Nobody rings you: what a margin call actually looks like in 2026 

A check runs against your equity every few seconds, all day and all night. By the time anybody tells you anything, the positions have usually already been closed. 

What is margin, exactly? 

The deposit you put up to open a leveraged position, expressed as a fraction of the position’s full value. The regulatory ceiling for a UK retail client is 30:1 on the major currency pairs, which puts the deposit at roughly 3.3% of the notional. Take a GBP 100,000 EUR/USD position and you commit around GBP 3,300. 

The important word is deposit. The GBP 3,300 stays yours; the broker holds it as collateral and returns it when the position closes, which is why it never appears on a cost calculation. What it buys is exposure to GBP 100,000 of price movement, and both the gains and the losses are calculated on the larger number. 

That asymmetry between what you committed and what you are exposed to is the whole of leverage, and it is where every margin call originates. 

What triggers the sequence? 

Losses eating into your equity. As the position moves against you, your account equity falls while the margin requirement stays roughly constant, so the ratio between them deteriorates. Brokers monitor that ratio continuously and act at defined thresholds. 

Under FCA rules the hard threshold is 50%. When your equity falls to half of the required margin, the broker must begin closing positions. This is the margin close-out rule, it is automatic, and it applies to every FCA-regulated retail account. The broker has no discretion in the matter and no interest in exercising any: the rule requires it to act, and it acts. 

The FCA close-out sequence for a retail account. Timings are not fixed; a fast move can take you from stage one to stage four inside a minute. 

How much warning do you get? 

In a normal market, some. Most platforms send a notification as equity approaches the threshold, and if you are watching you can add funds or close part of the position yourself. In a fast market, potentially none that is any use to you. The sequence runs at machine speed and a gap can carry price straight through the warning level to the close-out level. 

This is why the overnight and weekend cases are the dangerous ones. A position held through a Sunday reopen can gap past every intermediate stage while you are asleep, and the first you know of it is a closed position and a realised loss. The notification will be there. It will have arrived at four in the morning. 

Why the close-out fires at the worst price 

There is a structural nastiness in the timing of all this, and it is not an accident of implementation. The close-out fires precisely when the market has travelled furthest against you, because travelling furthest against you is what triggers it, so by construction it sells at close to the worst price of the whole episode. A position closed at the 50% level that then recovers within the hour is one of the most infuriating things that happens in retail trading, and it is the rule working exactly as designed rather than a platform malfunction. Complaints about it almost never go anywhere, and the reason is that nothing went wrong. 

What is negative balance protection? 

The backstop that stops the account going below zero. For retail clients at FCA-regulated firms, you cannot lose more than you deposited, and if a violent move takes the account negative before the close-out completes, the firm absorbs the shortfall rather than billing you for it. 

It is a genuinely significant protection and it was not always there. It also has a narrower scope than people assume: it caps your loss at everything you put in, which is a meaningful cap and still a total loss. Protection from owing more is not protection from losing all of it. 

Where the word protection runs out 

Hold that distinction when you meet the close-out rule described as a consumer protection, which it is. It stops a position running indefinitely and, with negative balance protection behind it, prevents a debt. What it does not do at any point is prevent a loss. It fixes the loss at a defined level, which is better than the alternative and a long way short of safe, and any account of these rules that stops at the word protection has told you half of what they do. 

Does adding funds to hold a position ever make sense? 

Occasionally, and it is one of the most reliably expensive decisions in retail trading. 

Topping up an account to avoid a close-out is a decision to increase your exposure to a position that is already losing, made under time pressure, usually by someone who does not want to be wrong. The base rates on that are not good. 

There are legitimate cases, mainly where the position is part of a planned structure and the margin pressure is a mechanical consequence rather than a signal. But the honest version of the advice is that if you are deciding at speed, in response to a warning notification, on a position you did not size properly, adding funds is usually the expensive option dressed as the brave one. 

How do you avoid the whole scenario? 

Size positions so that a routine adverse move does not approach the threshold, which in practice means using far less of the available leverage than the 30:1 cap permits. The cap is a regulatory maximum, not a recommendation, and treating it as a target is how accounts end up at stage three. 

Beyond that: know where your close-out level sits before you open, not after; account for financing accruing nightly against you, which erodes equity even in a flat market; and be aware of scheduled events that fall inside your intended holding period. The Investors Centre approaches the platform side of that from the account outwards. It opens and funds live accounts with its own money to test UK trading platforms, rather than compiling rankings from providers’ published fee schedules. 

The part nobody has tested 

Even so, there is a hole in the evidence underneath all of this, and it is worth naming. Nobody researching platforms deliberately runs an account into a close-out to see what happens, because it means destroying real money to produce a paragraph. So every published account of the last thirty seconds, including this one, is assembled from the rulebook, from platform documentation and from behaviour observed well short of the threshold. Funded testing is genuinely good at what a platform charges and how quickly it pays you out. It has nothing to say about what your screen does at the moment it matters most, and any description of a margin call that claims otherwise is describing something that did not happen. 

What a closed position leaves behind 

Statements first. A close-out is a normal sell order as far as the record is concerned, so it appears at the price it filled at rather than the price on the chart when you look afterwards, and on a fast move those can differ noticeably. Financing is charged up to the point of closure, so a position held for a fortnight and closed on the fifteenth night has fourteen nights of accrual sitting inside the realised figure. Reconcile both before deciding what the episode actually cost, because the headline loss on the platform’s summary screen and the arithmetic of the trade are rarely the same number. 

Then the account itself. Whatever remains is free margin again, which is the point at which people reopen at the same size and repeat the whole sequence with less capital behind it. If you do nothing else, work out what the surviving balance can now support at a sensible fraction of the 30:1 cap, and write that figure down before the next position rather than during it.