The conventional wisdom on 24/7 currency trading is that it is dangerous, and the danger is structural. Liquidity thins when the interbank desks in London, New York, Singapore and Tokyo are all dark at once. Spreads widen. Stop-losses execute at prices the trader never authorized. When Vantage publishes a risk disclosure warning its clients about liquidity and execution during extended sessions, it is repeating a position that every serious execution desk has held since at least January 15, 2015 — the morning FXCM discovered what a fifteen-minute liquidity vacuum does to a client book that was margined against a peg the market no longer believed in.
Why This Is Actually True
We should concede the argument in its strongest form before we do anything with it. The Vantage disclosure — that liquidity providers pull quotes during weekend and holiday sessions, that spreads on major pairs can widen to multiples of their weekday averages, that stop-loss orders may execute meaningfully away from their trigger price — is not a marketing device. It is what the execution record shows.
Every execution desk that survived the post-2015 rewiring of prime-of-prime relationships arrived at the same operating rule. When the top-of-book depth on EUR/USD collapses from the eight-to-ten-figure notional visible during the London-New York overlap to the sub-million-dollar quotes that populate Friday-night through Sunday-evening books, the mechanics of a stop order stop working the way retail traders imagine them working. A stop is not a price guarantee. It is an instruction to convert into a market order once a trigger is touched — and a market order into a thin book is a request for the worst reasonable price available.
The FXCM postmortem published in the weeks after the Swiss National Bank abandoned the EUR/CHF floor is the document to read here. FXCM described a client-facing execution stack that had performed within tolerance during normal-liquidity hours for years, then failed in a specific way when the quote stream fragmented. The failure was not a software defect. It was a reasonable system operating against an unreasonable book.
So when Vantage tells its clients that 24/7 trading — particularly the crypto pairs and the extended-hours CFD indices that keep quoting when the underlying spot desks close — carries execution risk that the client should quantify before sizing a position, the warning is defensible on its face. The desks that ignored this warning are, in most cases, no longer desks.
But here is what that framing misses entirely: the risk the disclosure describes is not the risk that actually did the damage.
Where It Breaks Down
The Vantage-style warning invites a specific mental model. The trader imagines a normal book, then a thin book, then a slightly wider spread, then perhaps a stop that fills a few pips beyond its trigger. The failure mode implied by the disclosure is quantitative — you lose a bit more than you thought you would.
The historical execution record describes something categorically different. When FXCM's book blew apart on January 15, 2015, the problem was not that stops filled two or three times worse than the client expected. The problem was that a subset of client positions filled into a market that had moved through more than 15% of the pair's price in under half an hour, with the client's margin covering approximately 2% of that move. The delta between what the margin covered and what the book actually settled at was, in aggregate across the FXCM client base, a nine-figure receivable from clients who no longer had the funds to make the broker whole. FXCM disclosed a client-side deficit of approximately $225 million in the immediate aftermath. The firm required a rescue loan of $300 million from Leucadia to remain solvent, on terms that eventually forced the sale of the U.S. retail business and the abandonment of the FXCM Inc. holding structure.
None of that is what the Vantage disclosure prepares a reader for. The disclosure prepares the reader for wider spreads. What actually did the damage — historically, on the record, with primary documents — was the failure of the entire concept of "spread" as a bounded quantity once liquidity vanishes on both sides simultaneously.
Consider the math. Suppose a trader carries a position of 100,000 units of a pair, uses 1:100 leverage, and posts $1,000 in margin. A 100-pip move against them consumes $1,000 — the full margin. A 200-pip move produces a $1,000 client-side deficit the broker must either absorb, chase, or write off. In a normal-liquidity failure, the broker's stop-out logic closes the position somewhere between 100 and perhaps 150 pips of adverse move — the deficit is small, the client is negative-balance protected in most jurisdictions, and the broker's exposure is limited. In the January 2015 failure, adverse moves on affected accounts ran to 1,500-3,000 pips before any counterparty was willing to quote at all. The stop-out logic did not fail. It simply had nothing to execute against.
That is the objection to the Vantage framing. The disclosure describes a graduated risk. The record describes a discontinuous one.
The Rule I Use Instead
The rule that survives a reading of the FXCM, MF Global and Refco postmortems is not "avoid trading in thin liquidity windows." That rule is too weak — it lets the trader believe that if they sit out Friday nights, they have solved the problem. The rule is: your position size is capped by the worst gap you can imagine, not by the spread you can currently see.
Concretely. Before entering any position that will be held through a low-liquidity window — weekend crypto, holiday sessions in a CFD index, an extended hours EM currency — the trader should identify the largest single-tick gap that pair has recorded historically, multiply it by their leverage, and confirm that the resulting drawdown does not exceed their comfort loss. Not their margin. Their comfort loss. If EUR/CHF can gap 1,500 pips in a session — because it did — and the trader is running 1:100 on a similar pair, the operative question is not whether the spread will widen. It is whether the trader can absorb a 15,000-pip loss on the notional they are carrying, because that is what the tape has done, in living memory, to a pair that most desks considered pegged.
This rule reframes what a broker's execution disclosure is for. A Vantage-style warning about spreads widening in 24/7 sessions is not the risk primer. It is the last-line reminder. The primary risk primer is the reader's own reconstruction of what the worst historical gap in each of their instruments looks like against their current leverage.
The reason this reframing matters is that it changes broker selection criteria. Under the Vantage framing, the trader chooses among brokers on the basis of spread quality during normal hours. Under the historical framing, the trader chooses on the basis of negative-balance protection posture, prime-of-prime concentration, and the specific language of the risk-transfer clauses in the client agreement — the same clauses FXCM's clients discovered, after the fact, gave them liability for a market movement the firm had not internally sized for.
When the Old Rule Still Wins
We should be honest about the limits of the reframing. For the trader who is scalping majors during London-New York overlap, sitting flat across weekends, and never carrying a position into a central bank announcement, the graduated-risk framing that Vantage's disclosure implies is a reasonable working model. Their exposure to a discontinuous liquidity failure is genuinely small. Spreads widening from 0.4 pips to 2 pips at 5pm New York on a Friday is, for a flat book, an academic observation.
The historical rule applies to the trader carrying overnight and over-weekend risk, particularly in pairs whose central banks have a documented history of policy discontinuities. It applies to anyone trading extended-hours products where the underlying spot desks close and the broker's own market-making becomes the entire book. It does not apply universally.
The reader who has read this far and still believes their trading style is genuinely inside the Vantage framing's bounds is probably right. The point of the objection is not that the warning is wrong. It is that the warning describes the wrong failure mode for the traders who most need to hear about the right one.