ANALYZED OHLC CLOSE · XAU/USD REFERENCE
$
—
Synthetic minute bars
—
SIMULATION MODE.
All prices, charts and forecasts are synthetic examples, not current gold quotes. Do not use to trade.
UTC --:--:--
LIVE REFERENCE PRICE · FREE / NO API KEY
$ —
Checking free OHLC candles…
Indicative spot price only · no broker bid/ask · separate from chart candles
CONNECTING
Free live reference:
Gold API updates roughly every 30 seconds. For 5–30-minute analysis use independently loaded, genuine 1-minute OHLC candles. A live spot tick is never pasted into synthetic chart history.
LAST 10 MINUTES
—
—
LAST 20 MINUTES
—
—
LAST 30 MINUTES
—
—
LAST 1 HOUR
—
—
Market structure & trajectory
Candles · EMA 9/21 · BB(20,2) · Fib · Monte Carlo scenario
Predictive scenario engine
Causal confluence + empirical bootstrap · no calibrated trading probabilities
✧ FORWARD SCENARIO
MODEL MEDIAN AFTER 10 MINUTES
—
$
—
INDICATOR CONFLUENCE
CALCULATING
Bearish alignment
Bullish alignment
Monte Carlo P10 scenario
—
Monte Carlo P90 scenario
—
Calculating trend, momentum and market structure…
Not validated
Directional accuracy requires genuine out-of-sample forward testing.
Directional accuracy requires genuine out-of-sample forward testing.
Historical signal replay & window comparison
Directional accuracy of fixed, causal technical signals · not a broker-trade win rate · no costs
| Horizon | Direction hit rate | Signals evaluated | Signals / eligible | Naive momentum baseline | Gross signed move¹ |
|---|---|---|---|---|---|
| Preparing historical results... | |||||
Chronological holdout uses the last 20% of the selected historical window.
Replay outcomes do not prove a profitable trading strategy.
¹ Mean movement in dollars in the predicted direction, before spread, commissions and slippage. Fixed confluence-score replay is a diagnostic, not independent validation of the Monte Carlo price range. Low signal counts make hit-rate estimates unstable.
Technical intelligence
Calculated from loaded candlesSmart money & price action
—Risk & data quality
SAFETY GATESNews-hold toggle is manual. No economic-calendar feed, broker spread, futures footprint or order-book feed is connected in this prototype.
Statistical drift & Monte Carlo
30-bar ordinary least-squares slope · 900 empirical 4-bar bootstrap paths
Regression drift
—
USD per 1-minute observation
Regression R²
—
Describes past fit, not forecasting power
Simulated paths
—
Block bootstrap; not independent data
Paths closing higher
Not a calibrated trading probability
Volatility & compression lab
Bollinger (20, 2σ) vs Keltner (EMA20 ± 1.5 ATR)
Squeeze state
—
Compression is not a breakout direction
BB width / KC width
—
Below 1 may indicate compression
Fibonacci & candlestick lab
Last confirmed 240-bar swing anchors, price pattern recognition
Latest candle pattern
—
Engulfing, pin bar, doji, inside bar etc.
Closest Fibonacci ratio
—
Informational confluence, not predictive alone
Complete indicator matrix
25 diagnostics · calculated only from the loaded candle sourceAdvanced external-data modules
Not inferred from spot candlesWhy the engine does not promise a win rate
EMA, RSI, MACD, structure and price momentum overlap. This prototype combines technical features, a 30-candle regression drift and empirical block-bootstrap simulations using a user-selected history window. It separately replays the fixed confluence-direction signal on a chronological holdout. It does not claim calibrated probabilities, a proven accuracy percentage, Elliott/Gann automation, true spot-gold footprint, or verified profitable trading. A production model must include broker spread, slippage, actual news releases, independent test periods and a minimum sample of settled forecasts.
Free official macro checks (manual):
Check major announcements before making decisions.
BLS CPI / NFP schedule
Fed FOMC calendar
FRED 10Y yields