Grid trading has a reputation problem.
Search for it online and you'll find two camps. Camp one says it's a guaranteed money printer. Camp two says it's a guaranteed account killer. Both are wrong, and the reason both exist is the same: most people confuse grid trading with Martingale.
Martingale doubles your position after every loss. One bad run and your account is gone. Grid trading doesn't do that. A properly built grid uses consistent lot sizes across levels, defines a maximum exposure from the start, and makes money from price oscillation, not from praying that a losing streak reverses.
So why does the confusion exist? Because a lot of so-called "grid EAs" on the market are actually Martingale EAs wearing a grid costume. They layer positions with increasing lot sizes, call it "grid trading," and hand you a smooth equity curve that hides a 60% drawdown waiting to happen. We've written about what those curves actually hide.
But here's the thing that gets lost in the noise: grid trading is one of the oldest surviving automated strategies in retail forex. People have been running grid EAs profitably for 15+ years. Not because it's easy money, but because range-bound conditions are more common than sustained trends in most forex pairs, and grid strategies are designed specifically for that environment.
That's the real reason grid trading won't die. Markets range more than they trend. And grid EAs are range-harvesting machines.
This guide is about using them correctly. When to turn them on, when to turn them off, what parameters matter, and what to watch so you're not the person writing "my account blew up" on a forum.
How grid trading EAs work — 30 seconds
You pick a price range. You place buy orders below the current price at fixed intervals, and sell orders above it. When price drops, buy orders fill. When it bounces back up, those positions close in profit. When price rises, sell orders fill. When it pulls back, those close in profit.
Price goes up and down, you make money on every swing. That's the core mechanism.
Think of it as fishing nets laid across a river at even intervals. Fish swim back and forth, nets catch them. You don't need to predict which direction the fish go. You just need them to keep moving.
The grid has a few fixed parts:
- Spacing: the distance between orders (e.g., 25 pips apart)
- Levels: how many orders on each side (e.g., 6 levels up, 6 down)
- Lot size: how large each order is
- Take profit: the target for each individual position (often equal to the spacing)
That's the whole strategy. The complexity is in the details.
Best forex pairs and conditions for grid trading
Grid trading is a range strategy. It prints money in sideways markets and bleeds in trends. Knowing when you're in which one is the entire game.
Where it works:
- Range-bound pairs that oscillate between support and resistance
- Low-volatility sessions (Asian session, European morning before data releases)
- Pairs with natural mean-reversion tendencies: AUDCAD, AUDNZD, EURGBP, AUDCHF
- Also works on EURUSD and GBPUSD during their frequent consolidation phases
Where it kills:
- Trending markets where price moves 300 pips in one direction without a meaningful pullback
- Around major news (NFP, rate decisions, geopolitical events)
- High-volatility pairs during active sessions
- Any time ADX is above 28-30
From testing grid EAs at FXTool, we've seen a consistent pattern: about 80% of the losses happen during trending periods that traders failed to avoid. The grid itself wasn't broken. The trader left it running when they shouldn't have.
Best pairs for grid strategies: GBPUSD, EURUSD, AUDCAD, and AUDNZD. These four have the right mix of liquidity, manageable spreads, and enough oscillation to keep a grid profitable. AUDCAD and AUDNZD in particular tend to range for extended periods due to the correlated nature of the Australian, Canadian, and New Zealand economies.
Grid EA parameters: the 5 settings that matter
Five settings determine whether your grid makes money or blows up:
| Parameter | What it controls | Setting principle |
|---|---|---|
| Grid spacing | Distance between orders | Match to volatility. Use ATR(14) as a baseline: spacing ≈ 1-1.5× ATR. Too tight = lots of fills but high exposure. Too wide = fewer trades, slower profits |
| Levels per side | Max orders in one direction | This × spacing = your maximum adverse move. 6 levels × 350-point spacing = 2,100 points of coverage |
| Lot size | Position size per order | Conservative: 0.01 lots per $1,000 of capital for the first level |
| Lot multiplier | How size increases on deeper levels | Fixed (1×) is safest. Graduated increase (1 → 1.5 → 2×) boosts recovery speed but increases tail risk |
| Equity stop | Kill switch for the whole grid | 10-15% of account. Mandatory. No exceptions. A grid without an equity stop is a time bomb |
The graduated lot multiplier deserves explanation. In EternalEngine EA, each level's lot size equals the previous level multiplied by a tier multiplier: 1× for levels 1-2, 1.5× for levels 3-5, 2× from level 6. Because each multiplier compounds on the previous level's lot, the actual sizes grow faster than the multiplier alone:
- Level 1: 1× → Level 2: 1× → Level 3: 1.5× → Level 4: 2.25× → Level 5: 3.38× → Level 6: 6.75×
Compare that to Martingale (1→2→4→8→16→32×). At level 6, you're at 6.75× your base lot vs Martingale's 32×. Still aggressive, but nearly 5× less exposure.
How much capital does a grid trading EA need?
This is where most people get it wrong. They see "0.01 lots" and think $200 is enough. It isn't.
Here's the math. EURUSD, 6 levels, 350-point spacing (35 pips), graduated lot sizing compounding from 0.01 base:
Worst-case floating loss when all 6 buy levels are filled (price at level 6 entry):
- Level 1 (0.01 lot): 175 pips underwater = -$17.50
- Level 2 (0.01 lot): 140 pips = -$14.00
- Level 3 (0.015 lot): 105 pips = -$15.75
- Level 4 (0.023 lot): 70 pips = -$15.75
- Level 5 (0.034 lot): 35 pips = -$11.81
- Level 6 (0.068 lot): just entered = $0
- Total floating loss: ~$75 per 0.01 base lot
Add a 2× buffer for overshoot: ~$150 per 0.01 base lot. EternalEngine's default auto-lot setting is 0.01 per $250 equity, which builds in that margin.
The rule of thumb: $1,000 recommended for a standard account running multiple pairs. For a single pair, $500 works.
Short on capital? You have options:
- Cent account: deposit $10, your account shows $1,000 cents. Lot sizes and P&L scale proportionally. The grid runs identically, but a blowup costs you $10 instead of $1,000. Great for learning and testing parameters with real market conditions.
- Micro account (10¢ per pip): deposit $100, similar concept. Smaller than standard but bigger than cent.
Both give you the real grid experience without needing $1,000 upfront.
For more on sizing your risk correctly, see our forex risk management guide.
Pure grid vs directional grid
There are two ways to run a grid.
Pure grid places orders in both directions. Price goes up, sells fill. Price goes down, buys fill. Works well in clean ranges. Problem: in a trend, one side accumulates large floating losses while the other side keeps closing small profits. The net effect is a slow bleed.
Directional grid only places orders in one direction. If you're bullish, you only run buy orders below the current price. If bearish, only sells above. You need a way to determine direction, but your worst case is cut in half.
In practice, the directional approach is safer. EternalEngine uses Bollinger Bands and RSI to assess market conditions and filter trade direction. It doesn't blindly grid in both directions. It picks its spots.
The smarter variation: multi-pair directional grid. Instead of running a deep grid on one pair, you run shallower grids across GBPUSD, EURUSD, AUDCAD, and AUDNZD simultaneously. The pairs don't all trend in the same direction at the same time, so diversification reduces the chance of all grids being underwater at once.
Grid trading EA results: 4 pairs, 3 years, 300% growth
Talk is cheap. Here's what a properly configured grid EA actually produces.
We ran EternalEngine EA across GBPUSD, EURUSD, AUDCAD, and AUDNZD on a 3-year backtest (2022-2025). Starting balance $1,000, 1:500 leverage, $7/lot commission, every tick based on real ticks modeling (MT5's highest quality).
Results:
- Growth: ~300% over 3 years, max drawdown ~25-30%
- Take profit: 60 points per position
- Grid spacing: 350 points
- Lot multiplier: 1 → 1.5 → 2
- Max orders per direction: 10
- Floating loss pause: 10% of equity
We're not going to pretend the equity curve was a straight line. There were drawdown periods. Some lasted weeks. The 300% came from consistent small wins in ranging periods compounding over time, not from one lucky run.
You don't have to take our word for it. EternalEngine supports full backtesting. Download the 7-day free trial, load it on the four pairs, set the strategy tester to 3 years, and verify the numbers yourself. We actively encourage this. If the backtest doesn't convince you, don't buy it.
What to look for in the backtest: not the final profit number, but the maximum drawdown and recovery time. A grid that makes 300% but has a 50% drawdown along the way might not suit your risk tolerance. That's fine. Better to know before you run it live.
For more on reading backtest results honestly, see our backtest vs live trading gap guide.
Grid EA backtest traps to watch for
Since we're talking about backtests, here are the traps:
Grid EAs almost always look good in backtests. Why? Because historical data contains lots of ranging periods. The grid harvests those ranges, the equity curve climbs, and by the end of a 5-year backtest you see impressive returns.
What to check:
Does the backtest include crisis periods? March 2020, September 2022, any period with sustained one-direction moves. If the backtest conveniently starts after or ends before those events, be suspicious.
What was the maximum floating drawdown, not just the closed-trade drawdown? Some backtest reports only show drawdown from closed trades. A grid can have 30% floating loss that never appears in that metric because the positions eventually recovered. "Eventually" can mean months.
Was the spread realistic? Grid EAs are sensitive to spread because they close lots of small profits. A backtest with 0.5-pip spread on EURUSD will look much better than reality where spread during Asian session might be 1.5 pips.
Our overfitting and curve fitting guide covers more on why backtests mislead and how to test more honestly.
When to shut it down
Knowing when to stop is more important than knowing when to start.
ADX crosses 28-30. Trend is forming. Pause new orders. If you're already in a grid, tighten your equity stop.
30 minutes before major news. NFP, FOMC, ECB rate decisions. Even if your grid is profitable, the volatility spike can fill multiple levels in seconds. EternalEngine includes a max spread filter (100 points default) that acts as a natural news gate, but manual caution is still smart. See our news filter setup guide.
Floating loss hits your threshold. EternalEngine pauses new entries at 10% floating loss and has an optional full close at a configurable percentage. Whatever EA you use, this feature is non-negotiable.
Friday afternoon. Weekend gaps happen. A gap through your grid can be catastrophic. Close or reduce before market close.
The most expensive lesson in grid trading: "it'll come back." Sometimes it does. Sometimes GBPUSD drops 800 pips in two weeks and your 6-level grid is 400 pips underwater with no sign of reversal. The equity stop exists for this exact moment.
Who should use this, who shouldn't
Grid trading fits you if:
- You have $1,000+ in capital (per pair)
- You can tolerate 15-20% floating drawdown without panicking
- You understand that some months will be flat or negative
- You're willing to monitor and turn off the grid when conditions change
- You want steady accumulation, not home runs
Grid trading doesn't fit you if:
- Your capital is under $500
- You're using a prop firm account (their 5-10% drawdown rules will trigger during normal grid operation)
- You want "set and forget" with zero monitoring
- You expect every month to be profitable
- You're not willing to learn when to turn it off
If you're not sure whether an EA approach is right for you at all, start with what an EA actually is and how to choose one.
The bottom line
Grid trading isn't a scam and it isn't a money printer. It's a range-harvesting strategy that works well in the market condition that occurs most often. The people who fail at it either confuse it with Martingale, run it with too little capital, or leave it running into a trend.
The strategy has survived 15+ years of retail forex because the underlying logic is sound: markets oscillate, grids capture oscillation.
Whether it works for you depends on whether you respect its limits.
Want to test it yourself? EternalEngine EA runs on GBPUSD, EURUSD, AUDCAD, and AUDNZD with built-in risk controls. 7-day free trial, full backtesting supported. Run the 3-year backtest, check the drawdown, decide for yourself.
Not sure grid is your style? Browse our full EA collection or read about other strategy types.
Frequently asked questions about grid trading EAs
Is a grid trading EA good for beginners? If you understand what floating drawdown means and you're okay watching your account go red for days before recovering, yes. If the word "drawdown" is new to you, start with our forex beginner guide first.
How much money do I need to run a grid EA? $1,000 recommended for a standard account across multiple pairs. Don't have that? A cent account lets you start with $10 (shows as $1,000 cents), and a micro account works from $100. Same grid, same logic, scaled-down risk.
Does grid trading work in trending markets? No. Grid trading bleeds in sustained trends. You need a way to detect trends (ADX above 28-30) and pause the grid. Leaving it running into a trend is the number one cause of grid blowups.
What's the difference between grid trading and Martingale? Classic Martingale doubles the lot size after every loss: 1→2→4→8→16→32×. A grid can use fixed lots across all levels, which is much safer. EternalEngine uses a graduated approach — tier multipliers of 1×, 1.5×, and 2× that compound level to level: 1→1→1.5→2.25→3.38→6.75×. At level 6, that's 6.75× your base lot vs Martingale's 32×.
Can I run a grid EA on a prop firm account? Not recommended. Most prop firms have 5-10% maximum drawdown rules. A grid EA in normal operation routinely hits 15-20% floating drawdown before recovering. You'll get cut before the strategy has time to work.
What are the best pairs for grid trading? Pairs that range frequently with good liquidity: GBPUSD, EURUSD, AUDCAD, and AUDNZD. The AUD crosses are especially suited because Australia, Canada, and New Zealand have correlated economies, keeping their exchange rates in extended ranges.
This article uses FXTool's EternalEngine EA to illustrate grid strategy mechanics. Nothing here constitutes investment advice. Trading involves risk, and past backtest performance does not guarantee future results.
About the author: The FXTool team builds and tests MetaTrader trading tools daily. We run every EA we sell on live accounts and publish the results. This guide reflects what we've learned from building 50+ EAs and working with thousands of retail traders.