← Captain Karat-Smash

Captain Karat-Smash — user manual

He waits patiently for the walls of consolidation to close in, and then BAM! Punches right through the resistance line to grab the gold. The rest of this manual explains exactly how he does it.

Version 1.11 · XAUUSD · attach to one H1 chart · hedging account

1. What it does

Captain Karat-Smash trades breakouts of intraday support and resistance on gold. It is not one strategy but seven independent ones running side by side on a single chart. They share one engine and differ in which timeframe they read, how strict their level definition is, and how their stop and target are proportioned.

Each strategy repeats the same cycle:

1. Find the level. It looks for the most recent unbroken swing high above price and swing low below it — a bar whose high (or low) is more extreme than the bars on either side of it, and which no closed bar has traded through since. Strategies 1 and 2 read daily bars; strategies 3 to 7 read four-hour bars.
2. Rest an order just inside it. A buy stop is placed slightly below the swing high, a sell stop slightly above the swing low. Price must carry through the level to trigger the order, which is what separates a break from a touch.
3. Bracket the fill. Every position gets a stop loss and a take profit the moment it opens, both set as a percentage of the entry price, so the geometry is the same whether gold trades at 1,800 or 4,300.
4. Manage the exit. Once the trade is far enough ahead, the stop moves to a locked level; from then on it trails behind each closed one-minute bar and never loosens. Two of the seven strategies also pull their target closer if the trade goes far enough against them, to exit on a bounce rather than at the full stop.
5. Expire unfilled orders. Every resting order has a lifetime, from 20 hours to 20 days depending on the strategy. When it expires it is re-placed at the same level if that level is still one of the two most recent unbroken swings, and dropped otherwise.

There is no grid, no martingale, no averaging down and no hedging of one position with another. Every position carries a hard stop loss from the moment it is opened. Losses are taken; they are not carried.

2. Installation

1. Copy CAPTAIN_KARAT-SMASH.ex5 into MQL5\Experts and restart the terminal. (An earlier build of this engine may still be present in some installs under a different file name; always install CAPTAIN_KARAT-SMASH.ex5 and check the Journal shows version 1.09 on its version line.)
2. Open one XAUUSD H1 chart and attach the EA. The chart timeframe is only a host — each strategy reads the timeframes it needs on its own.
3. Enable algorithmic trading.
4. Leave every input at its default. The compiled defaults are the validated configuration; no preset file is required. Set the Risk level and nothing else.

Requirements: a hedging account (the seven strategies hold independent positions), XAUUSD available with a normal contract size, and a stable connection or VPS. The EA prints a note if it is started on a netting account.

3. The only setting that matters

The figures below come from a Strategy Tester backtest, real ticks, BlackBull XAUUSD, 2022 to 2026.

The drawdown column is measured, not extrapolated, in a Strategy Tester backtest, BlackBull XAUUSD, 2022 to 2026: a balance of 4,000 behind each 0.01 lot produced a 14.8% peak-to-trough drawdown over 3.8 years of real ticks, the longest window tested and the only one containing a full flat year. Drawdown scales inversely with the balance figure, which is how the other levels are derived.

Position size scales with account balance, so the drawdown percentage stays roughly constant as the account grows or shrinks. The percentages come from a backtest on real broker ticks; plan for the future being worse than the past. A widely used rule of thumb among system traders is to assume the worst historical drawdown can double, and to stop and review a system if drawdown in practice reaches one and a half times its historical maximum.

Suggested minimum balance: 3,000 at Medium. The EA never refuses to trade on a small balance — it falls back to the broker's minimum lot — but below that figure a single minimum lot is a large share of the account and the drawdown percentages above no longer apply.

3f. What changed in version 1.11

• Trades identical to the previous version (proved deal for deal). It no longer sends an order or a stop change into a closed market (the broker's daily break, a holiday): the request the broker would have refused is skipped at the same point, so nothing about the trades changes and the journal no longer fills with 'Market closed' errors.

3e. What changed in version 1.10 - panel, chart labels, restart-safe

• Trades identical to 1.09 (proved deal for deal; see the MANIFEST). Nothing about when it enters or exits has changed.
• Faster backtests and less VPS load. A BlackBull backtest of 21 months (269 M real ticks) takes about 3 minutes instead of 17 (5.7x; Vantage 4.4x).
• A restart no longer changes a trade. Each position's furthest move for and against it, its lock stage and trend side, and each strategy's last swing are saved on the account and re-read after a restart. Before, a restart reset the London Grip and the bounce exit, and re-placed orders on swings whose orders had already filled - with a restart every hour, 1.09 traded 3,875 times instead of 1,148 at a 25 % instead of 5 % equity drawdown (backtest, restart arm, BlackBull, 2025-01 to 2026-09).
• Clearer journal. The start lines say CAPTAIN KARAT-SMASH, what the risk level buys at your balance (and when the broker's minimum lot sets the risk), what the smallest order needs in margin, and which of your inputs a prop account type overrides. When the broker's margin holds an order back, one line a day says so.
• A control panel and chart labels (below).

The panel and the chart (since 1.10)

A small panel in the top-left corner shows what the EA is doing, and the chart labels every level it waits at, every open trade's exit and every closed trade's result. Nothing is drawn in a plain Strategy Tester run; on a live chart and in the visual tester it refreshes every second. Inputs: InpPanel, InpDrawTrades, InpCleanChart (hides the grid while the EA runs and puts it back when it is removed), InpTextSize.

4 buttons. Every action is sent by the EA itself and written to the Experts tab; a panel close carries the comment "... panel close". The second-click buttons need the same button again within 5 seconds.

On the chart: each resting order as a dotted line at the swing level it rests on, labelled at the right edge with the strategies on it and their stop prices (buys above their line, sells below); each open trade's lock price, Grip start or Grip exit, or trailing stop as a line with its label. A closed trade shows as a dotted line from its entry to its exit, labelled with why it closed and its net result.

3d. What changed in version 1.09

London-session breaks are locked in; with-trend runners get more room. The London morning's first push is often the false move of the day: it runs, then fades once New York opens. From 1.09 a position opened in the London session (from 3:00 to 8:00 New York time) is protected by the Grip once it is half a stop in profit, and closed if it gives back half of its best move. Positions opened in any other session keep the trail. At the same time the trail on trades that go with the daily trend is a quarter wider (strategies 3, 4, 6 and 7); trades against the trend keep the 1.08 width.

Measured at 0.01 lot against 1.08 on four brokers' real ticks (January 2023 to September 2026):

The trade-off: the dips are shallower and shorter, but the ones that happen come faster - the deepest fall took about four weeks instead of several months. Proved before release: the 1.09 defaults place exactly the same deals as the tested configuration (208 trades, +671.63 on IC Markets real ticks, 1 May - 5 September 2026, where 1.08 made +346.60), and the prop account types carry it.

3c. What changed in version 1.08

The trail follows the daily trend. The wider 1.07 trail pays on trades that go WITH the daily trend and costs a little on trades AGAINST it: a counter-trend break tends to snap back, and the wider trail then gives back more. From 1.08 a position of strategies 3, 4, 6 or 7 that opens against the daily trend (the side of the previous day's close relative to its 20-day average) keeps the tighter pre-1.07 trail; with the trend it uses the 1.07 trail. Measured in a Strategy Tester backtest at 0.01 lot against 1.07 on four brokers' real ticks: net +3.9 % (BlackBull) / +3.5 % (OANDA) / +4.1 % (Fusion) / +11 % (IC Markets), and the worst five-day stretch at equal profit 5 to 12 % smaller on every one. Nothing else changes.

Proved before release: the 1.08 defaults place exactly the same deals as the tested trend-following configuration (208 trades, +346.60 on IC Markets real ticks, 1 May - 5 September 2026, where 1.07 made +254.31), and the prop account types carry it.

3b. What changed in version 1.07

A wider trailing stop. Strategies 3, 4, 6 and 7 now trail their stop twice as far behind price (for example strategy 3: 1.59 % of price instead of 0.80 %). The old trail was measured to close winners too early in quiet markets: measured in a Strategy Tester backtest across four brokers' real tick histories (BlackBull and OANDA 2023-2026, Fusion 2024-2026, IC Markets 2026) the wider trail earned more on every one - +9.7 % on BlackBull, +8.5 % on OANDA, +5.6 % on Fusion, +4.9 % on IC Markets, at 0.01 lot - and on BlackBull its deepest drawdown fell more slowly and was recovered in 28 trading days instead of 42. Trails wider still (x2.5 to x4) earned slightly more but lost faster and took about five times longer to recover, so they were not chosen. The entries, stops, targets and everything else are unchanged.

One screen of settings. The Inputs tab is now ordered 1 Start here (account type, risk level, a new virtual balance and the spread limit) / 2 Strategies / 3 Prop-firm rules / 4 Trading / 5 Custom sizing (chapter 4). The spread limit is now in USD (5.00): the old 500 points meant 5.00 USD on a two-decimal gold quote but only 0.50 USD on a three-decimal one. Both new inputs are off or unchanged at their defaults.

Proved before release: the 1.07 defaults place exactly the same deals as the tested x2.0 configuration (208 trades, +254.31, IC Markets real ticks, 1 May - 5 September 2026), the new inputs change nothing at their defaults, and the prop account types carry the new trail.

The yearly figures in chapter 5 were measured on version 1.05 (the older, tighter trail). They are being re-measured on 1.07; until then read them as the conservative case.

3a. Prop-firm accounts: one choice (since 1.06)

The first input, Account type (InpAccountMode), is all a prop-firm account needs:

Why strategies 4 and 6 are off in prop modes. They are the two that put the most loss into the worst days, measured in a Strategy Tester backtest. With both off, Karat's worst day on BlackBull 2023–2026 halved, from a 3.4 % to a 1.7 % loss of balance, and its net per unit of worst-week risk rose by a quarter (BlackBull) to two-fifths (Vantage). That is why the prop size can be LARGER than the personal default and still stay far inside a daily-loss limit.

What the measurements say — a simulation that replays the backtest's own days in random blocks (10,000 paths, the challenge rules applied at every trade close); a statement about tested history, not a prediction:

Read it as: the prop types never broke a daily-loss rule in these histories; the quiet stretches (2023–2024) only cost time, so a challenge without a time limit is the right kind. The simulation used the raw-spread feeds the product is built for; on a standard-spread account expect slower progress.

Proof that the prop types are exactly what they say: on Pepperstone 28 August – 19 September 2026, Prop firm 2-step placed the same 32 deals, to the cent, as version 1.05 with strategies 4 and 6 off at 3,667 per 0.01 lot (+261.37), and Prop firm 1-step the same 32 deals as 1.05 with strategies 4 and 6 off at Medium (+130.37). With Personal, 1.06 places the same 48 deals as 1.05 (+213.78).

4. All inputs

The Inputs tab is ordered by how often you should touch it (since 1.07).

1. Start here

2. Strategies on or off — see the table below. 3. Prop-firm rules — chapter 5a.

4. Trading — direction filter, magic number, comment, order cap, slippage (table further down).

5. Custom sizing (Risk level = Custom only)

Strategies — seven on/off switches, all on by default. Turning one off removes its trades and nothing else; the strategies do not interact.

Trading

5. What to expect

• Trade frequency. A few hundred trades per year across the seven strategies, mostly intraday. Median hold is well under an hour; some positions run for several hours.
• Win rate. The lock-then-trail exit closes many trades for a small gain rather than letting them reach the full target, which keeps the win rate above what the reward-to-risk ratio alone would suggest; the wider trail introduced in 1.07 lets more of them run than the original, tighter trail did.
• Losing streaks and flat periods. Normal and expected. A breakout system pays for its large moves with a steady stream of small losses; some months are flat or negative.
• Broker sensitivity. The system's edge per trade is measured in dollars, not cents, so it tolerates normal spread variation. It was tested on four different brokers' real tick histories, and the results differ between them — that difference is the honest range, not a defect.
• The returns are not evenly spread, and this is the most important line in this manual. The longest record tested, a Strategy Tester backtest on OANDA gold ticks reaching back to 2018 at Medium risk, earned slowly through the quiet years up to the end of 2024 - with stretches of up to two years without a new equity high - and earned most of its total in 2025 and 2026, while gold ran from about 2,658 to about 4,430.

This is a trend harvester, not a steady income system. It waits, pays rent in quiet and choppy markets, and earns when gold makes a real move. If a two-year flat stretch would make you switch it off, this is not the right system for you.

• Direction. Long and short trades are both enabled by default. In testing the long side was profitable in every period; the short side lost slightly from 2018 to 2024 and was strongly profitable in 2025 and 2026, while gold rose in all three periods. Over the full record the short side is net positive on both long-history brokers tested, which is why it ships enabled — but the switch is there if you hold a different view.

All figures in this manual come from the MetaTrader 5 Strategy Tester on real broker tick data. Backtest results do not represent trading on a live account and do not predict future performance.

5a. Prop-firm rules: your firm's own numbers (since 1.05)

The Account type (chapter 3a) chooses the size. The group right under it, Prop-firm rules, enforces your firm's limits. Choosing Prop firm 2-step or Prop firm 1-step already fills in the two loss limits for you (5 % daily and 10 % maximum, or 3 % and 6 %; both fixed, and the EA acts 0.5 % before each); type your firm's own numbers to override them. With Personal every rule is off until you type a number.

When a limit is about to be reached, the EA closes its own trades, deletes its own orders and stops opening new ones: until the firm's next day for the daily rules, for good for the maximum loss and the profit target. The Journal says which rule acted and at what equity:

CAPTAIN_KARAT-SMASH prop-firm rule: daily loss limit (equity 95512.40, day start 100020.00, firm limit 95020.00) - trades closed, orders deleted, paused until the firm's next day

and the start line lists what is on:

CAPTAIN_KARAT-SMASH prop-firm rules: account size 100000.00 | daily loss 5.00% of account size from the higher of day-start balance and equity - the EA acts at 4.50% of account size | maximum loss 10.00% (fixed) - the EA acts at 9.50% | ...

The inputs (group "3. Prop-firm rules (0 = rule off)")

Examples - what to type

From the LuxAlgo prop-firm directory on 23 September 2026 (the daily-loss wording is inferred by that directory; your firm's own rules page is always the authority, and firms change their rules):

All six count the daily loss as a share of the account size from the previous day's closing balance, open trades included. Leave counted from at the higher of balance and equity - it is never later than the firm's own meter.

Profit target: set it for the challenge phases only. When the target is reached the EA stops for good on that login; on the funded account set it back to 0.

The firm's day. Most firms restart the daily loss at midnight of the platform's own clock: leave 0. If your firm names another time, convert it to the Market Watch clock. Gold is closed for about an hour around the broker's midnight on most platforms, so for gold an hour's difference rarely matters; a market that trades through midnight (Bitcoin) needs the exact hour.

Restarts. The EA remembers the account size, the day's starting value, the highest balance and whether a rule stopped it, across restarts of the platform (terminal global variables named PR<login>_<magic>_...). A restart in the middle of the day does not give it a fresh daily allowance.

What these rules do not do

• They close this EA's trades only. The limits are measured on the whole account's equity, so trades from another EA or by hand count toward them; give every EA on the account the same numbers.
• The EA acts on the first price tick past its limit. A gap or a very fast market can carry the loss a little further before the closing trades fill - that is what the safety margin is for; do not set it to 0 on a real challenge.
• No news filter. Some funded accounts forbid opening or closing trades a few minutes around high-impact releases. The platform's economic calendar cannot be replayed in the Strategy Tester, so a news rule could not be proved the way every rule above was; pause the EA by hand around those releases if your firm requires it.
• No minimum trade duration and no minimum number of trading days.

Proof. With every rule off, this version placed the same deals as the version before it, deal for deal (32 trades, net +261.37, Pepperstone real ticks, 28 August - 19 September 2026). Each rule was then switched on at a level the same weeks reach; at all 8 triggers the EA closed its trades, opened nothing until the pause ended, and placed exactly the same deals as with the rules off up to the first trigger.

6. Troubleshooting

The EA prints a one-line census when it stops, for example:

Captain Karat-Smash: orders placed 632 | skipped - no level 2, spread 0, price inside stop level 1,
stop distance 0, lot 0, margin 0, send failed 0

No trades at all on a fresh chart is normal for the first hours: every strategy waits for a new confirmed swing before it places anything, and the daily strategies can wait days.

Complete reference — every input, its set-file key and default

Generated from the compiled source; enum defaults are the source names — the Inputs tab shows their plain labels.