Author: Matthew

A member sent me his statement last month with a one line message: “I don’t understand, I was right.” He had been right, in a sense. Gold had come back to where he first bought. The problem was that by the time it got there he had five positions open instead of one, and the account had been closed out two days earlier. Averaging down in gold trading is the most reasonable sounding way I know to lose an account, and the reason it is so hard to argue somebody out of is that it works most of the time.

That last part is not a figure of speech. I want to show you the arithmetic, because once you see how often it works you understand why nobody believes the warning, and once you see what it costs on the occasion it does not work, you stop needing the warning.

Everything numerical below is my own calculation from assumptions I state as I go. There are no gold price levels here, and nothing in this article is a recommendation to trade.

What averaging down in gold trading actually does to the arithmetic

The tactic is simple. Price goes against the position, so you open another one at the better price. Your average entry improves, so the distance back to break even shrinks. That part is real and it is the whole appeal.

Measure it properly and it is also where the trouble starts. Let me put distance in units of S, your original stop distance, and money in units of R, the loss the first position alone would take at that stop. One equal sized add, placed one S lower, means the combined position turns profitable while price is still below where you first bought. Two adds and it turns profitable a full S below your first entry. Every add makes the recovery look nearer.

Now the other column. Close the whole thing one step past your last add and the loss is not one R any more. One add makes it 3 R. Two adds, 6 R. Three, 10 R. Five adds and a 1 R plan has become a 21 R loss. If you double each add instead of matching it, the same six rows run 1, 4, 11, 26, 57 and 120 R.

That is the trade you are making, stated plainly: you are buying a shorter distance back to break even, and you are paying for it with a loss that grows faster than the number of positions. Nobody would take that deal if it were written on a form. People take it constantly when it arrives one decision at a time.

Why it feels like it works, and why that feeling is accurate

Here is the part that most warnings skip, and skipping it is why the warnings do not land.

Assume, generously, that gold is a coin flip: equally likely to move one step up or one step down. Under that assumption the chance of price touching one step in your favour before running n steps against you is exactly n over n plus one. That is the classic gambler’s ruin result and it is not controversial.

So if your account can absorb four steps against you, the tactic resolves in your favour 80 per cent of the time. With eight steps of room, 88.9 per cent. And that understates it, because once you have added, the combined position turns green well before price returns to your first entry. Simulating the realistic version, where you add at each step down and close as soon as the whole position shows one R of profit, I get a hit rate of 88.9 per cent with four steps of room and 96.0 per cent with eight.

Ninety six per cent. The person doing this is not deluded. They are describing their experience accurately.

The whole game, priced

A hit rate is not a result, though, and this is where it comes apart. I ran 200,000 independent decisions per rule, with a dealing cost of one tenth of a step per position opened, seed fixed at 20261001 so anyone can reproduce it.

The plain rule, one position and out at the stop, wins half the time, the average win is 0.90 R and the average loss is 1.10 R. With four steps of grid room the hit rate climbs to 88.9 per cent, the average win drifts down slightly to 0.83 R, and the average loss becomes 10.40 R. With eight steps of room: 96.0 per cent hit rate, 0.80 R average win, 36.80 R average loss.

Read those three numbers across each row rather than down the column. The hit rate goes up, the win stays roughly the same size, and the loss multiplies. Expected result per decision goes from minus 0.101 R on the plain rule to minus 0.687 R on the widest grid, on a game that was a fair coin flip before costs. The grid does not change your odds of being right. It changes the size of the bill when you are wrong, and it charges you a fresh dealing cost for every position you open on the way down.

Over a year of 200 decisions the deepest hole tells the same story. The plain rule digs a median worst drawdown of 27 R, with a worst case across 20,000 simulated accounts of 85 R. The eight step grid digs a median of 174 R and a worst case of 683 R.

Where the close-out actually lands

There is one more thing the grid does that people do not see coming, and it is not a market risk, it is a contractual one.

Under the ESMA product intervention measures, retail gold CFD leverage is capped at 20:1 and there is a margin close out rule set at 50 per cent of minimum required margin. Your broker is required to close positions when equity falls below that line.

Suppose one R is one per cent of your equity and your stop distance is half a per cent of the gold price. Those two assumptions fix the size of a position: a notional of twice your equity, needing ten per cent of equity as margin at the 20:1 cap. With one position open, the close-out sits 95 steps below your entry, which is so far away it may as well not exist. That is why the close-out feels like something that happens to other people.

Add positions and the line walks toward you. With two open it sits at 45.5 steps. With three, 29.3. With four, 21.5. With five open, the forced close-out is 17.0 steps below your first entry, and the arithmetic from the first section says you are already 15 R down by five steps.

The close-out is not a distant safety net you are drifting toward. It is a line that moves up to meet you every single time you add, and it fires at the broker’s discretion rather than yours. My member’s account did not choose to close. It was closed.

Why the lesson never arrives in time

Put the hit rate and the loss together and you get the thing that makes this so durable.

At 96 per cent, the expected number of decisions before the first failure is about 25. Twenty five straight successes. The probability of ten clean in a row is 66.8 per cent, and of twenty five clean in a row, 36.4 per cent. A trader running a wide grid will very likely have a long, quiet, entirely successful run.

That run is the teacher. By the time the failure arrives, the habit has been reinforced two dozen times, the position sizes have crept up because the tactic has been working, and the account is larger, so the loss is larger too. The lesson always arrives against the biggest position the habit has yet produced. That is not bad luck. It is the structure of the thing.

What I would do instead

I am not going to tell you where to put anything, and none of this is a rule for your account. But the survival logic here is not complicated.

The first thing is to notice that “averaging down” and “scaling into a planned position” are different animals wearing the same coat. If you decided before entering that the position would be built in three parts across a range, and you sized the whole thing so that the full position at its full stop is still one R, that is a plan being executed. If you are adding because price went against you and you did not want to take the loss, that is the grid, whatever you call it in your journal. The test is whether the total risk was decided before the first entry or after the second.

The second is that the number worth protecting is total open risk, not the risk of any one position. Five positions each sized at one R is a five R decision, and the arithmetic above says it is worse than that, because they are all the same trade in the same instrument moving the same way. I have written about how much to risk per trade and about position sizing for gold, and both of those assume the number they produce is the whole exposure rather than the first instalment of it.

The third is the uncomfortable one. The reason people average down is almost never analytical. It is that taking the loss means being wrong in a way that is recorded, and adding means the question stays open. A position you are still holding has not lost yet. That is the appeal, and naming it honestly does more than any amount of arithmetic. If this pattern is familiar, why you keep blowing your trading account covers the same reflex from the other direction.

Frequently asked questions

Is averaging down always wrong?

Not always, and the distinction is whether the total risk was decided in advance. A position planned as three instalments across a range, sized so the whole thing at its full stop is still the loss you budgeted, is a plan. Adding because price moved against you and the loss felt unacceptable is the grid described above, no matter how it is labelled afterwards.

What about averaging down in a long term physical gold holding?

Different question entirely, because the two risks that drive this article are absent. An unleveraged holding has no margin close out that can end the position on someone else’s timetable, and no dealing cost charged per add on the scale a leveraged account pays. Everything here is about leveraged gold positions.

Does it help if I use wider stops and smaller positions instead?

It changes the units rather than the shape. The arithmetic above is written in R and in steps of your own stop distance precisely so it holds whatever those distances are. A grid of five positions is 15 R by the fifth step whether that step is large or small.

Why does the hit rate in your simulation differ from the gambler’s ruin figure?

Because they model different exit rules. The n over n plus one figure is for price touching one step above your original entry. The simulation uses the realistic rule, closing when the combined position shows one R of profit, which after a couple of adds can happen while price is still below your first entry. That is an easier target, so the hit rate is higher. Both are stated so you can see which is which.

Where did the 96 per cent and the 36.80 R come from?

From a simulation of 200,000 independent decisions under the assumptions stated in the article: price moving one step up or down with equal probability, equal sized adds one step apart, a dealing cost of one tenth of a step per position, and a fixed seed of 20261001. Change any assumption and the numbers change, which is the point of writing them down.

My broker offers negative balance protection, so what is the worst case?

Losing the account rather than more than the account, which is a genuine protection and not the one people think they are getting. Negative balance protection caps your loss at what you deposited. It does nothing to make a 36 R loss smaller, and the margin close out will have ended the position long before it becomes relevant.

Where Gold Empire fits

Gold Empire is a free education channel about surviving in gold before trying to grow in it. Everything here is free to read and free to follow on Telegram. The one optional piece is the survival sheet above, a single page on protecting an account. We do not promise profits, we do not publish account results, and we do not believe anyone can tell you what your account will do. If you want the fuller version of the thinking in this article, risk management for gold trading is the piece everything else here hangs off.

About the author. Matthew writes Gold Empire. He spent a long stretch early on treating a losing position as a question that could be kept open by adding to it, and he stopped the year he sat down and worked out what the tactic had actually cost him across all the times it had worked. Nothing in this article is financial advice or a recommendation to trade, and no entry, stop or target discussed should be treated as a signal. Trading leveraged gold carries a substantial risk of loss and is not suitable for everyone.

Sooner or later somebody in the group asks what is hedging in gold trading, and almost always they ask it while holding a position that is down. That timing is the whole story. The question arrives dressed as a technical one, but underneath it is a request for a way to not take the loss yet. I want to answer the technical part properly, and then I want to be straight with you about the other part, because the arithmetic is unkind and almost nobody runs it before they press the button.

Everything below is either quoted from a rulebook you can open yourself, or arithmetic I have shown in full with the assumptions written out. There are no gold prices in this article and nothing here is a recommendation.

What is hedging in gold trading, and what people usually mean by it

In the textbook sense, hedging means holding one position to offset the risk of another, where the two things are genuinely different. A refiner with metal in a vault sells futures against it. A mining company with production coming in nine months locks in a price today. In both cases there is a real exposure sitting outside the trading account, and the market position exists to cancel it.

That is not what the question usually means. What retail traders almost always mean is opening a sell of the same size, in the same instrument, in the same account, while the buy is still open. Long one lot, short one lot, both live. The floating loss stops moving. People call this a hedge. The more honest name, and the one used in most trading rooms, is a lock.

The distinction matters because the two things do different work. A refiner’s hedge cancels a risk that exists elsewhere. A lock cancels nothing. It freezes a number on a screen, and the position it freezes is one you already own the consequences of. You have not removed the loss. You have paid to stop looking at it.

The rule that says the position cannot exist

Here is the part that surprises people, and it is the cleanest evidence that this is not a normal trading technique.

If your account is with a US registered forex dealer, the lock is not allowed. It is not discouraged, not penalised, not a matter of platform settings. It is prohibited by rule. NFA Compliance Rule 2-43, section (b), reads in full:

Forex Dealer Members may not carry offsetting positions in a customer account but must offset them on a first-in, first-out basis. At the customer’s request, an FDM may offset same-size transactions even if there are older transactions of a different size but must offset the transaction against the oldest transaction of that size.

Read that slowly. Offsetting positions may not be carried. If you are long a lot and you sell a lot, the sell does not sit alongside the buy, it closes the buy, oldest first. The rule was adopted effective 15 May 2009 and has been amended four times since, most recently on 15 September 2022, so it is neither new nor an accident of drafting.

A regulator does not write a first in, first out requirement into a rulebook because the alternative is merely inelegant. It writes it because carrying both sides at once obscures what the customer actually owns, doubles the transaction costs the customer pays, and makes the account statement harder to read than the position it describes. The rule exists to stop the thing this article is about.

Plenty of brokers outside that jurisdiction do allow it, and their platforms have a button for it. The button existing is not evidence that the manoeuvre is sound. It is evidence about where the firm is registered, which is a different fact, and one worth checking for its own reasons before you check anything else.

What the lock actually costs, with the assumptions written out

Let me do the arithmetic, because this is the part that gets skipped.

I am going to assume one standard lot of gold is 100 troy ounces, which is the usual contract convention. I am going to assume a spread of 0.30 US dollars an ounce and a combined overnight financing cost across both legs of 0.02 dollars an ounce per session. Both of those are assumptions I have chosen to make the arithmetic concrete, not quotes from any broker, and I will show you further down what happens when you change them. I am also assuming each position pays the spread once across its lifetime, which is the standard way to count a round turn.

One position, opened and later closed, costs 0.30 times 100, which is 30.00 dollars a lot.

A lock is two positions. That is 60.00 dollars a lot before anything else happens. You have doubled your transaction cost, permanently, in exchange for freezing a number.

Then the financing starts. Both legs are open, so both accrue overnight, every session, for as long as the lock exists. At the assumed 0.02 dollars an ounce a session, that is 2.00 dollars a lot a day, and it does not stop because you have stopped thinking about the trade.

Adding the second spread to the accumulated financing gives the total extra cost of the lock, over and above simply having one position. Hold it five sessions and it is 40.00 dollars a lot. Ten sessions, 50.00. Twenty sessions, 70.00. Thirty sessions, 90.00. Sixty sessions, 150.00. Ninety sessions, 210.00 dollars a lot, which is seven times the cost of the single round turn you were trying to avoid paying.

Expressed per ounce, so it is comparable to a price move, the thirty session lock costs 0.900 dollars an ounce and the sixty session lock costs 1.500 dollars an ounce. That is the distance the market has to travel in your favour after you unwind, purely to get you back to where you would have been if you had closed the position on day one and done nothing else.

Now change the assumptions, because yours will be different. At a narrower 0.20 dollar spread and cheap 0.01 dollar financing, a thirty session lock costs 50.00 dollars a lot. At a wider 0.50 dollar spread and 0.04 dollar financing, the same thirty sessions cost 170.00 dollars a lot. Across that range the answer moves by a factor of more than three, and every single version of it is a cost you would not have paid by closing the trade. There is no combination of inputs where the lock is cheaper. That is not a coincidence, it is structural: you are paying to hold two positions where one would do, and the arithmetic cannot come out any other way.

The decision you did not avoid

Cost is the easy objection. The harder one is that the lock does not do the job people believe it does.

When you lock, the loss is frozen but it is not resolved. At some point you have to take one leg off, and the moment you do, you are directional again, with exactly the same question you had before: which way is this going. You have not answered it. You have deferred it, at a price, and you have deferred it to a version of yourself who has now been sitting with the position for weeks and is more invested in being right about it, not less.

Worse, the decision has become harder rather than easier. Before the lock there was one position and one question. After it there are two positions, two entry levels, two financing lines and a choice about which leg to remove first, and the two legs no longer move in step once you start removing one. People who lock frequently describe the same experience, which is that unwinding it well turns out to require better timing than the original trade did.

Against that, consider what closing actually costs. You crystallise the loss, which is the part that hurts, and you pay one more spread when you want to be in the market again. That is 30.00 dollars a lot on the assumptions above. The thirty session lock costs 90.00, which is three times as much, and at the end of it you still have to decide. The loss was going to be the loss either way. The only variable you controlled was how much you paid on top of it, and the lock is the expensive branch.

This is the same logic that runs through everything in risk management in gold trading. The costs that ruin accounts are rarely dramatic. They are small, repeated, and incurred while avoiding a decision. It is worth understanding what swap in gold trading is charging you before you double it, and worth knowing what the spread in gold trading takes on a single position before you commit to paying it twice.

What the position size question looks like underneath

There is one more thing worth saying, because it is the actual problem in most cases where this comes up.

Nobody wants to lock a position that is comfortably sized. The urge appears when the position is large enough that closing it would hurt in a way the trader is not prepared to accept. The lock is not a market judgement at that point. It is a response to sizing, arriving several days too late to help.

For scale, take the afternoon gold auction published by the London Bullion Market Association, in US dollars. Across the 14,690 day to day changes in that record, the middle of the distribution is a move of 0.511 per cent. That is what an ordinary day does. If an ordinary day in a normal market is producing a floating loss you feel you have to freeze rather than accept, the position was too big when it was opened, and no amount of clever position management afterwards fixes a sizing decision made in the past.

The rules in Europe point at the same conclusion from the other direction. The measures agreed by ESMA for retail contracts for difference set leverage at 20:1 for gold, require providers to close out retail positions at 50 per cent of minimum required margin, and provide negative balance protection on a per account basis. Notice what a margin close out rule implies. The regulator has assumed that at some point a retail position will need to be closed rather than managed, and has written the threshold into the rulebook rather than leaving it to the trader in the moment. There is a reason for that, and it is not a flattering one, but it is realistic, and a locked position sitting on both sides of the book does not exempt anyone from it.

Frequently asked questions

Is hedging the same thing as a stop loss?
No, and they are close to opposites in what they commit you to. A stop is a decision made in advance about the point at which you are wrong, and it removes the position when that point arrives. A lock is a decision made in the middle of the trade to not decide, and it keeps both positions. One resolves the exposure, the other preserves it and charges you for the privilege.

My broker allows it, so why would it be a problem?
Whether a platform permits something and whether it serves you are separate questions. US registered forex dealers cannot carry offsetting positions at all under NFA Compliance Rule 2-43(b), quoted above, and the same manoeuvre is freely available elsewhere. That tells you about the regulatory perimeter your account sits inside, not about the merits of the trade. Both facts are worth knowing, and neither is a reason to use the button.

Are there situations where holding both sides genuinely makes sense?
Yes, when there is a real exposure outside the trading account that the market position offsets, which is the textbook case: a producer, a refiner, a jeweller with inventory, a business with a contracted delivery. In each of those the hedge cancels a risk that exists in the physical world. If both your positions are in the same account and there is no outside exposure, there is nothing being cancelled, and the description does not fit however it is labelled on the platform.

What if I lock and then the market moves a long way?
Then the two legs offset and your account equity barely changes, which sounds like the point until you notice that this is true in both directions. You have given up the recovery you were hoping for along with the loss you were avoiding. The frozen position captures nothing, and it accrues financing on both legs the entire time it is frozen.

Where do the numbers in this article come from?
The quoted rule text is from NFA Compliance Rule 2-43 on the National Futures Association’s own rulebook page, linked above. The leverage, margin close out and negative balance protection measures are from the ESMA announcement, also linked. The median daily move of 0.511 per cent is my own calculation across 14,690 day to day changes in the LBMA afternoon gold auction in US dollars, linked above. Every cost figure is my own arithmetic on two assumptions stated in the text, a spread of 0.30 dollars an ounce and combined financing of 0.02 dollars an ounce per session, with a sensitivity range shown. None of them is a quote from a broker, and no gold price level appears anywhere in this article.

Where Gold Empire fits

Gold Empire is free to follow. Daily gold analysis with the reasoning written out, the losing days included rather than quietly dropped, plus an optional Kit for people who want the method in one place. There is no promise of profit here, because nobody honest can make one.

If this article talked you out of one lock, it has more than paid for itself. The habit it belongs to is the boring one: decide the size before you enter, so that the moment where freezing the screen feels necessary never arrives. That sits with risk management in gold trading and with managing a gold trade after you enter, and if you want to know who is writing this, the about page covers it.

About the author

Matthew has spent a long time around gold, most of it learning things the slow way, and now writes them down so other people can learn them the fast way. He has a particular impatience with manoeuvres that feel like action while changing nothing, because those are the ones that quietly empty an account over a year. Nothing in this article is personal advice, and no entry, stop or target discussed should be treated as a signal.

---

This article is educational content about position management, trading costs and the rules that apply to retail forex accounts in different jurisdictions. It is not investment advice, not a recommendation of any broker, platform, jurisdiction or technique, and not a solicitation to trade. The rule text is quoted from NFA Compliance Rule 2-43 published by the National Futures Association, and the retail CFD measures referenced are those agreed by ESMA; rules differ by jurisdiction and by entity and they change over time, so check the rules that actually apply to your own account rather than relying on a general description. All cost figures are arithmetic on assumptions stated in the text and are not quotes from any firm. Trading carries risk of loss.

Most people choose a broker on spread, platform and withdrawal speed, and never once ask what happens if your forex broker goes bust. It is an uncomfortable question because it has nothing to do with trading well. You can have a sound method, disciplined sizing and a year of patient work, and still lose the account to an event that happens in a legal filing rather than on a chart. The good news is that this is one of the few risks in this business you can inspect before you take it, using documents the regulators publish for free.

This article is about where your money actually sits, who stands behind it, and what the published compensation limits genuinely cover. I have used the regulators’ own numbers throughout, and I have shown my arithmetic so you can redo it.

What happens if your forex broker goes bust is a custody question, not a platform question

Three different failures get filed under the same worry, and they have almost nothing in common.

The first is an outage. The platform stops responding, you cannot close a position, and you sit there watching. That is painful and it is operational, but the firm still exists and your balance is still recorded. The second is the firm being unregulated in the first place, where there is no filing to read because nobody required one. The third, the one this article is about, is the firm ceasing to exist as a going concern while it is holding your money.

Only the third raises the question that matters: is the money you can see in your account balance actually yours in a legal sense, held apart from the firm’s own funds, or is it a number on a screen that represents a claim on a company that no longer has assets? Those are very different positions to be in, and the answer is not decided at the moment of failure. It was decided when you opened the account and chose which jurisdiction you were dealing with, which is worth thinking about before how to open a gold trading account becomes a form you click through.

Six firms out of seventy four, which is what the CFTC file actually shows

In the United States, firms that hold retail forex money must file monthly financial reports with the CFTC, and the regulator publishes a summary spreadsheet every month. Anyone can download it. I pulled the file for the month ended 31 July 2026 and counted.

Seventy four firms are listed. Of those, exactly six report any retail forex obligation at all. Their combined retail forex obligation is 471,577,840 US dollars. Across every firm in that same file, customer assets held in segregation total 382,273,801,362 US dollars, so the entire regulated retail forex pool is 0.123 per cent of the customer money sitting in that system.

Sit with that proportion for a moment. The retail forex corner of the regulated American market is a rounding error next to the futures business it shares a rulebook with. It is not that retail forex is small worldwide, it plainly is not. It is that most of the firms advertising to retail traders are not in this file, because they are not registered here. When someone tells you a broker is large, the honest follow up is: large where, and filed with whom.

A firm’s own capital is not a fund that pays you back

The same file reports each firm’s adjusted net capital, which is the regulatory measure of the firm’s own financial cushion. It is tempting to read that as protection money set aside for clients. It is not. It is the firm’s own resources, and comparing it with what the firm owes retail forex customers produces a spread wide enough to be worth seeing.

Set each firm’s adjusted net capital against its retail forex obligation for that month and the ratios run like this. GAIN Capital Group reports an obligation of 204,121,663 dollars against adjusted net capital of 82,904,419, which is 41 per cent. OANDA Corporation reports 132,445,959 against 33,191,483, which is 25 per cent. tastyfx reports 49,950,447 against 67,394,162, which is 135 per cent. Charles Schwab Futures and Forex reports 55,996,755 against 301,491,825, which is 538 per cent. Trading.com Markets reports 2,872,026 against 24,180,804, which is 842 per cent. Interactive Brokers reports 26,190,990 against adjusted net capital of 10,238,479,537, which is roughly 39,092 per cent.

From the lowest to the highest, that is a factor of about 1,560.

Now the important part, because this is exactly where a number like that gets misused. A low ratio is not evidence that a firm is unsafe, and I am not suggesting it is. The ratio mostly reflects what else the business does. A firm whose main activity is retail forex will naturally show a smaller multiple than a diversified broker for whom retail forex is a side line, and both can be perfectly compliant with the capital rules that apply to them. What the spread tells you is something more basic: adjusted net capital is not a client protection pot, and a firm being well capitalised in the regulatory sense is a statement about the firm, not a guarantee attached to your balance. If you want to judge a firm, judge the custody arrangement and the authorisation, which is a different exercise from the one in best broker for gold trading and sits alongside knowing what an ECN account actually changes about execution.

What a compensation scheme promises, and exactly where it stops

Compensation schemes are the part people half remember and badly overestimate. In the United Kingdom the Financial Services Compensation Scheme covers eligible investment claims up to 85,000 pounds per eligible person, per firm, for firms that failed after 1 April 2019. The FSCS states the condition plainly: the PRA or the FCA must have authorised the provider or adviser, as well as regulated the service and product it provided. It may protect you where a provider goes out of business and there is a shortfall in the money or assets it was holding for you, and it does not pay for investments that simply lost value.

Two words in there carry most of the weight. Per firm means the cap is not per account and not per currency. Eligible means the authorisation has to be real and has to cover what you were actually doing.

Separately, across the European Union, the measures agreed by ESMA for retail contracts for difference include negative balance protection on a per account basis, alongside leverage caps and a margin close out rule. Negative balance protection answers a different question from a compensation scheme. It limits what you can lose to the funds in the account while the firm is running. It says nothing about recovering those funds if the firm stops running.

The arithmetic of a fixed cap is worth doing once, properly, because a cap does not scale with you. Assume a single eligible claim against one failed firm, and assume the whole balance is a genuine shortfall. At a balance of 10,000 pounds you are fully covered. At 50,000 pounds you are fully covered. At 85,000 pounds you are fully covered, right up to the line. At 150,000 pounds the cap covers 56.7 per cent and 65,000 pounds sits outside it. At 250,000 pounds it covers 34.0 per cent, leaving 165,000 pounds uncovered. At 500,000 pounds it covers 17.0 per cent, and 415,000 pounds is uncovered.

Which means the protection quietly stops being protection somewhere around the point where the account starts to matter to your life. Splitting changes the arithmetic, because the cap is per firm. The same 250,000 pounds held at one firm is 34.0 per cent covered, at two firms 68.0 per cent, and at three separately authorised firms it is fully covered. The catch is that splitting only works if the firms are genuinely separate authorised entities. Two brands owned by the same legal entity count once, and finding that out afterwards is an expensive way to learn it.

What this does and does not change

It does not mean any firm named here is in difficulty. Every one of them filed the report the rules require, and the figures are theirs. It does not rank brokers, and it is not a recommendation to use or avoid any of them.

The figures are also a snapshot. They describe one month, self reported by the firms, and they will be different next month. A monthly filing is evidence that a supervisory system exists and is producing public data, which is genuinely useful, rather than a live guarantee of solvency.

What it does change is the shape of the question. Custody and authorisation are decided when you open the account, they cost nothing to check, and they are invisible in every spread comparison you will ever read. That puts them in the same family as position sizing: unglamorous decisions made in advance that quietly determine whether an account survives an event nobody scheduled. It is the same logic as everything under risk management in gold trading, applied one level further out, to the firm rather than the trade.

Frequently asked questions

How do I check whether my broker is actually regulated where it claims to be?
Go to the regulator’s own register rather than the broker’s website, and match the exact legal entity name, not the brand. A group can hold a licence through one subsidiary while your account is opened with a different one in another jurisdiction. The entity named on your client agreement is the one that matters, and it is the one you should be searching for.

Does negative balance protection mean my money is safe?
No, and this is the most common mix up. Negative balance protection limits your loss to the funds in the account while the firm is operating normally, which is a real protection against a gap moving against you. It is not a custody arrangement and it is not compensation. It does not help you recover funds from a firm that has failed.

Is a bigger broker automatically safer?
Size tells you about the business, not about your claim. As the figures above show, the relationship between a firm’s own capital and what it owes retail forex customers varies by a factor of over a thousand, and a firm can be large in one activity while its retail forex book is small. Authorisation, custody and the applicable compensation scheme are what determine your position, and none of those is a function of size.

Should I split my account across several brokers?
That is a decision about your own circumstances and I am not going to make it for you. What I can tell you is the arithmetic: a per firm cap means the covered share of a fixed total rises as you split it across separately authorised firms, from 34.0 per cent at one firm to full coverage at three, on a 250,000 pound balance. Against that, more firms means more account admin, more platforms and more places to get something wrong.

Where did the numbers in this article come from?
The firm level figures are from the CFTC’s published Financial Data for FCMs, the FCM webpage update for the month ended 31 July 2026, linked above. Counting the firms, totalling the retail forex obligation, dividing it by total customer assets in segregation, and calculating each capital to obligation ratio are all my own arithmetic on that file. The 85,000 pound limit and its conditions are quoted from the FSCS page linked above. The coverage percentages are simple division under one stated assumption: a single eligible claim against one failed firm where the entire balance is a genuine shortfall. No gold price appears anywhere in this article.

Where Gold Empire fits

Gold Empire is free to follow. Daily gold analysis with the reasoning written out, losing days included, plus an optional Kit for people who want the method in one place. There is no promise of profit here, because nobody honest can make one.

Checking who holds your money is the least exciting thing you will do this year and one of the few that cannot be undone later. It belongs next to risk management in gold trading, and if you want to know who is writing this, the about page covers it.

About the author

Matthew has spent a long time around gold, most of it learning things the slow way, and now writes them down so other people can learn them the fast way. He would rather spend an article on a filing cabinet than on a setup, because the filing cabinet is what decides whether the account is still there to trade. Nothing in this article is personal advice, and no entry, stop or target discussed should be treated as a signal.

---

This article is educational content about broker custody, regulatory capital reporting and compensation schemes. It is not investment advice, not a recommendation of any broker or jurisdiction, not an assessment of the financial condition of any named firm, and not a solicitation to trade. Firm level figures are taken from the monthly Financial Data for FCMs published by the CFTC for the month ended 31 July 2026 and are self reported by the firms concerned; they describe one month and will change. Compensation limits and conditions are quoted from the FSCS, and the retail CFD measures referenced are those agreed by ESMA. Eligibility, limits and protections differ by jurisdiction, by entity and by product, and they change over time, so check the current rules of the scheme that actually applies to your account and read your own client agreement rather than relying on a general description. Trading carries risk of loss.

There is a particular kind of message that arrives in the Gold Empire group after a fast hour, and it always has the same shape. The stop was set. The stop was hit. The account lost more than the stop said it would. Somewhere underneath that message is a question the person has never had to ask before, because nobody warned them it was a question at all: what happens if a market order is not filled at the price you were looking at when you sent it.

The honest answer is that a market order is almost never rejected outright, and that is exactly why the problem catches people. It fills. It just fills somewhere else. That distance is not a glitch, not your broker being dishonest, and not something you can appeal. It is written into the definition of the order type, and most people trading gold have simply never read it. This article is about where the distance comes from, how large it gets, when it gets large, and what it does to an account that was sized on the assumption it would not exist.

What happens if a market order is not filled at your price, in the regulator’s own words

Start with the definition, because the definition contains the whole answer and it is a single sentence. The United States Commodity Futures Trading Commission maintains a public glossary of market terms. Its entry for a market order reads:

An order to buy or sell a futures contract at whatever price is obtainable at the time it is entered in the order book, ring, pit, or other trading platform.

Read the phrase “at whatever price is obtainable” again, slowly. There is no reference to the price on your screen. There is no reference to the price you clicked. A market order is an instruction about urgency, not about price. You are telling the market that you want the trade done now and that you are accepting, in advance, whatever the market charges for doing it now.

Which means the question “what happens if a market order is not filled” is slightly the wrong question, and finding the right one is the useful part. The order will get filled. The thing that can fail is the assumption you attached to it. You assumed the price would be the one you saw. Nothing in the instruction you sent contained that assumption.

People accept this reasonably calmly when it applies to an entry. They do not accept it calmly when they discover where else it applies.

Your stop loss is a market order wearing a different name

Here is the same glossary, a few entries further down, defining a stop order:

This is an order that becomes a market order when a particular price level is reached. A sell stop is placed below the market, a buy stop is placed above the market. Sometimes referred to as stop loss order.

That is the sentence that changes how you should think about your risk. The protective order most traders rely on, the one that lets them say “my risk on this trade is fixed”, is not a promise of a price. It is a trigger that converts into a market order, and a market order takes whatever is obtainable. Your stop defines the moment you leave. It does not define the price at which you leave.

This is not a flaw somebody should fix. It is the only way a stop could work. An order that guaranteed your price would have to refuse to trade whenever your price was unavailable, which is precisely when you most need out. That version exists, it is called a stop limit, and its failure mode is worse: it can decline to fill at all and leave you holding a position you had decided to close. You get to choose which of the two ways your safety equipment is allowed to disappoint you. You do not get to choose neither.

The exchanges treat cascading stops as a known hazard rather than an edge case. The same CFTC glossary documents a feature on CME’s electronic platform called Stop Logic Functionality, which deliberately pauses matching when triggered stop orders would otherwise push the market too far too fast. The mechanism exists because stops firing into thin liquidity is a normal, expected event that markets are engineered around.

Why the fill moves furthest exactly when you need it most

The distance between your level and your fill is not random across time. It concentrates. To show how much, I took the daily gold benchmark published by the London Bullion Market Association, every afternoon auction in US dollars from 1 January 1975 to 24 September 2026, which is 12,979 day to day changes. I measured each day’s absolute move, added them all together, and asked how that total was distributed.

The busiest one percent of days, 129 sessions out of 12,979, carry 7.4 percent of all the movement gold has made in fifty one years. The busiest five percent carry 22.7 percent. The busiest ten percent carry 35.7 percent. The busiest quarter of all days carry 61.1 percent, which leaves the remaining three quarters of the calendar sharing 38.9 percent between them.

The median day moves 0.533 percent. Days that move two percent or more happen 7.85 percent of the time, and when they happen their average size is 3.10 percent, which is 5.8 times the median day.

Now put the two facts side by side. A stop is a market order, so its fill quality depends on how fast price is travelling when it triggers. And gold’s travelling is concentrated into a small number of days. Your stop does not trigger on a random day. It triggers disproportionately on the days that are moving, because those are the days that reach it. The distribution of days on which your protection is tested is not the distribution of ordinary days. It is skewed towards exactly the sessions where “whatever price is obtainable” is furthest from what you were looking at.

This is why comparing your average fill to your expected fill across a calm month tells you almost nothing about the month that matters.

What the distance costs, counted in planned risks

Say your stop sits one planned risk away from your entry. Call that distance 1R, which is simply the amount you decided in advance you were willing to lose on the trade. If the fill arrives past your level, the realised loss is no longer 1R. Here is the arithmetic, and the only assumption in it is that your winners pay two planned risks:

• Fill 10 percent of the stop distance beyond your level, and the loss is 1.10R. The winners needed to recover it rise by 10 percent.
• Fill 25 percent beyond, and the loss is 1.25R. Recovery cost rises 25 percent.
• Fill 50 percent beyond, and the loss is 1.50R. Recovery cost rises 50 percent.
• Fill a full stop distance beyond, and the loss is 2.00R. You now need a whole winner to undo one loser, where you had planned on half of one.

The number that should get your attention is not any single row. It is that this tax is charged on losing trades only. Slippage on an entry can help you as often as it hurts. Slippage on a stop is asymmetric by construction, because the stop only triggers when the trade is going against you, and going against you fast is precisely the condition that widens the fill. A cost that only ever lands on your worst days compounds differently from a cost spread evenly across all of them.

Notice what this does to position sizing. If you size so that a 1R loss is a survivable fraction of the account, and the real distribution of your losses has a tail at 1.5R and 2R, you have sized for a distribution you do not have. The gap is invisible in a quiet month and expensive in the month that is not.

The four ways a market order actually goes wrong

Rejection is real but rare, and the four failure modes deserve separating, because the response to each is different.

It fills away from your price. The common one, described above. Not an error, not appealable, and the only one that reliably costs money on stops.

It is rejected for insufficient margin. The account lacks free margin at the moment the order arrives, usually because something else moved against you in the meantime. This one is silent in a way the others are not, because you may believe you are in a trade that you are not in.

It is rejected because the instrument is not trading. Weekend, a scheduled break, or a venue level halt. Your order waits or dies depending on the platform, and the price on the other side of the pause is whatever the pause produces.

It is requoted. Some venues answer a market order during fast conditions with a new price and a request to confirm. In the seconds you spend deciding, the market keeps moving.

The first is a cost. The other three are a different problem, which is believing you are protected when you are not. That is why “did my order actually execute” belongs on the same short checklist as “where is my stop”, especially after any minute where price moved quickly.

The regulatory backstops define the floor rather than the fill. When European regulators intervened in retail contracts for difference, the measures included a margin close out rule and negative balance protection, both applied per account, alongside leverage caps. The same ESMA announcement recorded that between 74 and 89 percent of retail accounts lost money on these products. Negative balance protection bounds the worst case at the account. It does not mean your stop filled where you wanted.

What to do about it without pretending you can control it

You cannot control the fill. You can control four things around it, and all four are decisions made before the market is moving.

The first is to stop treating 1R as a ceiling and start treating it as an estimate. If your sizing only survives when every loss is exactly the planned size, it is not sizing, it is a hope.

The second is to measure your own fills instead of assuming them. Every platform shows the price you asked for and the price you received. Two columns in whatever record you already keep will tell you more about your real cost of trading than any general article can, including this one. If you do not have that record yet, how to track my trades covers what is worth writing down.

The third is to know which sessions you are asking your stop to work in. Holding through a scheduled release means holding through the part of the day where the distance is widest, which is the subject of how to trade gold through high impact news. That is not an instruction to avoid news, it is an instruction to know the trade you are making.

The fourth is to choose the level before the market is in front of you, so that when the fill arrives past it you are dealing only with the distance and not also with the decision. Where to place a stop loss on XAUUSD deals with that half.

None of this makes the fill better. It makes the account able to absorb a fill that is worse than planned, which is the only version of this problem that has a solution.

Frequently asked questions

What happens if a market order is not filled at all?
Outright rejection is uncommon and has specific causes: not enough free margin, the instrument not currently trading, or a platform level refusal during fast conditions. In each case no position is created, which matters most when the order was meant to close a trade rather than open one. Check the order status rather than assuming, because a rejected closing order leaves you exposed while you believe you are flat.

Is a fill away from my price the same thing as my broker cheating me?
Usually no. A market order is defined as accepting whatever price is obtainable, so a fill away from your screen price is the order working as specified. What would be worth investigating is a pattern that only ever runs one way across many trades in ordinary conditions. Without your own record you have no way to tell an ordinary cost from an unusual one.

Does a stop loss guarantee my maximum loss?
No, and this is the single most expensive misunderstanding in retail trading. A stop becomes a market order when touched, so it fixes the moment of exit, not the price of exit. The loss can be larger than planned, and it is largest in exactly the fast conditions that triggered it. Size on the assumption that some losses will exceed the plan.

How is this different from a limit order not filling?
Completely different, and in your favour. A limit order that does not fill costs nothing at all, because there is no trade: no spread, no commission, no exposure, as covered in what happens if a limit order is not filled. A market order that fills badly has already cost you money. One is a missed opportunity, the other is a realised expense.

What about an order that only partly fills?
That is a third case again, with its own consequences for the risk you are actually carrying, and what happens if an order is partially filled works through it. The short version is that a partial fill quietly changes your position size, so the risk you calculated is no longer the risk you have.

Would a stop limit order solve this?
It solves the price problem by creating a worse one. A stop limit will not fill beyond the limit you set, which sounds like protection until the market moves past your band without trading inside it. Then the order does nothing and you still hold the position. You are choosing between a bounded exit at an uncertain price and an uncertain exit at a bounded price.

Where did the numbers in this article come from?
The two definitions are quoted from the CFTC glossary, linked above. The movement statistics are my own arithmetic on the LBMA afternoon gold auction in US dollars, also linked, using every observation from 1 January 1975 to 24 September 2026, which is 12,979 daily changes. Movement is measured as the absolute logarithmic change between consecutive auctions, then summed. The stated limitation is that one auction price per day cannot see what happened inside the day, so these figures describe the shape of the distribution rather than the size of any individual fill. The R arithmetic is straightforward division under the single stated assumption that winners pay two planned risks. No gold price level appears anywhere in this article.

Where Gold Empire fits

Gold Empire is free to follow. Daily gold analysis with the reasoning written out, losing days included, plus an optional Kit for people who want the method in one place. There is no promise of profit here, because nobody honest can make one.

Knowing that your protective order is an estimate rather than a guarantee is a sizing question before it is anything else, which is why this sits under risk management in gold trading. If you are still setting the account up, how to open a gold trading account covers the ground before any of this applies, and you can read more about who writes this on the about page.

About the author

Matthew has spent a long time around gold, most of it learning things the slow way, and now writes them down so other people can learn them the fast way. He would rather explain the mechanism that quietly widens a loss than the setup that fills a timeline, because the first one is what decides whether an account is still here next year. Nothing in this article is personal advice, and no entry, stop or target discussed should be treated as a signal.

---

This article is educational content about order types and execution mechanics. It is not investment advice, not a recommendation of any broker, platform or order strategy, and not a solicitation to trade. The order definitions are quoted from the public glossary of the CFTC. The price series is the daily gold benchmark published by the LBMA, and every statistic drawn from it is my own arithmetic under the assumptions written out above, including the limitation that a single daily auction price cannot observe intraday execution. Historical frequencies describe what has already happened and are not a forecast of what any future order will do. Order handling, rejection rules, requote policy and margin treatment differ between venues, so check your own platform’s documentation and client agreement rather than relying on a general description. Trading carries risk of loss.