XAU Squeeze Sniper 【BASE】｜Gold-specialized "Explosion Sniper" Breakout EA

Development History

“Recently, gold prices have been so volatile that ordinary EAs cannot withstand the drawdown…”

“And if you use a stability-focused EA, you might miss big trends…”

Do you have such concerns?

Due to inflation, geopolitical risks, and the活性化 of algorithmic trading, modern markets are rapidly entering a “high volatility” regime.

Old EAs that were designed for calmer markets (10 years ago) cannot handle the coming rough seas.

January 30, 2026. Gold moved $768 in one day.

“The spread widened.” “Stops didn’t trigger in time.” “Stopped out after waking up in the morning.” Such voices overflowed on social media.

In 2024, gold rose by +26% for the year. The daily average range was about $33.

In 2025, +65%. Daily average $60. Yet people still said it was getting rough.

And 2026.

January alone +13.33%. February +10.58%. March -16.56%.

Daily average range reached $170—five times that of 2024.

What would happen if you run averaging down or martingale strategies in this market?

Just imagining it sends a chill down the spine.

Currently Best Selling Products at GogoJungle

EA with “AI-equipped” or “win rate >●●%” claims dominate the top in the 30k-50k yen range.

Sales in itself isn’t bad. If there’s demand, it sells—this is market principle.

However, reading those reviews reveals another scene.

“Forward testing shows clean wins, but in my real account entries never occur.”

“I don’t understand why the AI entered here. If I have a losing streak, I can’t decide whether to keep it running.”

“A drawdown expands from averaging down, I can’t sleep at night. Yet it’s stable, so I can’t let go.”

What users truly want is not flashy numbers but the peace of mind to verify for themselves why it won and why it lost.

And while many top EAs remain MT4-only, users demand MT5 support.

So we developed 【XAU Squeeze Sniper】.

This system specializes in pinpointing the “explosive trend” unique to gold (XAUUSD) with a Squeeze Break and Trend logic*.*ONLY in FULL version

Furthermore, by incorporating our unique “hibernation filter (Regime Detection)” derived from a decade of backtesting analysis, we achieved overwhelming defensive power: aim for explosive profits in favorable markets and physically halt the system in losing markets.

XAU Squeeze Sniper is a “defense-first” design—a “bomb-dispersion sniper.”

Maximum drawdown 3.54% (absolute $366). This is the result after 10 years 7 months and 423 full tick backtests.

No averaging down. No martingale. Maximum position 1.

We chose not to chase flashy wins but to prevent exits.

Furthermore, in V3, two improvements derived from forward testing (April 2026~) are built-in by default. Strengthened hibernation filter threshold (ADX 25→30) and TP1 position management (preventing full rally from profits turning into losses). Both were confirmed to improve via historical data testing (three periods).

Why this design? Behind the validation data

This EA was not developed by “picking logic by feel.”

First, we compared both mean-reversion and trend-following strategies with data.

Backtest results for contrarian inside the band: best PF 0.89.

Squeeze breakout (trend-following): PF 1.26–1.46.

The numbers were clear. The conclusion to discard mean-reversion came not from intuition but from actual running results.

All backtests used real broker data. Not MetaQuotes demo (ideal spread) but history from actual market conditions. It includes a dynamic spread filter that automatically avoids entries when spreads widen to three times the normal mid, preventing wasted trades in spread-stretched phases.

And,

an exhaustive search over 200+ parameter combinations by AI.

Across Bollinger Band period × deviation × squeeze judgment period × ATR stop multiplier × risk-reward ratio × various filters—the total combinations exceed 200. Even if a fast test yields good numbers, only those that pass full tick-by-tick verification (98% history quality) are adopted. It’s not “cherry-picked numbers” but “tested and chosen from all.”

Three Core Themes

① Trend-following logic that discards contrarian—data-backed choices

This EA does not perform contrarian trading at all.

The decision not to do it was not based on instinct. We implemented the contrarian module for testing and confirmed with data that, in gold, trend-following underperforms over mid-to-long term.

This EA detects breakouts from Bollinger Band squeezes (contraction) and rides the initial move of the trend.

Actual entry conditions, exit conditions, and filter thresholds are disclosed in detail in the logic manual for buyers. You can verify for yourself why entries were made or omitted.

Stop loss:

Dynamic setting by ATR (×SL multiplier) *listed in the manual

Because SL width automatically adjusts with market volatility, even in volatile markets like 2026, unnecessary stop-outs are unlikely

Take profit:

Fixed value take profit by multiplying SL distance with risk-reward ratio (RR=○○*) *listed in the manual

A system that lets profits grow when trends appear and keeps losses small

Additionally from V3, after entry, if price moves in the squeeze range in favor, SL automatically moves to break-evenTP1 Break-even Managementis standard-equipped. This structurally blockss the most frustrating loss pattern of “profit turned back to break-even SL loss.”

② Five-layer defense—“hibernation filter” and four safety devices

The greatest enemy of breakout strategies is the back-and-forth in ranges with low volatility.

The EA constantly monitors higher-timeframe ADX (daily). If volatility stays below a threshold for a period, it marks it as a “hibernation period” and physically blocks entries.

Hibernation Filter (Winter Filter): In low volatility markets where daily ADX is below 30, do not enter. This is the most critical safety valve preventing false breakouts in ranging conditions. This single filter quietly carried us through the low-volatility hell from 2016–2021. In V3, threshold raised to 30 (from 25) to improve tolerance for ranges with volatility but no directional movement in 2026.

Time-of-day Filter: Automatically avoids time periods with frequent false signals

TP1 Break-even Management: When unrealized profits reach the squeeze width, SL is moved to break-even. Structurally blocks full drawdown losses (V3: standard ON)

Weekend Entry Stop: Automatic stop to new entries after Friday 8pm

Daily Guardian: If daily loss reaches 5% of margin, all trades for that day are stopped. The EA makes the exit point automatically, preventing emotional mistakes

“A defensively strong EA” may seem dull, but surviving traders are invariably quiet.

③ AI parameter optimization with “tested all”—settings chosen from 200+ combinations

Default values are not decided by feel.

BB period (20–35) × BB deviation (1.8–2.5) × squeeze judgment period (5–20) yields 64 BB grid configurations scanned rapidly.

Then add AR multipliers for SL and RR, analyze filter cost-benefit, the total combinations exceed 200. Setups after selection move directly on the XAUUSD M15 chart.

Backtesting Results (All Tick / History quality 98%)

Long-term 10 years 7 months testing(Robustness in low-volatility period from 2016–2021):

※ Note: If only 2016–2021 is used, net profit +$93 / PF1.13 / max DD $176 (including fees & swaps). This period was also used for V3 setting selection verification, so it is not completely unused data.

Profit timing bias: Of the backtest net profit +$832,about 94% occurred in 2026 (Jan–Aug)During the 2016–2025 decade, total +$47—a near-flat result focused on survival. This EA is not “one that earns every year,” but rather a personality ofprotecting and surviving, and taking profits when opportunities alignPlease understand this bias when considering.

In-Sample the last 4 years 7 months (optimization period):

※ Backtests were conducted in a real ECN environment. Spreads reflect execution prices, and net results exclude all trading costs. This EA is not about large profits but about small, steady survival. The profit per 0.01 lot is modest, but the maximum DD remains within the absolute total of $366, which is the core value.

Live Forward Results: This product has been undergoing GogoJungle’s forward testing (demo account) since April 2026. You can check the results anytime in the Forward Test tab at the top of this page (past results do not guarantee future performance).

Gold's “Anomaly” by the Numbers

2024 daily $33. 2026 daily $170. Unimaginable moves for the same gold.

In this environment, breakout strategies show their true value. Simultaneously, a defensively weak EA exits quickly.

Q&A

Q. Do you use averaging-down or martingale?

A. Not at all. We exclude logic that increases a position while in a drawdown.

Q. Will it work with brokers with wide spreads?

A. If spreads widen to three times the normal mid or exceed the absolute ceiling, entries are automatically skipped (dynamic spread filter). All backtests are in real ECN environments, validating under real market conditions including spread volatility and trading costs. Choosing ECN/STP accounts with low spreads and commissions is the simplest way to improve PF over the long term.

Q. Will it work with any broker?

A. If you can use MetaTrader 5,Hedging accountscan operate (netting accounts stop on startup; many domestic and international MT5 accounts are hedging).

Q. What margin do you recommend?

A. Start with 0.01 lot on $2,000 (~$30,000). The worst drawdown over the last 10 years 7 months is absolute $366 (18.3% of the account). Because DD% can look smaller on larger accounts due to percentage effects, we present the absolute amount.

Q. What changed in V3?

A. Two things: 1) Hibernation filter threshold increased from 25 to 30 (reducing wasteful entries in high-volatility but nondirectional ranges; improved in all three validation periods). 2) TP1 break-even management enabled by default (blocking full wipeouts from unrealized gains; at 0.01 lot, partial profit-taking is not performed; break-even move only). Both address forward-test-identified weaknesses with real measurements. You can revert to the old settings (see manual).

Q. Should I change parameters?

A. Generally, use the default. Defaults are chosen from scanning over 200+ combinations, balancing PF, drawdown, and trade count for stability. However, for those who want customization to suit risk tolerance or strategy, we offer a paid option (Parameter Tuning Guide) based on real backtest data.

Who this EA is for

• Have experienced capital blowups from averaging-down EAs
• Want priority on survival over win rate
• Want to exploit gold’s explosiveness but cannot devote time to discretion
• Dislike black-box EA; want to understand why entries occurred
• Looking for MT5-compatible gold EA

Who this EA is not for

• Expect daily entries (this EA is a “wait” strategy)
• Choose by win rate
• Believe backtests guarantee future results
• Prioritize larger profits (this EA is a capital-protecting, 0.01 lot with average annual net around $80; for aggressiveness, consider the FULL version)

For those seeking more aggressive options — XAU Squeeze Sniper FULL

BASE version is the “defensive” type with max absolute DD of $366 (3.54%).

On the other hand, XAU Squeeze Sniper FULL unlocks pyramiding: adding positions when a trend exists.

With the same 10 years 7 months backtest (net including fees & swaps), net profit grows from $832 to $4,168—about 5.0x.

Risk increases, but you can choose between offense and defense.

First, test the EA’s solidity with BASE version.

If satisfied, move to the next stage—