The mathematical proof: a comparison of two traders

Trader A: Perfect timing, poor sizing

Performance over 100 trades:

• 75 wins × 50 pips × 5% = +187.5%
• 25 losses × 40 pips × 5% = -50%
• Net return: +137.5%

Sounds good? Let's look at Max Drawdown:

At 5% per trade and 25 losses (which come in clusters):

• Worst case: 7 consecutive losses = -35% drawdown
• After -35%: You need +54% for recovery
• Psychological Pain → System Abandonment

Trader B: Average timing, professional sizing

Performance over 100 trades:

• 55 wins × 50 pips × 1.8% (avg) = +49.5%
• 45 losses × 40 pips × 1.8% = -32.4%
• Net return: +17.1%

Lower returns than Trader A? Yes. But:

• Maximum drawdown: 12% (instead of 35%)
• Recovery time: 8 trades (instead of 30+ trades)
• Psychological survivability: High
• Compound effect over the years: Superior

After 3 years:

• Trader A: 2 large drawdowns → system hopping → inconsistent returns → net +12% p.a.
• Trader B: Consistent 17% → Compounding → Net +17% p.a.

Result: Trader B ends up with 60% more capital—despite a lower win rate.

For Forex software for asset managers, this means:

Position sizing algorithms are more important than entry signal optimization. Top systems invest 70% of development time in risk management and only 30% in signal generation.

The five professional position sizing methods

Method 1 – Fixed-Fractional: The basis of all systems

What it is:

You risk a fixed percentage of your capital per trade.

Formula:

Position size = (account capital × risk %) / (entry – stop loss in pips × pip value)

Example:

• Account: EUR 100,000
• Risk tolerance: 2%
• Trade: Long EUR/USD at 1.1800, stop at 1.1750 (50 pips)
• Pip value for 1 lot: EUR 10

Calculation:

Position size = (100,000 × 0.02) / (50 × 10) = 2,000 / 500 = 4 lots

Critical: You lose a maximum of EUR 2,000 (2%) on every trade, regardless of the pair or volatility.

Advantages:

✓ Einfach zu implementieren ✓ Schützt vor Ruin (Kelly-Criterion-kompatibel bei <2,5%) ✓ Adaptiv: Wächst mit Konto, schrumpft nach Verlusten

Disadvantages:

✗ Ignores market volatility (50 pips in low volatility vs. high volatility carry different levels of risk) ✗ No adjustment for trade quality ✗ Can be too aggressive with volatile pairs

Best practices for foreign exchange trading for family offices:

• Conservative: 1% risk
• Moderate: 1.5% risk
• Aggressive: 2% risk (institutional maximum)

Never >2.5% – above this level, the probability of ruin increases exponentially.

Method 2 – Kelly Criterion: The mathematically optimal method

What it is:

A formula that calculates the mathematically optimal position size based on win rate and win/loss ratio.

Formula:

Kelly % = W – [(1 – W) / R]

W = Win rate (as a decimal, e.g., 0.60)

R = Win/Loss Ratio (Avg Win / Avg Loss)

Example:

Your strategy has:

• Win rate: 60% (W = 0.60)
• Avg-Win: 50 pips
• Average loss: 40 pips
• Win/loss ratio: 50/40 = 1.25 (R = 1.25)

Calculation:

Kelly % = 0.60 – [(1 – 0.60) / 1.25]
Kelly % = 0.60 – [0.40 / 1.25]
Kelly% = 0.60 – 0.32
Kelly % = 0.28 = 28%

28% per trade?! That's insane.

That's right. That's why nobody uses "Full Kelly."

Half-Kelly (standard among institutional traders):

28% / 2 = 14% per trade

Quarter Kelly (conservative):

28% / 4 = 7% per trade

Advantages:

✓ Mathematically optimal for long-term capital growth ✓ Balances risk vs. reward based on actual edge ✓ Prevents over-betting (when edge is low, Kelly recommends small sizes)

Disadvantages:

✗ Extremely sensitive to input parameters (small errors in win rate estimation → massive mispositioning) ✗ Full Kelly leads to 50%+ drawdowns (psychologically unacceptable) ✗ Requires a stable win rate (but markets change)

Real-world application of Forex algorithm for asset managers:

Step 1: Calculate Kelly based on the last 100 trades Step 2: Use Quarter-Kelly (1/4 of the result) Step 3: Cap at a maximum of 2.5%

Example:

• Kelly says: 20%
• Quarter Kelly: 5%
• But: Cap at 2.5% → You use 2.5%

Safety first.

Method 3 – ATR-based: Volatility adjustment

Problem with fixed fractional:

A 50 pip stop on EUR/USD in low volatility (ATR 60 pips) is different from 50 pips in high volatility (ATR 120 pips).

Solution: ATR-based sizing

Formula:

Position size = (account × risk%) / (ATR multiple × ATR × pip value)

Example:

• Account: EUR 100,000
• Risk: 2%
• EUR/USD: ATR(14) = 80 pips
• ATR multiple: 2× (stop loss at 2× ATR = 160 pips)
• Pip value: EUR 10

Calculation:

Position size = (100,000 × 0.02) / (2 × 80 × 10)
Position size = 2,000 / 1,600
Position size = 1.25 lots

Comparison with fixed fractional with 50-pip stop:

Fixed-Fractional would use 4 lots (see above). ATR-based only 1.25 lots.

Why?

Because ATR method understands: With 80-pip ATR, 160-pip stop is "normal." With fixed 50-pip stop, you risk normal volatility stopping you out.

Advantages:

✓ Automatically adjusts to market volatility ✓ Larger positions during low volatility, smaller positions during high volatility ✓ Avoids being "stopped out by noise"

Disadvantages:

✗ More complex to calculate ✗ ATR period must be selected (14? 20? 50?) ✗ In cases of extreme volatility, positions become very small (missed opportunities)

Best practices for premium Forex software:

Hybrid approach:

1. Calculate ATR-based size
2. Calculate fixed fractional size
3. Use the SMALLER of the two

Example:

• ATR-based: 1.25 lots
• Fixed fractional (2%, 50-pip stop): 4 lots
• Select: 1.25 lots (conservative)

Method 4 – Volatility Percentile Adjusted

The problem with ATR-based:

ATR only adjusts to absolute volatility, not relative volatility.

Example:

• EUR/USD: Current ATR 80 pips, historical median 70 pips → 14% above median
• GBP/USD: Current ATR 120 pips, historical median 95 pips → 26% above median

GBP/USD is relatively more volatile (26% vs. 14% above normal), but the ATR method only looks at absolute figures.

Solution: Volatility percentile adjustment

Formula:

Adjusted risk% = Base risk% × (1 – Volatility percentile factor)

Volatility percentile factor = (current ATR percentile – 50) / 100

Example:

• Base risk: 2%
• Current ATR is in the 75th percentile (higher than 75% of historical values)
• Factor = (75 – 50) / 100 = 0.25

Adjusted risk% = 2% × (1 – 0.25) = 1.5%

If ATR is in the 90th percentile (extremely high):

• Factor = (90 – 50) / 100 = 0.40
• Adjusted risk = 2% × (1 – 0.40) = 1.2%

If ATR is in the 30th percentile (low):

• Factor = (30 – 50) / 100 = -0.20
• Adjusted risk = 2% × (1 + 0.20) = 2.4%

Advantages:

✓ Takes relative volatility into account, not just absolute volatility ✓ Automatic reduction during volatility spikes ✓ Automatic increase during volatility compression

Disadvantages:

✗ Requires long historical volatility data ✗ Complex calculation ✗ Can be too conservative (miss opportunities during high volatility)

For automated foreign exchange trading strategies:

This method is ideal for systematic traders. However, it requires a robust database with historical ATR values covering 5-10 years.

Method 5 – Correlation-Adjusted Position Sizing

The overlooked risk: correlated positions

Scenario:

You are:

• Long EUR/USD (position: 3 lots)
• Long GBP/USD (position: 3 lots)

Your risk per trade: 2% → Total 4%? No.

EUR/USD and GBP/USD correlate at +0.85. When the dollar strengthens, BOTH positions lose value simultaneously.

Effective risk: ~7% (not 4%)

Solution: Correlation-Adjusted Sizing

Formula (simplified):

Effective risk = √(Risk1² + Risk2² + 2 × Risk1 × Risk2 × Correlation)

Example:

• Risk1 (EUR/USD): 2%
• Risk2 (GBP/USD): 2%
• Correlation: 0.85

Effective risk = √(0.02² + 0.02² + 2 × 0.02 × 0.02 × 0.85)
Effective risk = √(0.0004 + 0.0004 + 0.00068)
Effective risk = √0.00148 = 0.0385 = 3.85%

Almost twice as much as the 2% you thought!

Adjustment:

If you want a maximum total risk of 3%:

• Reduce both positions by a factor of 3% / 3.85% = 0.78
• New position sizes: 3 lots × 0.78 = 2.34 lots each

Advantages:

✓ Prevents "hidden" risk due to correlations ✓ Critical for multi-pair portfolios ✓ Protects against dollar strength/weakness events (all USD pairs move simultaneously)

Disadvantages:

✗ Very complex to calculate (for 5 pairs: 10 correlation pairs) ✗ Correlations change (2018: EUR/GBP +0.85, 2020: +0.72) ✗ Significantly reduces position sizes (less profit potential)

For Forex portfolio management:

Essential for 3+ simultaneous positions. High-end trading software should automatically update the correlation matrix and adjust position sizes.

Position sizing in practice: The 1000FTAD framework

The standard institutional method

What JP Morgan and top institutions use:

Basis: ATR-based fixed fractional with correlation overlay

The framework:

Step 1: Basic risk calculation

Base risk = 1.5% (conservative for institutional mandates)

Step 2: ATR adjustment

ATR multiple = 2× (stop at 2× ATR)

Stop distance = ATR(14) × 2

Step 3: Position size calculation

Position = (Capital × Base Risk) / (Stop Distance × Pip Value)

Step 4: Volatility percentile check

If ATR > 80th percentile: Reduce position by 25%

If ATR > 95th percentile: Reduce position by 50%

Step 5: Correlation check

If a long USD position already exists:

  Calculate correlation-adjusted risk

  If Effective Risk > 3%: Reduce new position

Step 6: Hard Caps

Maximum position per trade: 3% of capital

Maximum total exposure: 10% of capital

Real-world example: EUR/USD trade

Setup:

• Account: EUR 1,000,000
• Base risk: 1.5%
• EUR/USD at 1.1800
• ATR(14): 85 pips
• Stop: 2× ATR = 170 pips (entry – 170 pips = 1.1630)

Calculation:

Step 1: Base risk = EUR 15,000 (1.5% of 1 million)

Step 2: Stop distance = 170 pips

Step 3: Position = 15,000 / (170 × 10) = 15,000 / 1 ,700 = 8.82 lots

Step 4: Check volatility percentile

• Current ATR 85 pips is in the 65th percentile (median)
• No adjustment necessary

Step 5: Check correlations

• No other USD positions open
• No adjustment necessary

Step 6: Hard cap check

• 8.82 lots at 1.1800 = EUR 1,040,400 exposure
• That is 104% of the capital → Above cap (100% maximum)
• Reduce to maximum: 100% = 8.47 lots

Final position size: 8.47 lots

Risk check:

If stop is hit at 1.1630:

• Loss = 170 pips × 8.47 lots × 10 EUR = 14,399 EUR
• That is 1.44% of the capital ✓ (below the 1.5% target)

For institutional forex trading software:

This entire process must be automated. With manual calculation: 5-10 minutes per trade. With software: 0.3 seconds.

The psychological pitfalls of position sizing

Case 1 – "Martingale thinking" after losses

What traders do:

After 3 consecutive losses: "The next trade MUST win" → Increase position size to 5%.

Mathematics:

Each trade is independent. Probability of the next win: unchanged.

The result:

If the fourth trade also loses (probable with a 55% win rate system): Catastrophic loss.

The rule:

Position size is based on capital and volatility. Never on "feelings" or past trades.

Case 2 – Overconfidence after winning

What traders do:

After 5 wins: "I'm hot!" → Position size to 4%.

The reality:

5 wins with a 60% win rate system occur in 7.8% of all cases. This is normal distribution, not a "hot streak."

The risk:

Increased position size precisely when mean reversion is likely (statistically, a winning streak is followed by a losing streak).

The solution for Forex risk management software:

Position sizes are calculated algorithmically. No manual overrides are permitted. Emotions are eliminated.

The five core principles of position sizing

1. Position sizing beats entry timing

An average system with excellent risk management outperforms an excellent system with poor risk management – over the long term.

2. Choose your method based on context

• Fixed Fractional (1-2%): Basis for all traders
• Kelly Criterion (Quarter-Kelly): For mathematically oriented traders with stable win rates
• ATR-based: Standard for institutional traders
• Volatility percentile: For adaptive systems in volatile markets
• Correlation-Adjusted: A Must for Multi-Pair Portfolios

3. Never more than 2.5% per trade

Regardless of the method: 2.5% is the mathematical limit. Above that, the probability of ruin increases exponentially.

4. Automation is not optional

Manual position size calculation leads to errors, inconsistency, and emotional overrides. Fully automated forex trading with programmed risk parameters is the standard.

5. Correlation awareness is critical

In multi-pair trading: correlations make your risk 2-3 times higher than you think. Ignoring this = ruin during dollar stress events.

The truth: Position sizing isn't sexy. It doesn't generate spectacular profits. But it's the difference between 10 years of profitable trading and bankruptcy after 18 months.

Professional capital management in foreign exchange trading – with the 1000FTAD software

Our automated foreign exchange trading strategies undergo a rigorous 6-month validation framework – developed with 17 years of institutional trading experience.

What our system offers:

✓ Foreign exchange trading for family offices: Fully automated, rule-based, disciplined
✓ Strategic trading for currency pairs without emotional decisions
✓ Forex portfolio management with institutional risk parameters
✓ Software for foreign exchange trading with risk management and compliance reporting
✓ Trading experience combined with former JP Morgan trading desk experts
✓ Continuous optimization based on market regime changes

1000FTAD stands for controlled, technology-driven foreign exchange trading—with a focus on substance, discipline, and long-term asset stability.

Find out more in a personal consultation:

📧 info@1000ftad.com
📞 +41 71 588 03 40

Exclusively for family offices, asset managers, and institutional investors with minimum assets of EUR 10 million

FAQ: Frequently asked questions

Q: Isn't 2% risk per trade too conservative? Can't I grow faster with 5%?

A: Mathematically: Yes, 5% delivers higher returns with the same win rate. But: (1) Drawdowns become 2.5 times larger (with 7 consecutive losses: -35% instead of -14%), (2) Psychological stress leads to system abandonment, (3) Probability of ruin increases exponentially. Studies show: Traders with >3% risk have a 68% higher failure rate over 3 years. Institutional limits: 2% maximum.

Q: Which position sizing method is the "best"?

A: Depends on the context. For beginners: Fixed fractional (1.5%). For experienced traders: ATR-based. For quants: Kelly criterion (quarter Kelly). For multi-pair: Correlation-adjusted is mandatory. Premium forex software should support and be able to compare all methods.

Q: Do I have to recalculate position size for every trade?

A: Yes – but software does that automatically. Your capital changes after every trade, ATR changes daily, correlations shift. Manual calculation: 5-10 min/trade = impractical. Algorithmic: 0.3 seconds. Customized Forex trading solutions automate this completely.

Q: What do I do if the Kelly criterion recommends 15% (even quarter Kelly = 3.75%)?

A: This means your system has an extremely high edge (win rate >65% AND win/loss ratio >2). Nevertheless: cap at 2.5%. Why? (1) Kelly assumes perfect win rate knowledge (impossible), (2) 15% leads to 50%+ drawdowns, (3) your edge can change. Use Kelly for confirmation ("system has edge"), but follow institutional caps.

Q: How do I deal with pairs that have different levels of volatility (EUR/USD vs. GBP/JPY)?

A: That's exactly what ATR-based sizing is for. With fixed fractional with fixed pip stops: GBP/JPY becomes massively undersized (too small a position) or EUR/USD oversized (too large a position). ATR method normalizes: Both pairs get position sizes that match their natural volatility. Forex analysis software for executives should automatically integrate pair-specific ATR values.

Q: Can I backtest position sizing?

A: Absolutely—and you should. Test your strategy with different sizing methods. Often, a strategy with 1% fixed fractional has a Sharpe ratio of 1.8, while the same strategy with 3% fixed has a Sharpe ratio of 0.9 (worse despite higher absolute returns). Backtesting shows which sizing method optimizes risk-adjusted return. FX software for professionals should include sizing optimization as a feature.

Copyright © 2025 1000FTAD AG – All rights reserved

Note: This article does not constitute investment advice. It is a market assessment for professional investors.

Find out more now: www.1000ftad.ch 📩 Contact: https://1000ftad.ch/kontakt/

Risk warnings: 1000FTAD products are only suitable for professional investors and qualified investors. Further information: https://1000ftad.ch/rechtlicher-hinweis/