The Zone Recovery Strategy Explained: How the Hedging EA Works, the Maths Behind It and Where It Breaks

What zone recovery is, in one paragraph

Zone recovery is a hedging method. You open a trade without a stop loss. Instead of exiting at a loss, you define a zone, a fixed distance in price, and place an opposite order at its far edge. If price reaches that edge, the opposite trade opens and it is larger than the first. If price then reverses across the zone, a third trade opens in the original direction, larger again, and so on. All the trades share the same two take profit levels, one above the zone and one below it. Because the lots grow with each switch, the first time price travels far enough in either direction, the whole set closes together with a small profit. The method is also called zone recovery hedging, the zone recovery algorithm and, in some sales material, a never-lose system. The last name is the problem this guide addresses.

The vocabulary you need

Points and pips differ. On a five-digit EURUSD quote, one pip is ten points, and one point on one standard lot is worth about 1 dollar. So 0.01 lots earns about 1 cent per point, and a 300 point target on 0.01 lots is about 3 dollars, which is roughly 2.60 euros at a rate of 1.15. Keep that small number in mind. The whole machine is built to earn it.

How a cycle works, step by step

1. 1

   Step 1: Leg 1: open the initial trade

   The EA opens a buy at the top of the zone with a take profit 300 points above it and no stop loss. In many EAs the entry comes from an indicator or a manual click.

2. 2

   Step 2: Place the hedge order

   A pending sell stop is placed at the bottom of the zone, 200 points below the entry. It acts instead of a stop loss.

3. 3

   Step 3: If price rises to the take profit

   The buy closes with its set profit and the pending sell is cancelled. The cycle is over. This is the good case.

4. 4

   Step 4: If price falls to the bottom of the zone

   The sell stop triggers and leg 2 opens, larger than leg 1. A new buy stop is placed at the top of the zone as the next hedge.

5. 5

   Step 5: If price rises back to the top

   Leg 3, a larger buy, opens. A new sell stop is placed at the bottom. The buys and sells now overlap inside the zone.

6. 6

   Step 6: Repeat until a take profit is hit

   When price finally moves 300 points beyond either edge, the take profit of the side that is ahead is reached. Because that side holds more lots than the losing side loses, the combined set closes in profit and the cycle ends.

Look at the shape of the diagram. Inside the shaded zone the account is locked: every extra buy is offset by a sell. The profit or loss is frozen at a number set by the zone width and the lots, and it does not change while price stays inside the zone, apart from swap and costs. The only things that move the result are price leaving the zone, or the trader closing the set early.

The lot-size formula

The core of the method is how big each new leg must be. The rule is simple to state: size the new leg so that, if price reaches the take profit in its direction, the profit from all the trades on that side exceeds the loss on the other side by the set profit. Write Z for the zone width and T for the take profit distance, both in points, and S for the total lots on each side. When a new leg opens at one edge of the zone, price must travel T points beyond that edge for its side to reach take profit, while the trades of the other side, which were opened at the opposite edge, would then be Z plus T points in loss.

With the invented numbers Z = 200, T = 300 and a set profit of 300 points on 0.01 lots, the first hedge is ((500 × 0.01) + 3) / 300 = 0.0267 lots, which rounds up to 0.03. The next buy is ((500 × 0.03) + 3) / 300 − 0.01 = 0.05. The pattern continues, and each size depends on everything already open. Some EAs use a simpler multiplier, such as 1.6 or 2.0 times the previous leg, instead of solving the equation. The result is similar: exponential growth. Compare it with doubling after each loss, which is described in our guide to martingale and grid EAs. Zone recovery grows more slowly than doubling, about 1.65 times per leg in this example, but it is the same family of method.

A ten-leg worked example in euros

The table applies the formula to EURUSD with a 200 point zone, a 300 point take profit and an initial lot of 0.01. Losses are in euros at a rate of 1.15, margin is at the 30 to 1 retail leverage limit, and the floating result is what the set would lose if you closed everything the moment the leg opened, before spread. All figures are arithmetic, not a forecast.

Read the table in two directions. Down the lot size column, the legs grow from a cent a point to nearly two lots, a 184-fold increase, to chase a 2.60 euro target. Across the last two columns, the margin required rises from 33 euros to over 15,000, while the floating loss at the moment of each hedge is modest: under 300 euros even at leg 10. That asymmetry is why the strategy feels safe for a long time. The loss that is locked in at any moment is small compared with the exposure that has to be held to get out of it.

What the end of a cycle looks like

There are three ways a cycle ends. The first is the one the sales pages show: price leaves the zone, hits the take profit of the larger side, and everything closes with the set profit. The second is a manual or automatic exit: you, or a drawdown rule in the EA, close the set at a loss that is larger than the profit target. The third is the broker closing it for you, because margin fell below the close-out level. The chart shows the first and the third on the same rules.

The strategy is therefore not a way to avoid losses. It is a way to postpone a loss, and to enlarge it, in exchange for a high frequency of small wins. That trade is attractive in a backtest, where the tail event may not appear, and it is painful in live trading, where it eventually does.

How likely is each leg? A simple probability model

A rough model helps. Assume price moves like a driftless random walk, with no trend and no costs. At each edge of the zone, the chance that price travels T points further before it travels Z points back across the zone is Z divided by Z plus T. With a 200 point zone and a 300 point take profit that is 200 / 500 = 0.4. The chance of reversing across the zone is the other 0.6. So in this model the cycle ends at the first take profit with 40 percent probability, and each further crossing happens with 60 percent probability.

Two readings follow. First, most cycles end early. About 64 percent end by leg 2, and 79 percent by leg 3. A backtest or a short live run will show many small wins, which is the attractive part. Second, deep legs are rare but not rare enough. One cycle in 13 reaches leg 6, and one in a hundred reaches leg 10. If the EA starts a new cycle every day, a leg 6 cycle appears about every two or three weeks, and a leg 10 cycle about every five months. Real prices are not driftless. Trends, news and volatility clustering make long runs of one-way movement and repeated reversals both more likely than the model says, which is why the real tail matters more than any table.

The expected result is close to zero before costs

Now put money on the probabilities. Suppose you cap the EA at N legs and close the whole set at a loss if price crosses the zone one more time. A cycle that ends in a take profit earns about 3 dollars. A cycle that hits the cap loses the floating loss at that moment. The table gives the loss and the expected value per cycle in the random-walk model, before costs.

The numbers show the central fact about the strategy. The growth in the loss at the cap, roughly 1.65 times per leg, is almost exactly cancelled by the fall in the chance of reaching it, 0.6 per leg. The product stays near one. Adding legs buys you a higher win rate and a bigger loss in the same proportion, so the expectation does not improve, it only moves from frequent small outcomes to rare large ones. That is the signature of a martingale-type system. Before costs the edge is zero by construction. After costs it is negative, and the next section shows how fast.

Costs: why a winning cycle can lose money

Every leg pays spread when it opens, and many accounts also pay commission. The lots grow, so the cost grows with them. The target stays at about 3 dollars. Take an all-in cost of 1 pip per lot, which is 10 dollars per lot, close to a standard account with a modest spread. Add up the lots open when the cycle ends and compare.

From leg 5 onward, a cycle that reaches its take profit still loses money after costs. With 1 pip of cost, any cycle that runs past leg 4 is paid at a loss after costs, even though it reaches its take profit, and the deeper ones are far worse. With a raw-spread account the cost per lot might be 0.3 pips of spread plus commission, but the same logic holds: costs scale with lots while the target does not. Slippage on stop orders, swap on positions held through rollover and wider spreads during news add to it. See spread, slippage and swap and weekend gaps and rollover.

The margin wall: what your account can actually carry

The deeper the cycle, the more margin it holds. A retail account in the European Union can use at most 30 to 1 leverage on major pairs, and the broker must close positions when margin falls to 50 percent of the required level, as covered in EA trading in Europe. The chart shows the balance you need at each leg to stay clear of that close-out, counting the margin and the floating loss.

Notice what the table says about the usual advice to start small. An initial lot of 0.01 looks cautious, but the chain it triggers is sized for a 300 point target, not for your account. A 1,000 euro account that starts a cycle every day meets a leg 7 event in under five percent of cycles, which is about once a month. When that happens the loss is large relative to the account, and years of 2.60 euro wins may not repair it. The margin calculator and the account sizing guide let you test your own numbers.

Why real markets are harder on it than the model

• Trends do not cooperate. A strong trend can sweep through the zone and past the take profit repeatedly, but it can also produce several crossings in a day when volatility is high, and each crossing opens a larger leg.
• News gaps. A release can move price several times the zone width in seconds. Stop orders fill beyond their price, so the hedge opens late and at a worse level, and the set no longer closes at the planned profit.
• Spread spikes. A wide spread at the moment of a hedge can trigger an extra leg that a clean price chart would not, or fill it far from the intended level.
• Broker rules. Some brokers restrict hedging, limit the number of open positions or reject orders that are too close to the market. An EA that depends on a long chain of hedges depends on all of these working.
• Gaps over the weekend. A set left open over a weekend is exposed to a gap with no way to hedge. See the weekend gaps guide.
• Netting accounts. On a netting account the opposite trades merge instead of coexisting, so the strategy does not work as designed. See hedging versus netting.
• Volatility changes. A zone that was right for a quiet week is too narrow for a volatile one, so crossings come faster and legs stack up sooner.

Common variants and what they change

Commercial zone recovery EAs add features around the core. Understanding what each does helps you read a product page without taking it at face value.

If an EA says that it never needs a stop loss, ask what its loss-taking policy is. A system with no cap has an unbounded tail. A system with a cap has a bounded one, and the cap level, in euros, is the single most important number on the product page. The vendor should state it, and you should confirm it in the backtest and in the EA's inputs.

When, if ever, does it make sense?

Zone recovery is not useless. It converts a stream of small, frequent wins into an occasional large loss, which can suit a trader who understands the trade and sizes it so that the large loss is acceptable. It works best in markets that oscillate within a range or reverse often, where cycles end early. It struggles in strong trends and around news. The sensible way to use it, if you use it at all, is with a hard leg cap, a loss limit sized to a small fraction of the account, and an initial lot that makes the capped loss tolerable.

The cap loss is the floating result at the moment the next leg would have opened, so it is the cost of saying stop. A trader who wants to risk about 1 percent per cycle on a 2,000 euro account can run a cap near leg 4 in this example. At that cap the strategy is far less attractive: it wins about 87 percent of cycles in the random-walk model, with tiny gains and tiny costs, and the expected result is still negative after spread. That is the honest picture. The high win rate that sales pages quote is bought with tail risk, and removing the tail removes the appeal.

Prop firm challenges add another reason for caution. Daily and overall loss limits can be breached by one deep cycle, and some firms restrict hedging and grid methods. Read using EAs with prop firms, EAs not suitable for prop firms and the comparison in prop firm versus your own account.

How to evaluate a zone recovery EA before you buy

A zone recovery EA is easy to make look good and hard to judge. The backtest that matters is the one that shows the worst cycle, not the average. Use this list on any product.

Be especially wary of a record that shows only a smooth rising balance. Zone recovery EAs produce smooth balance curves by design, because most cycles end in a small win, while the equity curve underneath may show deep floating losses. Compare the two on the account page, and see how backtests can mislead.

How to test one properly

1. 1

   Step 1: Backtest on real ticks with your broker's costs

   Set the leverage to the real figure, 30 to 1 on EU retail major pairs, and the deposit to what you will trade. See the Strategy Tester settings guide.

2. 2

   Step 2: Find the deepest cycle in the report

   Look at the largest position, the maximum drawdown by equity and the longest losing sequence. If the maximum margin used in the test exceeds your account, the test would not have survived on it.

3. 3

   Step 3: Stress the settings

   Widen the zone, narrow it, change the multiplier and see how the worst cycle changes. A system that only works at one setting is fitted. See walk-forward analysis.

4. 4

   Step 4: Add costs and delay

   Include commission and a realistic execution delay. Compare the result with the tester's ideal run.

5. 5

   Step 5: Run it on a demo through a news week

   Include a high-impact event and a weekend. Watch how many legs open and what the equity does. See forward testing on demo.

6. 6

   Step 6: Set the cap and loss limit before going live

   Use the smallest lot the broker allows and a cap whose loss is a small share of the account. Write the stop rule down. See when to turn off your EA.

Zone recovery compared with grid, martingale and a plain stop loss

The pattern is clear. The methods with high win rates are the ones that postpone and enlarge losses. The method with a stop loss has a lower win rate, a smaller tail and a visible cost. None of them removes risk. For the wider comparison see EA strategy types compared.

Five claims worth checking

1. Price must eventually reach a take profit. True of a price with unlimited time, but your account has limited margin, and the broker closes you out first.
2. You never take a loss. Only if you never close the set and never get closed out. An unrealised loss is still a loss, and the set can be forced shut.
3. Hedging locks the loss, so risk stops growing. It stops the price risk inside the zone, but you keep paying spread and swap, and each new leg adds margin and cost.
4. It is smarter than martingale. The growth factor is lower than doubling, but the shape is the same. The risk is slower to arrive, not absent.
5. A backtest with no losing months proves it. A smooth record can come from a system that has not yet met its deep cycle. Check the equity curve, the largest lot and the stress tests.

If you want to study the idea, run it on a demo account with the smallest lot and watch how the legs behave through a volatile week. You will learn more from three deep cycles on demo than from any sales page. If you then decide to trade it live, set the cap first and the lot second, and treat the sums in this guide as the minimum homework. The wider routine is in the MT5 Expert Advisors guide.