Almost every article on trading risk says the same four things. Size small. Use a stop. Aim for decent reward. Do not overtrade. All correct, and all of it treats risk as a fixed property of a position. It is not. The same trade carries different risk on a quiet Tuesday than on the afternoon before a Fed decision. |
1 | The Problem With Standard Risk Advice |
The conventional framework: decide what percentage you are willing to lose, work out where your stop belongs, size the position so the distance to the stop equals that percentage, repeat. There is nothing wrong with it, and most people who blow up an account never adopted even this much. But it assumes your risk is determined by the position and the stop, and that nothing else matters. In macro markets, three things can change the risk of a position you have not touched. The calendar, because a binary event changes the distribution of outcomes entirely. Correlation, because what looks like four trades may be one trade wearing four costumes. And liquidity, because a stop is a promise that depends on somebody being there to take the other side. None of those appear on a chart. |
|
2 | Your Book Is Probably One Trade |
Imagine a trader who is long gold, short USD/JPY, and long EUR/USD. Three positions, three markets, each sized carefully. It feels diversified. It is not. Long gold is a bet against the dollar. Short USD/JPY is a bet against the dollar. Long EUR/USD is a bet against the dollar. One hawkish Fed surprise and all three lose simultaneously, and the total loss is roughly three times what any single position was sized for. |
|
What happened last week The Fed raised rates and published projections showing the median expectation for end-2026 rising to 4.1% from 3.8%, with the whole forecast path shifting up. The dollar strengthened against every major at once. Anyone holding a spread of short-dollar positions took a correlated loss across all of them, in a single afternoon. |
|
The fix requires honesty rather than complexity. Before you count your risk, ask what single variable would hurt every open position at the same time. If the answer is a hawkish Fed, or risk-off sentiment, or higher oil, then that is your real exposure, and your real risk is the sum of those positions rather than the largest one. Most macro pairs reduce to a handful of underlying bets: the dollar, risk appetite, energy, and a specific rate differential. |
|
3 | The Calendar Changes Positions You Already Hold |
On an ordinary day price moves in something approximating a continuous fashion, and your stop is likely to be respected roughly where you put it. On an FOMC afternoon or an inflation release, the distribution changes shape. Price can move a long way very quickly, and the gap between where your stop sits and where it fills can be substantial. This is not a reason to avoid holding through events. Plenty of good swing trades require it, because the event is what resolves the thesis. It is a reason to decide consciously. The question worth asking on Sunday is simple: which of my open positions has a scheduled event this week, and am I comfortable holding through it at this size? Sometimes the honest answer is that you would be comfortable at half the size. |
|
4 | Gap Risk, and Where It Comes From |
The obvious source is the weekend. Markets close Friday and reopen Sunday, and anything in between arrives as a gap rather than a move you could have managed. Geopolitical developments are particularly prone to landing there. The less obvious source is central bank meetings outside your trading hours. The Bank of Japan decides during Asian hours. The ECB decides during European morning. Last week made this concrete: the Fed decided Wednesday afternoon, the Bank of England Thursday, and the Bank of Japan across Thursday and Friday. Three decisions in roughly 48 hours across three timezones. For anyone in sterling or yen pairs, the Fed's move was repriced again before it had settled, and at least one of those repricings happened while most people were asleep. |
|
5 | Liquidity Is Not a Constant Either |
A stop loss is an instruction, not a guarantee. It executes at the best available price, and around major releases the best available price can be considerably worse than the level you set. Spreads widen, sometimes by a multiple of normal, and orders fill away from where you expected because there is nobody on the other side for a few seconds. A position sized for a one percent loss can produce a larger one purely because of when the loss occurred rather than how wrong the idea was. There is no clever solution. Reduce size ahead of the release, accept the slippage consciously, or be flat through it. What does not work is assuming your stop will behave the way it does on a normal Tuesday. |
|
Know What Is Coming Before It ArrivesEvery Sunday, FedAndMarkets covers the macro calendar and what is actually moving seven markets, so you know what your positions are exposed to before the week starts. |
|
6 | Sizing to the Calendar, and the Thesis Stop |
In a week with no major scheduled events, your risk is closer to what your framework assumes. In a week with a central bank decision or a major inflation print, the distribution is wider and your effective risk is larger than the number in your spreadsheet. A fixed percentage applied identically every week is not actually a constant level of risk. It only looks like one. On stops, most placement advice is technical. For a macro swing trade there is a better question: what price level would tell me my reason for being in this trade was wrong? A stop that sits where the thesis breaks is a stop you can actually hold, because most people move or cancel stops they do not believe in. And a thesis can break without price hitting your stop at all. If you are long the dollar on a hawkish Fed and the following data comes in soft enough that rate expectations start unwinding, the reason you entered has expired even though the position has not stopped out. Exiting there is not impatience. It is consistency. |
|
|
|
Sort your positions by underlying bet. Not by market. By what they actually express: dollar direction, risk appetite, energy, a specific rate differential. Count exposure per bet, not per position.
Check which positions have a scheduled event. Note which ones land while you will be asleep.
Decide consciously about each event. Full size, reduced size, or flat. Any of those is defensible. Drifting into it without deciding is not.
Write the invalidation, not just the stop. What would have to happen for your reason to no longer be true? Sometimes a price. Sometimes a data release.
Expect worse fills around releases. Assume slippage in the minutes around a major event and size with that in mind.
Review what actually hurt you. Separate losses from a wrong idea and losses from correlation, gap or slippage. They need completely different fixes, and most people treat them as one problem.
|
|
Key TakeawaysRisk is not a fixed property of a position. The calendar changes it, correlation between your own positions changes it, and liquidity around scheduled events changes it. The most common hidden exposure is correlation. Several positions across different markets frequently express one underlying bet, usually on the dollar. You cannot control whether any individual idea works. You can control whether being wrong once costs you more than it should. |
|
Go Deeper
|
Get This Context Every WeekEvery Sunday, FedAndMarkets covers the macro calendar and what is actually moving seven markets. No signals. No predictions. Free every Sunday · 7 markets · No spam |
|
You cannot control whether any individual idea works. You can control whether being wrong once costs you more than it should. — Fed'n Markets |
|