Safe Grid Settings for Gold: Basket Stop and Lot Caps

You open a XAUUSD grid on Monday morning. Price chops sideways for two hours, the grid racks up a handful of small wins, and everything looks fine. Then London opens, gold drops $25 in twenty minutes, and suddenly you have eight positions underwater with no exit plan. The account is bleeding, and you’re staring at the screen wondering where the stop was supposed to be.

If that scene feels familiar, the problem isn’t your grid logic. It’s the settings that are supposed to protect you when the grid goes wrong. Safe grid trading for gold isn’t about picking better entry levels. It’s about deciding, before you click buy, exactly how much pain the account can take and what happens when gold decides to trend.

Here are the settings that matter most.

1. The Basket Stop Is the Only Stop That Matters

Most grid traders set individual stop losses on each position. That’s a mistake. Individual stops on a grid defeat the purpose of averaging, because the first stop-out strands you with a partial basket and a realized loss you now need to recover from scratch.

The basket stop works differently. You define a maximum floating loss for the entire group of positions—say, 5% of account equity. As long as the basket’s combined loss stays under that number, the grid runs normally. Once the basket loss hits the threshold, everything closes at once. No negotiation, no waiting for a bounce.

The MQL5 forum discussion on gold grid risk control describes a nine-layer system where the most critical layer is an equity circuit breaker that triggers a full close when drawdown reaches 30% and cannot be disabled from the client side . The reasoning is blunt and honest: “把止损的决定权交给一个会犹豫的人，等于没有风控” — give the stop-loss decision to someone who will hesitate, and you have no risk control at all .

For gold specifically, a basket stop of 15–25% of equity is a reasonable starting range. Gold regularly moves $30–$50 in a session, and a grid without a basket stop is just a martingale with extra steps. The exact number matters less than the fact that it exists and that you don’t override it when the drawdown starts feeling uncomfortable.

This is where a tool like Camovia Tray becomes practical. If you’re running a gold grid and want to keep an eye on floating exposure without staying glued to the MT5 window, the tray hover display shows open positions and current P&L. You can see the basket loss climbing in real time, from the system tray, before it becomes an emergency.

2. Lot Caps and Progression: The Difference Between a Grid and a Time Bomb

The second setting that separates a survivable gold grid from an account killer is how lot sizes grow as the grid deepens.

The default approach in many grid EAs is geometric: double the lot size with every new level. A GitHub gold grid EA for XAUUSD uses a default LotSize of 0.1 and allows up to 20 concurrent orders, but the risk warning is explicit: “Grid trading carries significant risks… no stop loss protection… potential for large drawdowns during strong trending markets” . Twenty orders with doubling lots after the tenth level is no longer a grid. It’s a countdown.

A safer gold grid settings template looks more like this:

Fixed or gently progressive lots. Start with 0.01 for the first level, 0.01 for the second, 0.02 for the third, and so on. The progression is linear, not exponential. The deepest level might be 5–10x the first lot, not 500x.

A hard maximum lot. No single grid position should exceed a predetermined size, regardless of how many levels the grid has added. One MT5 basket manager EA built for XAUUSD scalping includes a “maximum lot” parameter as a built-in safety check, alongside basket-level stop-loss management . The cap prevents a situation where one “recovery” order at the bottom of a deep grid carries more risk than the original position.

A maximum number of levels per side. This is your lot cap in disguise. If you allow 30 levels, the math of lot progression will eventually produce a position size that can wipe the account. Cap the grid at 5–8 levels per direction. If gold has moved that far without pulling back, the grid thesis is wrong, and the basket stop should be doing its job anyway.

The relationship between grid step and lot cap is also worth checking. A $2 grid step with 0.01 lots and a $500 account is different from a $2 step with 0.05 lots and the same account. The xauusd grid parameters that matter are not just the step and the lot size individually, but the ratio between them and the account equity.

3. Grid Step and Volatility: Don’t Use a Fixed Step on a Moving Market

One of the more common mistakes in gold grid settings is using a fixed dollar step regardless of market conditions. Gold’s volatility is not constant. The Asian session might see $1–$2 ranges; the London open can produce $10 swings in minutes.

A fixed $2 step will trigger almost every level during a volatile session, loading the basket with positions before the real move even starts. A fixed $10 step might never trigger during a quiet Asian range, leaving the grid idle.

The better approach is to let the step adapt to volatility. Several XAUUSD grid EAs use ATR-based spacing precisely for this reason . When ATR is elevated, the step widens; when volatility compresses, the step tightens. This keeps the grid from over-firing during spikes and from sitting dormant during quiet periods.

A practical middle ground: set a minimum step of $3–$5 for gold, and let an ATR multiplier adjust upward from there. The MQL5 forum post referenced earlier uses grid steps of 1.5 and 2.0 price units for its two channels, but that system is running on a specific gold price range with a calibrated lot progression . For most traders, wider steps are safer because they reduce the number of positions the basket carries at any given time.

Grid spacing also connects directly to the basket stop. A tighter grid triggers more levels, which means a larger basket exposure, which means the basket stop needs to be tighter too—or the lot sizes need to be smaller. These three settings are not independent. They form a system.

4. What to Check Before the Grid Goes Live

You can have the best basket stop and the most conservative lot progression in the world, and still get caught by something you didn’t check.

Spread at your broker. Gold spreads vary wildly between brokers and between sessions. A grid that triggers on every tick during a 50-point spread environment will accumulate losses before price moves at all. The MT5 basket manager EA mentioned earlier includes spread filtering as an execution safety feature . If your grid opens positions during wide spreads, the math is against you from the first fill.

Margin headroom. A grid with a basket stop at 20% drawdown still needs enough free margin to survive the path to that drawdown. If you’re using 80% of available margin on the first few grid levels, the basket stop may trigger as a margin call before it triggers as a voluntary exit. The cent account recommendation seen in multiple grid EA listings exists for this reason: it multiplies the nominal equity buffer without changing the trading logic .

The terminal connection. A grid strategy that relies on EA-side basket management requires the terminal to be running and connected. If MT5 loses connection during a news event, the basket stop is not watching . This is one of those invisible risks that doesn’t show up in backtests but shows up in live accounts at the worst possible time.

For gold grid traders who want to reduce the “terminal always open” requirement, Camovia Tray offers a middle path. The tray hover display shows live quotes and open position P&L without pulling up the full MT5 window. You still need the terminal running for the grid EA to function, but you don’t need to keep the chart visible on a second monitor to know whether the basket is in trouble. The position management view lets you see floating exposure and, if needed, close positions from the tray.

That last point is worth sitting with. A grid’s worst moments happen fast. Being able to see the basket loss without switching windows doesn’t change the math, but it changes your reaction time.

The Short Version

Safe grid settings for gold come down to three numbers: the basket stop, the lot cap, and the grid step. The basket stop defines your maximum loss. The lot cap defines how deep the grid can go. The grid step defines how often it fires.

Get those three right, and the grid is a tool. Get them wrong, and the grid is a bet.

The basket stop is the one that matters most. It’s also the one most traders either don’t set or override when the drawdown starts. If you’re running a gold grid and you don’t have a basket-level exit, that’s the first thing to add—not a better entry, not a tighter step, not a new EA. The exit is the strategy.