Elite Gold XAU/USD: Complete Strategy Points

• H4 bias → H1 pullback → M15 entry trigger. The agent watches M15; the human confirms H4/H1 and support/resistance.
• Max 2 acted signals per day.
• 1% daily loss cap — then the platform closes.

• Exports 300 M15 bars (~3 days) with EMA-20, EMA-50, RSI-14, ATR-14, refreshed every 15 seconds.
• Also streams live spread in points for the spread filter.
• Why M15: it is the trigger timeframe of the strategy above — the agent watches the trigger, the human confirms the higher timeframes.

• Events graded LOW / MEDIUM / HIGH / CRITICAL.
• Blackout on HIGH/CRITICAL: no signals from 30 min before to 15 min after.
• Times stored in UTC, alerts shown in PKT.
• Actual vs Forecast vs Previous: the surprise moves gold, not the number alone.
• Never assume "war = gold up". Combine the headline with DXY, yields, oil, volatility and price reaction.
• Multiple confirmations required before any setup.

• EMA and RSI lag. They confirm moves, they don't predict them.
• The engine can't see support/resistance or chart structure. That is the human's job.
• In sideways markets it can flip signals; the 32–68 RSI bands reduce this.

• Each alert carries: signal, live price, EMA-20, EMA-50, RSI, ATR, the reason, and news-filter state.
• Every alert ends with "SIGNAL/DEMO ONLY, no automatic order execution."
• No Telegram bot configured means no messages.
• Planned: alert only when the signal changes.

• Now (v1): signals only. The code cannot send orders to MT4.
• Next: real news/calendar API.
• Then: H1/H4 confirmation and paper-trade logging.
• Then: 3+ months of logged demo results.
• Only then: consider an execution bridge with small size, daily loss cap and kill switch.

Times shift by an hour with daylight saving. Verify with your broker's server time.

• Risk per trade 0.5%; position size derived from it.
• Spread filter: skip signals above 80 points.
• News blackout: 30 min before / 15 min after.
• Multi-timeframe: M15/H1/H4/D1 agreement required before promoting a signal (planned).
• Human rule: max 2 acted signals/day, 1% daily loss cap.

• Mark support/resistance on D1 and H4 first. Lower timeframes only for timing.
• Round numbers (e.g. 4,000, 4,050) act as magnets and reaction zones in gold.
• Note previous day high/low, weekly open, and Asian session range.
• Prefer pullbacks to a level in the direction of the H4 trend over chasing breakouts.
• Wait for candle confirmation (rejection wick, engulfing, close beyond level) before entry.
• Beware fakeouts: a break that closes back inside the level is a warning, not a signal.

• Gold math: 1 lot = 100 oz, so a $1 move = $100 per lot (0.10 lot = $10 per $1).
• Lot size = (account × risk %) ÷ (stop distance in $ × 100). Example: $1,000 × 0.5% = $5 risk; with a $10 stop, lot = 5 ÷ 1,000 = 0.005 — round down to the broker minimum or skip.
• Stop placement: beyond structure (swing high/low), roughly 1–1.5× ATR. Never an arbitrary tight stop.
• Minimum reward:risk 1:2. Skip setups where the next level is too close.
• Never widen a stop. Moving it to breakeven after a clear structure break is fine.
• Consider partial profits at 1R and trailing the rest behind structure.

• Spreads can widen 5–10× and slippage is common on CPI, NFP, FOMC. Stay out of the blackout window.
• Don't trade the first spike. Wait for the retest after volatility settles.
• Close or reduce positions before high-impact events if you can't monitor them.
• Friday late session and weekend gaps: reduce size and avoid holding large positions.

• Central banks bought a net 289 t in Q2 2026 — the highest Q2 on record, up 62% year-on-year.
• First-half total was about 345 t, the lowest H1 since 2022, because Q1 was weak (revised to 57 t).
• Biggest Q2 buyer: National Bank of Poland (51 t). China's central bank (PBoC) kept accumulating. Bank of Russia was the largest seller (22 t).
• Intentions: in the WGC reserves survey, 45% of central banks plan to increase gold reserves over the next 12 months.

• Outflows of about 45 t in Q2 as prices eased.
• ETF flows follow price and rates — they are a short-term sentiment signal, not a leading one.

• 327 t in Q2 (571 t in H1) — large physical and OTC demand.
• Asia, especially China and India, matters for physical demand. Watch their buying seasons and price sensitivity.

• Down about 17% year-on-year in Q2 because of high prices.
• India is the key market to watch — festivals and wedding season drive swings.

• An AI stock boom pulls capital into equities (risk-on) — that can reduce investment demand for gold.
• An AI selloff can send money back to gold.
• Bubble and valuation fear: concerns about AI valuations or a tech correction can raise safe-haven demand. But remember — a sharp equity crash can hit gold first through forced liquidation.

• Interest rates and USD: heavy AI investment can support US growth, a stronger USD and higher yields — negative for gold. If AI spending raises productivity and lowers inflation, rate cuts become more likely — positive for gold.
• Inflation channel: AI data-centre buildout raises demand for power, copper and other commodities, which can feed inflation and inflation hedging.
• Real demand: gold is used in electronics, chips and connectors. Technology is a small share of demand (the WGC reports modest growth) — not a main price driver.
• Market structure: algorithmic and AI-driven trading speeds up moves and creates sharp spikes and stop hunts around news and levels. Expect faster reactions and more fakeouts.
• Fintech access: gold ETFs, tokenised gold and mobile trading make it easier for retail and Asian investors to buy, increasing flows.

How crypto and gold relate

• "Digital gold" is not a stable rule. Bitcoin–gold correlation keeps changing: in 2026 it ranged from strongly negative (about -0.9 on a 30-day basis in March) to above +0.5 on a 90-day basis by late August. Long-term average is only about +0.1.
• Competing store of value: some investors choose Bitcoin or crypto ETFs instead of gold. Strong crypto inflows can reduce part of gold's retail and speculative demand, especially among younger investors.
• Common drivers: both react to USD, liquidity, real yields and Fed expectations. Currency-debasement fears can lift both together.
• Risk behaviour differs: crypto usually behaves like a high-risk asset and falls hard in risk-off or liquidity stress. Gold is the defensive asset — but in a sharp liquidation both can fall together.
• Forced selling: big crypto crashes, exchange hacks or leveraged liquidations can make traders sell gold to cover margin — a sudden dip in gold that is not about gold itself.
• Stablecoins and USD demand: stablecoins are backed mostly by US Treasuries, adding some demand for USD assets. Tokenised gold gives crypto users an easy route into gold.
• Regulation news: spot crypto ETF decisions, bans or tax changes move money between crypto, gold and equities.

How to use it

• Treat crypto as a risk-sentiment and liquidity gauge, not a signal. Don't buy or sell gold just because Bitcoin moved.
• Bitcoin crash with DXY rising: usually pressure on both. Be cautious with gold longs.
• Bitcoin and gold both up with DXY falling and yields easing: debasement or liquidity theme. Supports gold bias.
• Bitcoin falling while gold rises: money moving to safety. Confirm with price action at your levels.
• Check the rolling correlation (30 and 90 day) before assuming a link, and re-check it often.
• Crypto trades 24/7, including weekends. A weekend crypto move can hint at Monday sentiment, but gaps in gold are driven mainly by news.
• Watch Bitcoin around major macro events (CPI, FOMC) to see how risk assets react before gold's own reaction settles.
• Crypto is highly volatile — never size gold trades using crypto signals.