Anti-Martingale System — Sizing Up Positions on Wins

In her first year trading on her own, Anna opened every EUR/USD position at the same size — three mini lots, whether her account stood at €5,000 or €15,000. She finished the year with roughly €5,000 in profit and a 55 percent win rate, but her drawdowns piled up painfully because every losing trade kept hitting an unchanged position size. In the second year she switched to an anti-martingale system based on a fixed fraction of equity — 1.5 percent per trade — and her annual result climbed to €8,000. This article explains why this simple position-sizing mechanism is the standard choice of most professional traders and how to implement it in your own practice without falling into the usual traps.

What an anti-martingale system actually is

Where a martingale strategy doubles up after every loss in the hope of clawing the loss back, the anti-martingale does the opposite. After a string of wins the trader gradually increases exposure; after a string of losses he or she scales it back. The logic is both mathematical and behavioural: capital that has already grown can safely carry a larger position, while capital that has been eroded deserves protection from further damage. The same euro, at different points in the equity curve, carries very different risk.

Martingale versus anti-martingale

Fixed fractional — the most popular variant

The most widely used form of an anti-martingale system is the fixed fractional method — risking a constant percentage of current equity on every trade rather than a constant nominal amount. You can read a complete breakdown of these sizing principles in the risk management guide on ForexMechanics.com. The bigger the account, the bigger the position; the smaller the account, the smaller the position. The risk profile, expressed as a percentage of capital, never changes.

The position size formula is straightforward: the amount risked equals equity multiplied by the chosen risk percentage, and the lot size equals that amount divided by the product of the stop-loss distance in pips and the pip value. For a €10,000 account, a two percent risk and a 50-pip stop on EUR/USD, the calculation gives €200 divided by €500 per lot, which is 0.4 lots. When the account grows to €12,000, the formula automatically prescribes 0.48 lots at the same two percent risk. When it falls to €8,000, the position shrinks to 0.32 lots and the remaining capital is shielded.

The mathematics of risk of ruin

An anti-martingale system protects the account only as long as the risk taken on each individual trade stays moderate. The table below shows how long a streak of losses an account can absorb at different risk levels — and why moving from two percent to five percent is not a slightly bigger risk but a step into an entirely different category of bankruptcy probability.

• Two percent risk across ten consecutive losses produces a drawdown of about 18 percent — painful but easily survivable.
• Two percent across twenty consecutive losses takes the account down around 33 percent — a worst-case scenario, but still within reach of recovery.
• Two percent across thirty losses ends with a drawdown of roughly 45 percent — the edge of psychological and financial endurance, yet the career goes on.
• Five percent across ten losses already gives a 40 percent loss; twenty in a row leaves only about a third of the original capital.
• Ten percent across twenty losses wipes out 88 percent of the account — in practice, the end of trading on that balance.

The fixed ratio method of Ryan Jones

A second variant of the anti-martingale family is the fixed ratio method, developed by Ryan Jones in his 1999 book The Trading Game. Instead of holding a constant risk percentage, the trader adds another lot every time cumulative profit reaches a predefined increment, known as the delta. A larger delta produces slower scaling and lower exposure to a sudden drawdown immediately after an upsize; a smaller delta produces faster geometric growth at the cost of higher stress.

The Kelly criterion and its practical limits

The third variant — the Kelly criterion — is a formula derived from Claude Shannon’s information theory and adapted to gambling by the mathematician John Kelly in 1956. The optimal fraction of capital per trade equals the win probability multiplied by the reward-to-risk ratio, minus the loss probability, all divided by the reward-to-risk ratio.

For a strategy with a 60 percent win rate and a reward-to-risk ratio of 2:1, full Kelly prescribes risking around 40 percent of equity per trade. The number is mathematically correct, but in practice psychologically unbearable — at that setting, 50 percent drawdowns become a routine event rather than an exception. Professional traders therefore use Half-Kelly (half of what the formula suggests) or Quarter-Kelly (a quarter). Half-Kelly cuts the typical drawdown by about 75 percent in exchange for only a 25 percent reduction in growth rate, and most retail traders eventually settle back on the safer 1-to-2 percent fixed fractional anyway.

“I always reduce my position size during a drawdown. It is not a matter of taste — it is the difference between those who survive ten years in the markets and those who quit after two bad months.” — Ed Seykota, interviewed by Jack Schwager in Market Wizards, 1989.

How to implement the system in practice

Implementing an anti-martingale system comes down to seven concrete steps, best taken in the order below. Skipping any of them — particularly the historical-testing step — is the most common reason that excellent position management on paper turns into a losing account in practice.

1. Choose a variant matched to your experience. A beginner should start with a fixed fractional system at 1 percent; an intermediate trader can move up to 1.5 or 2 percent.
2. Backtest the strategy on at least one hundred historical trades. Confirm a positive expectancy before adding any position-sizing logic on top — the edge must exist before it can be optimised.
3. Build a position-size calculator. An Excel sheet or a MetaTrader script that takes the account balance, the risk percentage and the stop-loss distance as inputs and returns the lot size as output.
4. Keep a trade journal with a position-size field. Without it, you cannot later verify whether the system was applied mechanically or whether intuition crept in.
5. Recalibrate the account balance weekly. Each week you feed the current equity into the calculator, so that position size automatically follows the equity curve.
6. Run a monthly review of position-sizing impact. Were winning streaks captured with larger positions? Were drawdowns dampened with smaller ones during cold periods?
7. Carry out a quarterly optimisation of the risk percentage. A smooth equity curve allows for a modest increase; sharp zig-zags signal the need to step down.

The most common mistakes in applying the system

Years of working with individual traders show that the same mistakes appear again and again, among beginners and many intermediates alike. Knowing them in advance is the cheapest way to avoid paying for them in losses.

• A fixed lot size unrelated to the account balance. Half a lot on a €5,000 account and half a lot on €15,000 represent two completely different risk levels — about 10 percent and 3 percent respectively — whereas the same strategy ought to risk the same percentage of capital, not the same number of lots.
• Sizing up after a loss in revenge. This is a classic psychological martingale, even if the rulebook nominally specifies an anti-martingale system.
• Risking 5 to 10 percent of capital per single trade. The risk-of-ruin maths shows clearly that this is the fast track to wiping out the account during the first serious losing streak.
• Skipping the historical-testing step. Without a positive expectancy, the best position management in the world only slows the rate at which the account leaks.
• Neglecting the weekly recalibration. Within a month the actual balance drifts away from the figure in the calculator, and position sizes lose their connection to real equity.
• Confusing position sizing with the entry strategy. Fixed fractional optimises an existing edge, it does not create one — an important distinction that, when ignored, leads to the false comfort of “my position sizing is excellent, so the strategy must be working”.
• Switching variants every other week. Hopping between fixed fractional and the Jones method on a weekly basis is a lack of discipline, not a search for the optimum.
• Ignoring correlations between currency pairs. Three long positions on pairs with the US dollar in the denominator behave like one position three times as large whenever the dollar makes a sharp move.
• Not adjusting for volatility. The same risk percentage in a calm market and in an environment full of NFP, CPI and central-bank decisions represents two very different regimes — an ATR-based check on stop-loss distance helps to keep position size sensible.
• Discretionary overrides of the formula. “I feel good after two winners, so I will trade a bit bigger” turns a mechanical anti-martingale into intuitive chaos.

Case study — Anna over a five-year horizon

Let us return to Anna, whose story opened this article. Once she had moved to a fixed fractional system at 1.5 percent in her second year, she finished with a profit of €8,000 and a worst drawdown of around €800. The third year added another refinement: she began scaling into the strongest setups (pyramiding in the direction of an established trend) and closed the year with €12,000 of profit and a drawdown of roughly €1,000. In her fourth year she experimented with a hybrid — fixed fractional as the foundation, with a fixed-ratio overlay for the highest-conviction setups — and reached €15,000 of profit with a drawdown of €1,200. By the fifth year the system had matured to a yearly result of €18,000 while preserving a healthy rhythm of work and rest.

Across the five-year horizon her annual result rose from €5,000 to €18,000 — a 260 percent increase compared with the starting year, with each subsequent year showing drawdowns consistently softened by the anti-martingale mechanic. This is not a spectacular get-rich-quick story; it is a portrait of how compounding a market edge and proper position sizing actually look on the scale of several years.

What to do tomorrow

1. Analyze your trading history from the past month. Open your trading journal or broker records and check if your position sizes were fixed, variable, or completely random. Calculate how much your drawdowns would have been reduced if you had consistently applied a fixed fractional risk model of 1 percent per trade.
2. Create a simple position-size calculator in a spreadsheet. Set up a formula that automatically calculates the correct lot size for EUR/USD once you input your current account equity, desired risk percentage (such as 1.5 percent), and stop-loss distance in pips. Keep this spreadsheet open on your desktop before every session.
3. Establish a strict risk limit for the upcoming trading week. Write down and display under your monitor the following rule: „My maximum risk limit per trade is 1.5 percent of equity, and after three consecutive losses, I will cut my size in half”. Having this rule physically visible will help you resist emotional martingale revenge sizing.

Related reading: the martingale trap — the opposite of an anti-martingale and the reason it never pays to size up in revenge; the Kelly criterion — the mathematical justification of optimal position size; position size for different stop-loss distances — detailed calculations across currency pairs and scenarios.