Atlas Bitcoin

Atlas Bitcoin is a defined-risk, multi-strategy BTCUSD Expert Advisor for MetaTrader 5. Every trade has a hard stop-loss. No grid. No martingale. No averaging down. Six carefully selected, complementary strategies operate through one shared portfolio-level risk engine, offering adaptability to different market conditions.

Unlike grid-based Bitcoin robots that risk catastrophic loss when the market does not recover, Atlas caps the risk on every position and reports its performance transparently. Backtested from Jan 2018 to August 2026 across 1,519 positions, Atlas produced a 4,685% total compounded return with ~25% maximum balance drawdown and a 1.45 profit factor at 1% risk/strategy. Higher and lower risk profiles are available, and their corresponding returns and drawdowns are disclosed below.

Atlas runs on one BTCUSD H1 chart using a hedging account. Three straightforward risk profiles, an on-chart risk dashboard, a free Strategy Tester demo and a complete illustrated user guide make it easy to evaluate and operate. 

WHY ATLAS BITCOIN IS DIFFERENT

• A hard stop-loss on every trade, always. You always know your worst case per position.
• No grid, no martingale, no averaging. Losers are cut, never doubled.
• Six independent strategies behind one shared risk engine, each built for a different market condition, so no single regime dominates the book.
• Strategy rules are frozen. The EA verifies them at start-up and refuses to run if anything has been altered. There is nothing to over-optimise.
• Honest reporting. The real drawdown, the real trade count, and the price of every risk setting stated plainly.

WHERE IT COMES FROM

Atlas is a research program first and a product second.

Nine research campaigns went into the search for strategies that genuinely hold up on Bitcoin, and most of them promoted nothing at all. That is the method rather than a confession: an idea that cannot survive being attacked has no business managing anyone's money, and the great majority did not survive.

Six did. Each passed a ten-gate validation process and was then set against sealed data it had never been allowed to see, on the principle that the only honest test of a strategy is one it cannot have been fitted to.

Those six ship as one Expert Advisor, already decided. You choose the risk; the research settled everything else.

SIX STRATEGIES, ONE RISK ENGINE

Bitcoin does not behave the same way from one month to the next. It trends, it breaks out, it grinds sideways, it capitulates. A single approach that shines in one of those regimes tends to bleed in the others.

Atlas runs six independent strategies that are complementary by design, each suited to different conditions and each largely uncorrelated with the rest. When one is out of favour another is usually in it, so the equity curve is steadier than any single strategy could deliver alone.

The part most multi-strategy systems get wrong is what happens next. Six strategies that each size themselves independently will happily stack six positions into one direction at the worst possible moment. In Atlas they do not decide alone: they share a single risk budget and negotiate for it.

RISK MANAGEMENT

• Hard stop-loss on every trade, no exceptions.
• No grid, martingale, or averaging.
• Percent-risk position sizing calculated from each trade's own stop distance, so a wider stop produces a smaller position and the money at risk stays constant.
• A shared risk budget across all six strategies, with a cap on total open risk and a limit on how one-sided the book may become.
• A drawdown throttle that scales every position down in measured steps as drawdown deepens, and a hard stop that halts new trades entirely at a level you set.
• A spread guard that blocks entries when spreads blow out, for example around news.

Across 1,519 backtested positions, not one lost more than 1.6% of the account.

PERFORMANCE (backtest, BTCUSD, compounding)

• Duration: 2018-01-01 to 2026-08-20 (8.6 years, 1,519 positions)
• Server: IC Markets demo, 1-minute OHLC model
• Costs: spread, swap and commission measured on a live broker feed, not assumed

At the shipped default of 1% risk per strategy:

• Total return: 4,685%
• Maximum drawdown: 25.2%
• Profit factor: 1.27

At the more conservative 0.17% the research itself validated:

• Total return: 108%
• Maximum drawdown: 4.4%
• Profit factor: 1.45

The same six strategies produce both, and one input moves you between them. One number is the reward and the other is what you have to sit through to collect it. The average winner is roughly two and a half times the average loser, at about three trades a week.

Drawdown above is balance drawdown, peak to trough, which is the figure MetaTrader reports as Balance Drawdown Maximal. Equity drawdown, which marks open positions to market between bars, runs a few points higher. Both appear in the tester report and neither is hidden here.

Note also that MetaTrader counts each order as a trade, and this EA splits large entries across several orders. Its report will show roughly 3,000 trades where we say 1,519 positions. Same record, different unit.

In 20,000 Monte Carlo resamples of the trade record, none finished at a loss.

Past performance, while worth assessing rigorously, never guarantees future results. Markets can always deliver never-before-seen conditions, volatility or news that no backtest could anticipate.

THREE READY-MADE PRESETS

The presets differ in one input - your risk per strategy per trade - so they are trivial to set by hand in the Inputs tab:

• Validated: Risk = 0.17. The exact level the research validated. Steadiest, and the smallest drawdown.
• Balanced: Risk = 0.5. A middle position.
• Standard (default): Risk = 1.0. Faster growth, materially deeper swings; only if you understand the drawdown figures above.

USER GUIDE AND PRESET FILES

Every buyer gets a full illustrated User Guide (PDF) and the three ready-made preset (.set) files.

To get the files, just leave a comment or send a message on this page and I will share them with you directly. You can also set all three presets by hand in seconds using the risk values above, no download required.

REQUIREMENTS

• Account type: HEDGING. Atlas holds several independent positions on BTCUSD at once, so it requires a hedging account. Netting accounts are not supported and the EA refuses to start on one.
• Symbol: BTCUSD, on one H1 chart. The EA builds its own decision grid internally, so it is not affected by your broker's server time.
• Broker: a low-spread broker is recommended; spread and fills matter to results.
• VPS: a 24/7 VPS is recommended for uninterrupted operation. Leverage 1:100 or higher.

RISK AND YOUR BALANCE (please read)

Because every position is sized from its stop rather than from a fixed lot, the account has to be large enough to fund a proper stop at your chosen risk.

Measured across full 8.6-year runs at each starting balance:

• 2,000 and above: the account takes essentially every trade. A 2,000 balance declined four orders out of roughly 2,900, which is the same as a 100,000 balance did.
• 1,000: 3,976% return, declining about a hundred orders it could not fund.
• 500: 2,126% return, and a lower drawdown than the large accounts, because the trades it declines are exposure it never takes.

Bigger is smoother, but the floor is far lower than most position-sized systems manage.

What matters more is what the EA does about it. By default it declines those trades and tells you why, rather than quietly taking a minimum lot and handing you several times the risk you set - the usual silent failure in position-sized systems, and the one that empties small accounts. A marginal balance gives you fewer trades, never larger ones.

If you would rather trade the smaller account anyway, a single input switches that behaviour on and the EA takes the minimum lot instead of declining. Be clear about what you are accepting: this deliberately exceeds the risk you set, and by the widest margin at exactly the moments volatility is highest and stops are widest. It ships switched off, turning it on is entirely your own decision, and with it on the risk you choose becomes a floor rather than a ceiling.

The on-chart panel shows your live risk per trade and your current drawdown, so you always know where you stand.

GETTING STARTED

1. Attach Atlas Bitcoin to one BTCUSD H1 chart and enable Algo Trading.
2. In the Inputs tab, load a preset (start with Validated or Balanced).
3. Set the broker GMT offset for your server. This is the one input you must get right.
4. Check the panel's risk line against your balance (see above).
5. Run it on a VPS and let it work. One instance, one BTCUSD chart.

INPUTS (PARAMETERS)

Account

• Base magic number: unique ID the EA stamps on its own trades; change only if you run a second copy on the same account.
• Broker GMT offset: how many hours your broker server runs ahead of UTC. A fact about your broker rather than a preference.

Risk

• Risk % per strategy per trade: your main growth and risk dial. Six strategies at 1% means 6% of the account at risk when all six hold a position.
• Cut risk from this drawdown %: where the throttle begins scaling positions down, in three measured steps.
• Stop trading at this drawdown %: the hard stop. Once reached, the EA opens nothing further until a human clears it by name.
• Skip entries above this spread: declines new entries when the spread is abnormally wide. Ordinary conditions sit far below the default.
• Trade min lot if balance too small: the small-account override described above. Off by default, and it raises your real risk when enabled.

Strategies

• Six on/off switches, one per strategy, all on by default. They were validated together as a portfolio, so turning any of them off is not recommended.

Display and maintenance

• Dashboard on or off.
• Optional diagnostic log file. This is the only way to see per-strategy results, since no standard report can tell which strategy opened a trade.
• Two inputs used to clear a latched stop, which requires a name on the record.

The strategy engine is pre-tuned and locked - there are no strategy parameters to configure or get wrong. You control risk; Atlas handles the trading.

HONEST LIMITATIONS

• Atlas is not a high-win-rate robot. It wins under half its trades, but the average winner is far larger than the average loser, which is how this style profits. If you need a "95% win rate", this is not that (and be wary of anything claiming it).
• It is patient. Around three trades a week, positions held for days rather than minutes, and occasional quiet stretches of a fortnight. A quiet week is the system working, not the system broken.
• Choppy, directionless stretches cost a little. Atlas pays a small toll in those regimes rather than taking blow-up risk to avoid it.
• The 1% default is six times the level the research validated, and the drawdown figures above show exactly what that costs. The choice is yours, and it is stated openly rather than buried.
• It trades Bitcoin only, from price and volume. No cross-asset inputs, no news trading.

SUPPORT

Questions and setup help are welcome; leave a comment or send a message. A full user guide and the preset files are available on request (see above).

Trade responsibly. Trading leveraged instruments carries significant risk of loss; only risk capital you can afford to lose.