Pegasus Gold — verifiable 11-strategy XAUUSD portfolio cBot for cTrader
Pegasus Gold Pegasus Series
A diversified XAUUSD book, built to compound.
Why it adds value
It's not a promise: it's a cTrader-native backtest (m1, real commission and swap) you reproduce yourself, with a free edition to verify it before paying. Proof over promises.
Verifiable and reproducible
Every Store figure redraws in your own cTrader with the exact recipe. A free backtest-only edition lets you check it on your broker's costs without paying.
A book, not a single bet
11 decorrelated strategies on XAUUSD in a single cBot — diversified by behaviour (trend, breakout, momentum, mean-reversion, volatility, seasonal), each isolated by its label.
Integrated money management
Position sizing, per-asset & global risk caps and a 25% drawdown guard baked in. Sell it as Compound; the risk layer ships calibrated.
Backtest↔live parity
The same signal engine in backtest and live. The anti-copy uniqueness (60–120 s jitter) acts only on live/demo, never in backtest: you reproduce the exact figure.
What changed
Every release since the current book was published, newest first. Version numbers only ever go up for something you can notice.
v2.3 · August 2026 — Guards that remember
The book is untouched again: same strategies, same parameters, same risk, same published figures. This one is about everything the bot used to forget the moment it restarted — and about a guard that could arm on a reading that was never real.
- A halt now survives a restart. If the max-drawdown guard stops the bot, it stays stopped. Before, restarting the platform brought it back to life — precisely what that guard exists to prevent. The daily-loss budget behaves the same way: a restart mid-session no longer reopens the whole day's allowance.
- A re-adopted trade keeps its history. After a restart the bot rebuilds each open position's age and its trailing high-water mark from the broker's own record, so a time-based exit fires when it was meant to and a trail never gives back progress it had already locked in.
- A guard can no longer arm on a bad reading. During a broker maintenance window a platform can hand the bot an account equity of zero for a fraction of a second. That is now rejected as unusable instead of being read as a 100% drawdown — and if an emergency close ever is warranted, it checks the market is open and spaces its attempts instead of retrying in a tight loop.
- New per-strategy Mode — retire a sleeve without orphaning its trades. Manage only stops a sleeve taking new entries while it keeps trailing, breakeven and exits running on what it already has open. Open only is the mirror image. Until now, switching a sleeve off left its open positions with nobody managing them.
- The preset now tells you what it overwrote. Selecting a named preset re-applies its values at every start — that has always been true, and it is what reproduces the published figure. What changes is that the log now names every field it wrote over, instead of discarding what you typed without a word. Set Preset to Custom and your own numbers hold.
- More room on the guards. The max-drawdown guard gains an on/off switch and a wider range, and the margin cap accepts 0 to turn it off — the code always treated 0 that way; it was the slider that would not let you type it.
v2.2 · August 2026 — Connection resilience
The book is untouched: same strategies, same parameters, same risk, same published figures. This one is about what happens when your platform loses its broker connection.
- After a reconnect the bot takes back control of the trades it left open — it rebuilds each entry stop distance from the broker's own record and resumes trailing, breakeven and time-based exits. Before, those trades kept their broker stop and target and nothing else, with nothing to tell you.
- Connection log — a lost session is reported the moment it happens, and the recovery line gives the length of the gap and how many positions were taken back.
- Build stamp — the first line of the log names the version and release date you are running, so you can confirm an update actually took effect.
- Wider symbol lookup — more broker naming schemes resolved, so a renamed instrument is less likely to leave a strategy sitting idle.
v2.1 · July 2026 — Engine reliability
Also with the book untouched. We run these cBots on our own live accounts, and that is where we found that a strategy can fall quiet with no error at all.
- Start-up verdict — one line at launch confirming every strategy has the history it needs, naming any that does not. A thin data feed now looks like one.
- History requested, not assumed — it asks for what its indicators need instead of taking whatever the platform hands it.
- Restart-proof positions — time exit and trailing high-water mark rebuilt from the broker's own record of the entry. Redeploy, reboot or change machine: open trades keep managing exactly as designed.
- Trailing on more brokers — where a broker hides its minimum stop distance, the bot derives a safe one instead of having the change refused.
v2.0 · July 2026 — New book
Both columns are the cBot's OWN cTrader backtest over the same window, on the same costs — the version you may already own, and the one on sale now. Reproduce either yourself with the free edition.
| Metric | v1.0 | v2.0 |
|---|---|---|
| ROI | +142.0% | +156.1% |
| Profit factor · per trade | 2.08 | 2.33 |
| Max DD · equity | −8.85% | −8.48% |
| Calmar · CAGR/DD | 4.81 | 5.39 |
| Trades | 707 | 639 |
Strategies — composition
11 decorrelated sleeves on ONE market, each isolated by its label. We show the behaviour family and timeframe — enough to understand the book, without revealing the alpha.
What stays proprietary: the entry/exit rules, the indicators and their parameters, the thresholds, the stops/targets and the exact per-sleeve weights are not published. The above is portfolio context, not a copyable recipe.
Watch it in action
A full walkthrough in cTrader's own backtester — install the free edition, set the period, hit run, and reproduce the exact numbers yourself.
Opens on YouTube (channel @realbacktesting).
Money management — built in
The strategies find the entries; the money-management layer decides how much lives and dies on each one. It ships calibrated to reproduce the published figure.
Contained drawdown
Max equity drawdown held to −8.48% across the 2024–2026 backtest, with a configurable 25% drawdown guard as the hard ceiling.
Per-sleeve ATR trail & breakeven
Each sleeve manages its own exit: an ATR-scaled trail locks in a winner and a breakeven move takes risk off once a trade is working.
Asset & global risk caps
A per-asset cap stops the book concentrating risk; a global cap bounds the summed risk-to-stop across every open position.
Additive or Compound
Ship it Additive for a steady band, or Compound to let the account grow the size — the growth mode the headline uses.
The whole risk layer is exposed as parameters. Keep the defaults and you reproduce the published figure. See every configurable parameter →
Pricing — you pay per strategy
One-time on the cTrader Store, no subscription. The price is simply $20 per strategy — a bigger book means more edge and more diversification, so the ladder is transparent: you pay for exactly the number of strategies you get. Today it's $220 (11 × $20).
The per-strategy price starts at $20 and rises +$5 every 5 sales, up to a $75 ceiling. Today's price is the floor — the earlier you buy, the less you pay.
At the $75 ceiling this book would read $825 — so early access is a real discount, not a countdown gimmick.
Free edition (backtest-only)
The paid cBot has a free “Backtesting only” twin: the identical full portfolio — same strategies, engine, parameters and parity — compiled with a lock.
- The cTrader Backtester and Optimizer run the FULL logic → you reproduce every published figure on your own broker's costs before paying.
- In real time (live and demo) the free edition prints a notice and stops: it cannot place a single order.
- It's a try-before-you-buy funnel published as a separate €0 Store product; the paid edition unlocks live/demo trading.
Quick usage manual
From install to reproducing the published figure.
Verify first with the free edition
Install the Pegasus Gold — Backtesting only (free) edition and run the Backtester with the recipe: m1 (opening prices) · compound base 80,000 · commission auto · your broker's spread. You should see a figure in the neighbourhood of the published one.
Buy and install the full edition
Add Pegasus Gold (from $220) from the cTrader Store to your account.
Attach to ONE chart
The cBot manages its timeframes (H1–D1) internally. Attach it to a single H1 chart of XAUUSD; it opens the rest.
Check the symbol mapping
Gold is XAUUSD on virtually every broker, so no mapping is needed. If yours uses a suffix, set it in Symbol overrides; it auto-maps raw/ECN suffixes.
Account and sizing
Works on any standard/live account in EUR or USD. Default: Compound sizing with an 80,000 base. Adjust Initial balance or the Risk scale for your account size.
Pick the preset (optional)
Leave the shipped preset, or read the parameters page to understand every guard before you touch a dial.
Run it 24/5 on your own server
Optional, but this is how we run ours.
Running the cBot on your own computer means it stops whenever you close the laptop or the machine restarts. If you want it trading continuously — and sending you Telegram alerts when it opens, closes or trips a risk guard — you can run it on a small VPS with Docker.
One caveat worth knowing before you choose: cBots executed on cTrader Cloud have no outbound internet access, so external notifications do not work there. Your own server does not have that restriction.
Docker, pinning the engine version, systemd, resource limits and monitoring — the exact setup behind the live statistics on this page.
Honest notes
- Figures are a cTrader-native backtest over 2024–2026 (m1, real commission and swap, FP Markets' live spread), reproducible in your cTrader. See proof →
- Costs vary between brokers — your spread, swap and commission won't match FP Markets to the decimal. Verify with the free edition on your own account before you buy.
- This is a retail, real-money product — not a prop challenge. There is no FTMO rule set to breach; you run it on your own capital.
- Gold has been in a strong bull regime over 2024–2026 — a 2.5-year window is shorter than our 5-year prop backtests, and past performance never guarantees the future.
- The anti-copy uniqueness (60–120 s jitter) acts only on live/demo, never in backtest → you reproduce the exact figure.