Risk Management·Portfolio Management & Drawdown
Correlation & Diversification
More Trades Does Not Equal More Diversification
Having 5 open positions does not mean you have diversified risk. If all five are correlated — moving in the same direction in response to the same news — you're effectively trading one big position with five stops. True diversification means spreading risk across uncorrelated instruments. Understanding correlation is the difference between a portfolio and a concentrated bet disguised as a portfolio.
In March 2020, when COVID-19 pandemic fears triggered a global market crash, correlations across nearly all asset classes spiked to near 1.0. Stocks, commodities, emerging market currencies — everything sold off simultaneously. Traders who thought they were 'diversified' across different currency pairs found that every position moved against them at once. The lesson: correlations increase during stress events, which is precisely when you need diversification the most. Plan for crisis correlations, not calm-market correlations.
Definition
Correlation Coefficient
A number between -1 and +1 measuring how two instruments move in relation to each other. +1 means they move identically (buying both doubles your risk). -1 means they move in opposite directions (buying both hedges your risk). 0 means no relationship. Correlations are not static — they change over time and especially during market stress.
Definition
Concentration Risk
The risk that arises from having too much exposure to a single theme, sector, currency, or market driver. If 80% of your open risk is exposed to 'USD strength,' a single unexpected Fed announcement can hit all your positions simultaneously. Concentration risk is often invisible until it causes damage.
Definition
Beta
A measure of how much an instrument moves relative to a benchmark (usually the broader market). A beta of 1.5 means the instrument moves 1.5x as much as the benchmark. In forex, you might think of individual pair volatility relative to overall market volatility. High-beta pairs (GBP/JPY, NZD/JPY) amplify both gains and losses.
How to check correlations
Most trading platforms and websites offer correlation matrices that show the current correlation between currency pairs over different timeframes (daily, weekly, monthly). Check correlations on the same timeframe you trade. A pair that is +0.90 correlated on the daily chart might only be +0.50 correlated on the 5-minute chart. For swing traders, the daily and weekly correlation matters most. For day traders, the 1-hour and 4-hour correlations are more relevant.
| Correlation Range | Classification | Portfolio Implication |
|---|---|---|
| +0.80 to +1.00 | Strongly positive | Essentially the same trade — only take one |
| +0.50 to +0.79 | Moderately positive | Some diversification but significant shared risk |
| +0.20 to +0.49 | Weakly positive | Decent diversification — acceptable to hold both |
| -0.20 to +0.19 | Uncorrelated | Best diversification — independent risk profiles |
| -0.50 to -0.21 | Weakly negative | Natural hedge — reduces total portfolio volatility |
| -0.80 to -0.51 | Moderately negative | Strong hedge — but may cancel out both profits and losses |
| -1.00 to -0.81 | Strongly negative | Near-perfect hedge — holding both is nearly flat exposure |
Common Forex Correlations
| Pair 1 | Pair 2 | Typical Correlation | Implication |
|---|---|---|---|
| EUR/USD | GBP/USD | +0.85 to +0.95 | Highly correlated — holding both = double USD short exposure |
| EUR/USD | USD/CHF | -0.85 to -0.95 | Inversely correlated — holding both partially hedges |
| AUD/USD | NZD/USD | +0.80 to +0.90 | Highly correlated — both commodity/risk-sensitive currencies |
| USD/JPY | EUR/JPY | +0.70 to +0.85 | Both move with risk sentiment, JPY is the common driver |
| EUR/USD | USD/JPY | -0.50 to -0.70 | Moderate inverse — USD strengthens both, in different directions |
| USD/CAD | WTI Crude Oil | -0.60 to -0.80 | Oil strength = CAD strength = USD/CAD falls |
| AUD/USD | Gold (XAU/USD) | +0.50 to +0.70 | Both risk-sensitive, both benefit from USD weakness |
| GBP/JPY | S&P 500 | +0.60 to +0.80 | Both 'risk-on' instruments — rise together in optimistic markets |
The double-risk trap
If you're long EUR/USD and long GBP/USD simultaneously, you have effectively doubled your USD short exposure. A sudden USD strengthening news event (e.g. a hawkish Fed statement) will hit both trades at once, potentially doubling your expected loss. This is the most common portfolio-level mistake retail traders make.
Correlations spike in crises
During the 2008 financial crisis, correlations between virtually all risk assets approached +1.0. During the SNB event in 2015, all CHF pairs moved in lockstep. During COVID in March 2020, everything correlated with 'risk on/off.' Your diversification is only as good as it is during the worst markets — not during calm ones. Always stress-test your portfolio against a scenario where all your positions move against you simultaneously.
How to Build an Uncorrelated Portfolio
Building Diversified Exposure
- 1
Identify your core view
Do you have a directional bias? (e.g. USD strength, EUR weakness, JPY safe-haven demand). Start with one trade that expresses this view cleanly.
- 2
Check correlations before adding trades
Before entering a second trade, check its correlation to your first. If correlation is above +0.7 or below -0.7, the second trade is not truly independent risk. Use a correlation matrix tool.
- 3
Choose uncorrelated instruments
Look for pairs or instruments that respond to different drivers — e.g., USD/JPY (risk sentiment / BoJ policy) vs. USD/CAD (oil prices / BoC policy) vs. EUR/GBP (relative EU/UK growth).
- 4
Size each position independently
Apply the 1-2% risk rule to each individual position — but also consider your total portfolio exposure. If 3 trades are all effectively long USD, your total USD risk is 3x your per-trade risk.
- 5
Monitor and adjust
Correlations change over time. What was uncorrelated last month may become highly correlated this month due to macro regime shifts. Review your portfolio's correlation structure weekly.
| Portfolio Type | Example Positions | Effective Diversification | Total Risk Exposure |
|---|---|---|---|
| Concentrated (bad) | Long EUR/USD, Long GBP/USD, Short USD/CHF | None — all three are USD short | 3x intended risk on USD |
| Partially diversified | Long EUR/USD, Short USD/JPY, Long AUD/NZD | Moderate — mixed USD exposure + one non-USD trade | ~2x USD risk + independent AUD/NZD risk |
| Well diversified | Long EUR/GBP, Short USD/CAD, Long AUD/JPY | Good — different drivers for each | Spread across EUR, GBP, USD, CAD, AUD, JPY |
The total portfolio risk rule
As a general rule, limit your total correlated exposure to any single theme (e.g. USD direction, risk-on/off) to 5-6% of account equity at any time. This ensures even a major macro shock doesn't cause catastrophic losses. Some professional risk managers go further: no more than 3% of total equity exposed to any single currency.
The LTCM lesson on correlation
Long-Term Capital Management believed their positions were diversified because they held trades across many different markets (bonds, currencies, derivatives). However, when the 1998 Russian debt crisis triggered a flight to quality, all their positions correlated simultaneously. Spreads they expected to converge instead widened further. Their 'diversified' portfolio behaved as a single massive bet — and it lost $4.6 billion in weeks. True diversification must account for stress-scenario correlations, not just normal-market correlations.
Stress Testing Your Portfolio
Before holding multiple positions overnight, ask yourself: 'What happens if the USD surges 2% overnight on an unexpected event?' Then calculate the worst-case scenario for all your open positions combined. If the answer exceeds 5-6% of your account, you are overexposed regardless of how 'diversified' the positions look on paper.
Portfolio Stress Test Process
- 1
List all open positions and their risk amounts
Write down every open trade, its current risk in dollars, and the currency pair. Example: Long EUR/USD risking $150, Long GBP/USD risking $100, Short USD/JPY risking $120.
- 2
Group positions by common currency exposure
Identify which positions share common risk factors. In the example above, all three positions lose money if USD strengthens. Total USD-short exposure: $370.
- 3
Apply a stress scenario
Assume a 2-3 standard deviation move in the common factor. If USD surges 2% in one session, all three stops may be hit simultaneously. Total loss: $370 = 3.7% of a $10,000 account.
- 4
Evaluate if the stressed loss is acceptable
Is 3.7% acceptable for a single event? For most traders, yes — but barely. If a fourth USD-short trade were added, total stressed loss would exceed 5%. The stress test reveals when you need to reduce positions.
Entry
0.8580
Stop
0.8540
Target
0.8660
Diversification in action: If you already have a long EUR/USD position, adding a long EUR/GBP trade gives you EUR exposure WITHOUT adding more USD exposure. EUR/GBP is driven by the relative performance of the eurozone vs. the UK — a different driver than EUR/USD. This is genuine portfolio diversification. Stop at 0.8540 (40 pips), TP at 0.8660 (80 pips), RR = 1:2.
Knowledge check
A trader is long EUR/USD and long GBP/USD with 1% risk each. What is their effective USD exposure?
Knowledge check
During a market crisis, what typically happens to correlations between risk assets?