

    Signals are limit-of-intention style: LONG trigger = confirmation-candle high + 0.1× ATR (buy-stop); entry zone = trigger − 0.5× ATR → trigger. A signal stays WAITING_TRIGGER until real price trades through the trigger. Stop = swept invalidation structure ± 0.5× ATR buffer (never a fixed percentage; rejected if > 3.5× ATR). Targets are the nearest real swing-liquidity levels offering RR ≥ 2.5, preferring ≥ 3R when structure allows. Setups whose nearest realistic liquidity cannot fund the RR are rejected — TP is never fabricated.

    Confluence scoring (100 points)

    HTF alignment 20 · market structure 20 · liquidity/SMC 20 · momentum 15 · volume 10 · volatility 5 · risk/reward 10. Tiers: A+ ≥ 90, A ≥ 85, B+ ≥ 80 — anything below 80 is never published. The score ranks confluence quality; it is NOT a win probability.

    Hard rejection rules

    Regardless of score, a setup is rejected when: 4H and 1H both oppose the direction; RR < 2.5; 4H regime is RANGING; ADX < 15 with no emerging-trend evidence (DI aligned + ADX rising / DI spread ≥ 3); relative volume < 1.2×; stop distance abnormal; entry pressed against major opposing structure; price has already displaced too far (late entry); data stale or candles missing; the symbol already has an active same-direction signal; a duplicate setup is inside the cooldown window; or any data-integrity check fails. Signals can never depend on future candles (replay-verified).

    Lifecycle & outcome engine

    WAITING_TRIGGER → TRIGGERED (price trades the trigger; gap opens fill at the open) → WIN (TP) / LOSS (SL) / AMBIGUOUS (both touched in one candle and 1-minute data cannot determine order — never counted as a win) / CANCELLED (max holding 4h) / EXPIRED (no trigger within 12 × 15m candles). Signals are immutable after generation: entry, trigger, SL and TP are never rewritten (no repainting).

    Data source & failure policy

    Binance USDⓈ-M public market data only (no API key, no account access). If data cannot be retrieved, the scan is marked degraded/failed, previous valid data is retained, and no simulated values are ever produced. Frontend prices are labelled LIVE only when the browser successfully refreshed them from the exchange.

    This system provides analytical signals only and does not guarantee profitability.

    9 · System Logs (latest scan events)

    —