The franc event is the cleanest teaching case in retail trading because nothing about it required an unusual position. EURCHF had traded in a narrow band above 1.20 for three years, held there by a stated central bank commitment. It was the textbook low volatility carry position. When the commitment was withdrawn without warning, price fell through the floor in minutes with no functioning market underneath, and the losses were sized by the gap rather than by anyone's stop.
Brokers went under. Some pursued clients for negative balances, some absorbed them, and the difference came down to jurisdiction and policy rather than to anything the client had done.
Why normal risk maths understates the tail
Most position sizing assumes returns behave roughly normally: small moves common, large moves rare, extreme moves effectively impossible. Financial returns do not behave that way. The distribution has fat tails, meaning the extremes occur far more often than the bell curve predicts, and when they occur they are larger than the model allows for.
The practical consequence is that a stop distance calibrated on recent volatility is calibrated on the middle of the distribution. It tells you what a normal adverse move looks like. It says nothing about what happens when the market that is supposed to fill your stop is not there, and it is systematically most reassuring in exactly the low volatility periods that precede shocks.
Correlations converge when it matters
Portfolio diversification rests on positions not moving together. In a shock they do. Assets that showed low correlation for years move as one because the driver stops being anything specific to them and becomes a single question: who needs to raise cash right now.
A trader holding six positions across different pairs believes they hold six risks. During a crisis they often hold one risk, expressed six times, and the account moves six times as fast as expected. Checking currency correlations in calm conditions is worth doing, provided you remember that the numbers you are reading describe calm conditions and will not hold during the event you are protecting against.
A black swan is defined partly by being outside the range of what you modelled. Any protection that depends on knowing the size or timing of the shock in advance is not protection. The defences that work are the ones that do not require a forecast.
Defences that do not require a forecast
Total exposure caps. Not risk per trade, but the sum of everything open at once, measured as the fraction of the account that a simultaneous adverse gap in all positions would cost. Most traders can state their per trade risk instantly and have never calculated the aggregate.
Overnight and weekend policy. A shock that arrives while markets are closed reaches you as an opening gap with no opportunity to react. Reducing exposure through closed periods costs some opportunity and removes the worst version of the outcome. Our note on weekend gaps covers the mechanics.
Negative balance protection. In several jurisdictions retail clients cannot lose more than their account balance, because the rule requires the firm to write off the excess. Elsewhere it is a commercial policy that can differ by account type or client classification. Read the terms rather than assuming, since this is precisely the clause that decides whether a bad day ends at zero or in a demand letter.
Capital held away from the trading account. Money at the broker is exposed to the broker as well as to the market. Client fund segregation and investor compensation schemes reduce that exposure without eliminating it, and 2015 showed that firm failure and market shock arrive together rather than separately.
What the shock does to a firm, beyond the trader
If you run a brokerage or a prop firm, the same event looks different. Client losses that exceed deposits become your losses under negative balance protection. Hedged exposure can fail if your liquidity provider stops quoting or if the fills you receive are worse than the fills you gave. Margin calls to your own counterparties arrive on the same day your clients cannot fund theirs.
The firms that survived 2015 mostly had two things: exposure limits that were enforced automatically rather than reviewed weekly, and enough capital sitting above the regulatory minimum to absorb a bad day without a fire sale. That is a treasury and risk system question, and it is why real time exposure monitoring is a core function of any broker back office rather than a reporting nicety. Aggregate client exposure per instrument, updated continuously, is the number that tells you whether the next headline is survivable.
Living with uncertainty you cannot price
None of this is a forecast, and any tool sold as one deserves suspicion. The honest position is that the timing and shape of the next shock are unknown, that leveraged trading carries a high risk of loss even without one, and that the only variable under your control is how much of your capital is standing in front of the move when it comes.
The traders who came through 2015 with an account intact were not the ones who read the SNB correctly. They were the ones whose position size meant that being completely wrong was survivable. That is a duller skill than analysis and it is the one that compounds, because it is the only one that requires you to still be trading next year.
"Everybody agrees the tail risk is real and everybody sizes as though it is not. The only test that matters is whether your account survives a move you did not think possible, because you will get one."
— Alex Onta, Executive Director, SINGUARD
Key Takeaways
- The 2015 Swiss franc move produced account destroying losses from ordinary position sizes, because the loss was set by the gap and not by the stop.
- Return distributions have fat tails, so volatility based sizing describes normal conditions and understates the extremes.
- Correlations converge during shocks, turning what looks like six independent positions into one risk repeated six times.
- Aggregate exposure caps, reduced weekend risk, negative balance protection and capital held off the platform work without needing a forecast.
Frequently Asked Questions
What is a black swan event in trading?
It is a market event that falls outside the range of outcomes a participant considered possible, arrives with little or no warning, and has an outsized effect. The removal of the Swiss franc cap in January 2015 is the standard retail example, since it moved a pair that had been stable for years by an amount no model allowed for.
Can a stop loss protect against a black swan?
Only partly. A stop triggers an exit attempt once the level trades, but if there are no bids between your level and a price far below it, the fill happens far below. In the 2015 franc event stops were filled hundreds of pips away from where they were set. A guaranteed stop, where offered, fixes the exit price for a premium.
What is negative balance protection?
It is the rule or policy under which a client account cannot go below zero, with the firm writing off any shortfall. It is a regulatory requirement for retail clients in several jurisdictions and a voluntary commercial policy elsewhere, so it can vary by account type, client classification and jurisdiction. The account terms are the place to check.
About the Author
Alex Onta is an Executive Director at SINGUARD. He built eTrader, the terminal, the mobile apps, eTrader Broker, Copytrading, Business and Community, along with the worldwide clustered-server infrastructure it all runs on, with his brother Roman Onta helping on the design, and he leads that division today. Together with Roman he builds the Prop Firm CRM, the Broker CRM, Scalegram and CopySignals, and the two of them carry worldwide compliance, payment processing and international business structuring side by side. He lives and works in Dubai for most of the year. Meet the executive duo leading Singuard's five divisions.