Solitaire
- Experts
-
Simon Reeves
With over ten years of experience in financial markets I bring institutional-level risk management principles to retail algorithmic trading.
My Trading Philosophy:
Risk management comes first, profits follow
Thorough backtesting and forward testing before deployment - Version: 1.0
- Activations: 8
One position. One risk. One clear exit.
Solitaire by Starpoint Trading — the one-position Gold EA for MT5. A solitaire is a single stone, set alone. No cluster, no halo, no pavé — nothing beside it to share the load, and nowhere for a flaw to hide. The stone is either good or it is not, and you can tell which from across the room. That is the entire design brief for this EA.
Live Signal: https://www.mql5.com/en/signals/2384410
User Manual: https://www.mql5.com/en/blogs/post/773462
Come chat with us in our public MQL5 channel! https://www.mql5.com/en/channels/starpoint
The Solitaire Rule
At most one Solitaire-managed XAUUSD position may be open at any time. While that position is active, Solitaire will not add another trade.
No grid. No martingale. No averaging into a loser. No recovery basket. No hedge opened against an open position. Every entry is submitted with a broker-side stop loss attached to the order — and if that stop is ever found missing, Solitaire writes it back on the next tick.
One position means there is no second entry to average a bad one and no basket to bury it in. Whatever Solitaire is holding, you are looking straight at it. That is not automatically safer than a grid — it is simply legible, and a buyer who can see exactly what he owns is in a better position than one who has been handed a smooth curve and asked to trust it.
Max open positions: 1 | Grid: No | Martingale: No | Averaging: No | Hedging: No | Stop loss: broker-side, every trade | Profit exit: fixed TP or trailing stop, your choice
Every stone has an inclusion. Here is Solitaire's.
Read this before you read the performance figures. Solitaire's default stop loss is wider than its profit target: it risks 100 basis points of price to arm a profit exit at 15. That is deliberate, and it is the most important thing to understand about this EA.
Across the 18-month test below, the realised figures were an average win of $63.79 against an average loss of $410.03. From those two numbers you can calculate the win rate the strategy needs simply to break even:
410.03 ÷ (410.03 + 63.79) = 86.5% breakeven win rate.
It achieved 94.18%.
That gap is the entire edge. It is real, it is measurable, and it is what you should monitor. If the strike rate drifts into the mid-80s the equity curve flattens; below that, the strategy loses money regardless of how good the individual entries look.
Three ways to bank a winner. One dropdown.
- Trailing Stop — Virtual (default). The trail level is held inside the EA and enforced by closing at market. Nothing is written to the broker, which is what makes a trail this tight possible at all — a stop that close would sit inside most brokers' minimum stop distance and be rejected. The broker-side protective stop stays underneath it throughout. Requires the terminal to keep running; use a VPS.
- Trailing Stop — Broker. The same trail, written as a real stop loss order and advanced as price moves in your favour. Coarser, because every level must respect your broker's stops and freeze levels — in exchange, the exit survives a terminal disconnect.
- Fixed Take Profit. No trail. A real broker take profit placed with the order and refreshed as swap accrues. The most predictable and the least dependent on your terminal — but it caps every winner at the target.
All three aim at the same net dollar target, adjusted for accrued swap and for round-turn commission measured from your own deal history. The trail arms by money rather than distance — it does not exist until net profit reaches the target value on the traded lot — then tracks the best price reached and never moves backwards. An armed trail can only ever close in profit: Solitaire refuses any trail level at or worse than entry.
The setting scales with the stone
Stop loss, take profit, trailing distance and trailing step are set in basis points of the entry price, where 1 bp is 0.01%: price distance = entry price × bp ÷ 10000. Defaults are SL 100 bp and TP 15 bp — at gold near $4,100 that is a $41.00 stop and a $6.15 target; at $2,600 it is $26.00 and $3.90. A stop fixed in dollars gradually becomes a different fraction of the daily range as the metal moves from $2,000 to $4,000, and the strategy quietly changes character without anyone touching a setting. Budget for the consequence: at a fixed lot size, your dollar risk per trade rises as gold rises. One input switches every distance to absolute broker points, and Solitaire refuses to run if a value entered in one unit was obviously meant for the other.
Position sizing
- Fixed Lot Size (default) — always trades your set lot. Ships at 0.01.
- Fixed per Balance — one unit lot per set amount of balance. At the defaults, 0.01 lots per $1,000.
- Automatic — an eight-step ladder from Very Low (0.01 per $4,000) to Extreme (0.01 per $500). Ships at Low, which is 0.01 per $3,000.
A sizing note worth reading twice. The shipped default of a fixed 0.01 lots on the recommended $1,000 minimum deposit is the same exposure ratio as Medium-High on the ladder above — moderately aggressive on $1,000, conservative on $5,000. The backtest below was run at exactly this ratio, so its drawdown figures apply directly to a $1,000 account running the defaults. Two consecutive losses were the worst run in 18 months, and on a fresh account at that ratio it is close to a 10% dent. If that is more than you want, use Automatic / Low or halve the fixed lot.
Filters and controls
News filter (optional, ships disabled). ForexFactory weekly calendar as primary source, MetaTrader's built-in calendar as automatic fallback, plus 933 baked-in events covering 2020–2027 so filtered backtests are reproducible. Block high impact, medium impact, or both, with separate before and after windows; a keyword list promotes under-rated medium events such as FOMC, NFP and CPI, and overlapping windows resolve to the latest end time. Full feed coverage requires adding https://nfs.faireconomy.media/ to MT5's WebRequest whitelist — see the manual. The published backtest was run with the filter off.
Friday control. Block Friday entries entirely, or set a cut-off hour after which no new trade opens — default 12:00 server time, against thinning liquidity and weekend gap risk. Set −1 to trade the full session. Spread filter: entries refused above your maximum; 0 disables. Directional filter: buy only, sell only, or the default of both.
The control panel
Drawn as a single canvas bitmap rather than chart objects, so an indicator or trade manager attaching after Solitaire cannot bury it. It shows live and daily P&L, latched daily drawdown — the worst excursion below today's equity high, not how far below you happen to be right now — cumulative P&L, balance, equity, a full configuration readout, and a named reason whenever no trade is open. One-click close exits Solitaire's own position without removing the EA; your other trades are untouched. Ships off so backtests run faster.
Backtest results
Everything needed to reproduce this run is stated below. One broker, one symbol, one eighteen-month window, news filter disabled. It is a record of what happened, not a forecast of what will.
| Broker / server | Pepperstone-MT5-Live01, terminal build 6061 |
| Symbol / timeframe | XAUUSD, M1 |
| Period | 1 January 2025 – 30 June 2026 (18 months) |
| Modelling | Real ticks, 99% history quality — 155,225,571 ticks / 526,570 bars |
| Deposit / leverage | $10,000, 1:500 |
| Exit configuration | Trailing Stop — Virtual, TP 15 bp, SL 100 bp, trail 1.00 bp, step 0.15 bp |
| Filters | Max spread 60 points, slippage 60 points, Friday 12:00 cut-off, news OFF |
| Position size | Fixed 0.10 lots — same exposure ratio as the shipped default on $1,000 |
| Net profit | $27,713.86 (+277.1%) |
| Profit factor | 2.48 |
| Total trades | 773 (approximately 43 per month) |
| Win rate | 728 won / 45 lost — 94.18% |
| Average win / average loss | $63.79 / $410.03 |
| Largest loss | $573.15 |
| Longest losing run | 2 trades (−$958.60) |
| Maximum balance drawdown | $959.65 (2.83%) |
| Maximum equity drawdown | $1,294.60 (4.73%) |
The identical 773 trades on Automatic sizing at Medium-High returned $140,892.28 with a maximum equity drawdown of $12,321.41 (10.73%). That is not the headline, because the strategy did nothing different — compounding inflates return and drawdown together and flatters an eighteen-month window in a way that would not survive a bad start. Judge Solitaire on the fixed-lot result.
Trading Recommendations
| Symbol / timeframe | XAUUSD (suffixes handled automatically) — any timeframe |
| Minimum / recommended deposit | $1,000 at 0.01 lot — $2,000+ for a gentler exposure ratio |
| Account type | Raw / ECN spread with commission (IC Markets, Pepperstone, Vantage, BlackBull) |
| Leverage | 1:100 minimum — 1:500 recommended |
| VPS | Strongly recommended, required in practice for the virtual trail |
| Tester mode | Every Tick Based on Real Ticks |
What Solitaire is not
Solitaire is not a guarantee of profit. Some trades lose — and by design the losses are larger than the wins. Some days produce no trades at all, and gaps, slippage and spread widening can produce a fill worse than the intended stop. It will not suit you if you want high-frequency trading, a recovery basket after losses, a guaranteed number of trades per day, or an EA that never has a losing month. It will sometimes sit quiet for a day and a half, and that is not a fault. It is built for traders who would rather own one thing they can see clearly than a drawerful of things they cannot. The complete manual and default preset are published before purchase: https://www.mql5.com/en/blogs/post/773462
Three other high-quality EAs are available from Starpoint Trading: Impulse (six-strategy XAUUSD momentum grid), Antipode (Live Signal 12 Months) and Ducat.
DISCLAIMER: Trading foreign exchange and CFDs carries a high level of risk and is not suitable for everyone. Past performance, backtested or live, is not indicative of future results. A stop loss is an instruction, not a guarantee — gaps and slippage can produce a fill worse than the intended level, and weekend gaps can exceed it substantially. Execution differs between brokers, and spread, commission, swap and fill quality all change the result. Always test on a demo account before deploying live. Nothing in this listing is financial advice or takes account of your circumstances. Trade only with money you can afford to lose.
One stone. Set alone. Nothing behind it.
