When five positions are really one position, five times over.
intermediate · 2 min read · 15 XP
You size each trade carefully at 1% and open five of them. You believe you are risking 5%. Depending on what you bought, you might be risking 5% — or you might effectively be risking 5% on a single bet.
You are long EUR/USD, long GBP/USD, long AUD/USD, short USD/CHF and short USD/JPY.
Five different pairs. One trade: short the US dollar, five times over.
If the dollar rallies, every one of those positions loses simultaneously. Your carefully sized 1% risks combine into a single 5% loss on one view. That is not diversification — it is concentration wearing a disguise.
The question to ask before every new position
"If this new trade goes wrong, will my existing positions go wrong at the same time?" If the answer is yes, the new trade is not adding a position — it is increasing the size of an existing one.
Correlation runs from +1 to −1:
Typical relationships (they shift over time, so measure rather than memorise):
| Pair combination | Typical correlation | Meaning |
|---|---|---|
| EUR/USD and GBP/USD | Strongly positive | Both are short-dollar |
| EUR/USD and USD/CHF | Strongly negative | Long one ≈ long the other |
| AUD/USD and NZD/USD | Strongly positive | Both commodity-linked, both short-dollar |
| EUR/USD and USD/JPY | Variable | Depends on the risk regime |
Note the second row: long EUR/USD and short USD/CHF is the same trade twice, because a strong negative correlation means an opposite position is a duplicate.
Think in currency exposure, not pairs. Add up your net exposure per currency. Long EUR/USD and long GBP/USD is: long EUR 1 unit, long GBP 1 unit, short USD 2 units. That USD number is your real position.
Cap exposure per currency, not just per trade. A rule like "no more than 3% total risk against any single currency" prevents the accidental five-fold dollar bet.
Prefer genuinely different bets. Long EUR/USD and short AUD/JPY are far less related than long EUR/USD and long GBP/USD. If you want several positions, choose ones that can be wrong independently.
Re-measure. Correlations change, sometimes abruptly. In a risk-off panic, almost everything becomes a dollar trade regardless of what the historical numbers said last month.
No — it is only bad when unmeasured. Deliberately expressing a strong dollar view across several pairs is a legitimate strategy, provided you sized it as one trade. The failure is not correlation; it is believing you have five independent positions when you have one.
What to remember
Correlated pairs turn several carefully sized trades into one large one. Think in net currency exposure rather than pair count, cap total risk per currency, and remember that correlations shift — in a panic, almost everything becomes a dollar trade.