Concentration risk is the gap between how many positions your trade list shows and how many bets your account is actually making. For most of us trading gold, that gap is enormous, and it is the reason a "normal" day occasionally takes a bite out of the business that no single trade was ever supposed to be able to take.
Every business I have ever run had a version of this problem. In my first one, I had eleven customers and felt diversified right up until I noticed that four of them were subsidiaries of the same parent company. I did not have eleven customers. I had eight, and one of them was very large. Nothing about my invoice list said so.
Your trade list does the same thing to you.
What Concentration Risk Means for a Trading Business
In any business, concentration risk is what you carry when too much of your outcome depends on one thing. One customer, one supplier, one product, one market. The business looks busy and varied from the inside, and then the one thing moves and everything moves with it.
A trading account is the purest version of this I have ever seen. Most of us here trade one instrument. Gold. That is one product, sold into one market, priced by one set of forces. There is no second product line to carry a bad quarter.
That alone is worth knowing, and it is not a criticism. Specialising in one market is a perfectly sound business decision, and I have argued for it. Depth in one instrument beats shallow familiarity with nine. But a specialist business has to be run differently from a diversified one, and most retail traders run theirs as though the diversification is there.
The Trade List Lies to You
Here is where it gets concrete. You have three positions open. Three rows on the platform, three tickets, three separate stop levels. It feels like three trades, and your brain does the natural thing: it assumes the risks are spread across three independent outcomes.
They are not independent. They are the same instrument, usually in the same direction, taken for reasons that came from the same chart in the same hour. If gold turns against you, it does not turn against one of them. It turns against all three at once, because they are all the same bet wearing three tickets.
The platform will never tell you this. It shows you rows. Rows look like diversification.
The Arithmetic, With the Assumptions Written Out
I ran the numbers, and I want to be open about the assumptions so you can push back on them. Assume three open trades. Assume each risks exactly 1R. Assume each is a straight 50/50, win 1R or lose 1R. Ignore costs. These are simplifications, and the conclusion does not depend on them being exact.
If those three trades were genuinely independent, the chance of all three losing is one half cubed, which is 12.5 percent. Unpleasant, and rare enough that you would treat a full 3R day as a bad-luck outlier.
Now make them what they actually are: the same market, the same direction. There is only one outcome to be had. Either the market goes your way or it does not. The chance of losing the full 3R is no longer 12.5 percent. It is 50 percent, the same as the chance of a single trade losing.
The 3R day did not become slightly more likely. It became four times more likely, and nothing on your screen changed.
There is a second way to see the same thing that I find even more useful. Measure the swing of the day rather than the worst case. Three independent 1R bets have a standard deviation of about 1.73R, because independent risks add up along a square root rather than in a straight line. Three perfectly correlated 1R bets have a standard deviation of exactly 3R.
Work backwards from that and you get the line I keep coming back to. To find how many independent trades would swing as hard as three correlated ones, you solve for the number whose square root is three. The answer is nine. A three trade day in one instrument moves your account like a nine trade day would if the trades were genuinely separate.
It scales worse than most people expect. At two open trades, the full loss is twice as likely as the independent case. At three, four times. At four, eight times. At five, sixteen times. Every position you add in the same market roughly doubles the odds of the worst case, while your trade list just grows by one tidy row.
Why This Is a Business Problem, Not a Trading One
The reason I frame this as concentration risk rather than "don't overtrade" is that the underlying idea is older and better established than anything in retail trading, and it did not come from a trading educator.
Harry Markowitz published Portfolio Selection in the Journal of Finance in 1952 (volume 7, issue 1, pages 77 to 91, doi:10.1111/j.1540-6261.1952.tb01525.x). Its central insight is the one every operator needs and most retail traders skip: the risk of a collection of holdings is not the sum of their individual risks. It depends on how they move together. Two holdings that move together are, for risk purposes, closer to one holding at double the size.
That is a seventy year old finding, it underpins how every serious institution measures exposure, and it applies with full force to three gold tickets on a retail platform.
It is worth adding the blunt regulatory backdrop too. When European regulators restricted leveraged retail products in 2018, they published their reasoning: analyses across EU jurisdictions found that 74 to 89 percent of retail accounts typically lose money, with average losses per client ranging from 1,600 to 29,000 euros. Accounts do not usually die from one carefully sized trade. They die from several at once that turned out to be the same trade.
The Four Places Concentration Hides
When I audit my own exposure, I look for four overlaps, in this order.
Same instrument. The obvious one. Three gold positions are one gold position. If you only trade gold, treat total open risk as a single number rather than a list, and set a ceiling on that number the way you would set a credit limit for a customer.
Same direction. Long gold and long gold are the same bet. Long gold and short gold are not diversification either, they are mostly a hedge that pays you nothing while costing you spread on both sides. This is worth thinking through rather than assuming either one solves the problem.
Same reason. This is the one people miss. Two positions in different instruments can still be one bet if you took them for the same underlying reason. If you are long gold and short the dollar because you expect the same rate move, that is one macro position with two tickets on it. Different symbols on the screen, one thesis in your head.
Same time. Positions taken in the same hour tend to share whatever you believed in that hour, including your mood. A cluster of trades opened within twenty minutes of each other is usually one decision, repeated.
What an Operator Actually Does About It
The fix is not exotic, and it is not "go and trade more markets". Adding instruments you do not understand in order to look diversified is a worse problem than the one you started with.
What an operator does is measure total exposure rather than counting tickets. One number: how much of the account is at risk right now if the market goes against the thesis all these positions share. That number, not the number of rows, is what you cap. If you have built a plan using the one page trading business plan, this belongs in it as a hard line, and it should be written down before the day starts rather than negotiated at the moment you want a fourth position.
The practical consequence for most people is smaller individual size when several positions are open, or simply fewer positions. If your rule is a fixed risk per trade, and you keep three open at once, then your real risk per decision is three times what you told yourself it was. The arithmetic in position sizing for gold trading only protects you if the sizes are not stacking behind your back.
The second thing an operator does is hold a reserve that is genuinely sized for the correlated case, not the independent one. If your worst realistic day is 3R rather than the 1R you had in mind, then the cash reserve has to be built for that day. This is the same logic as surviving a market shock, applied to an ordinary Tuesday rather than a crisis.
The third is to record it. Add a column to your journal for total open risk at the moment of each entry. Within a month you will see whether your worst days are single bad trades or clusters. In my experience it is almost always clusters, and until you track it you will keep filing those days under bad luck. Your key metrics should include the shape of your losses, not just their size.
What This Does Not Mean
A few things I am not saying, because this topic invites overcorrection.
I am not saying you should never hold more than one position. Sometimes scaling into a thesis is a reasonable thing to do, as long as you are honest that it is one thesis and you size the total accordingly.
I am not saying specialising in gold is a mistake. It is one of the better decisions a small operator can make. It just means the concentration is a permanent feature of your business, so it has to be managed permanently rather than noticed occasionally.
And I am not saying any of this predicts what gold will do. Nothing here tells you whether a position will work. It tells you how much of your business is riding on the same answer, which is a different and more controllable question. It also has nothing to say about where to enter or exit, and it should not be read that way. Setting a sensible ceiling on total exposure is part of managing risk in gold trading, not a method for finding trades.
Frequently Asked Questions
Is concentration risk just another way of saying overtrading?
No, although they often show up together. Overtrading is about how often you act. Concentration risk is about how much your outcomes depend on the same thing, and you can carry a lot of it while taking very few trades. Three carefully considered positions in one market is not overtrading, and it is still one bet at triple size.
Does hedging with an opposite position solve it?
It changes the problem rather than removing it. An opposing position reduces directional exposure, but you now pay costs on both sides and you have two decisions to get right instead of one. Whether that is worth it depends on your situation, and it is not the simple neutraliser it is often presented as.
How do I calculate my total open risk?
Add up what you would lose across all open positions if every one of them hit its stop. That single number is your real exposure. Compare it to the risk figure you believe you are running per trade, and if the total is several times larger, you have found the gap this article is about.
Should I trade other markets to diversify?
Only if you would trade them on their own merits and you understand them properly. Adding an unfamiliar instrument to reduce a number on a spreadsheet usually introduces more risk than it removes. Reducing total size is the simpler and more reliable lever, and it does not require you to learn a new market.
Where did the 12.5 percent and 50 percent figures come from?
I calculated them from the assumptions stated above: three trades, each risking 1R, each a 50/50 outcome, costs ignored. They illustrate how the mathematics behaves rather than measuring anyone's actual results. Change the assumptions and the exact figures move, but the direction of the finding does not.
Does this apply if my positions have different stop distances?
Yes, and it is why measuring in money rather than in tickets matters so much. Different stops mean different amounts at risk per position, so counting rows tells you even less. Add up the money, not the trades.
What I Would Want You to Take Away
Count exposure, not tickets.
That is the whole thing. Your platform is built to show you rows because rows are what it has to show. Your business needs one number, and that number is how much of the account is riding on the same answer right now.
The reason I keep hammering the business framing is that this particular mistake is invisible from inside the trade. Every individual position can be correctly sized, every stop can be sensible, every entry can be defensible, and the day can still take 3R out of you because all three were the same trade. There is no error to spot at the level of any single decision. It only appears when you look at the portfolio, which is exactly the view a business owner is supposed to take and a trader usually does not.
If you want the structure that makes this a habit rather than a good intention, the free one page trading business plan has a line for total exposure. Fill it in before the market opens, when the number is theoretical and easy to be honest about. Deciding your ceiling while you are staring at a fourth setup you like is not a decision, it is a negotiation, and you already know who wins it.
About Rex
I'm Rex. Before I ever placed a trade I spent five years running a real business, and the lesson that transferred hardest was about measurement: for a whole year I tracked units sold and congratulated myself, while the number that decided whether I made rent sat in a column I was not looking at. Trading gave me the chance to make the identical mistake a second time, in a different unit, and I took it. Counting distance is comfortable because distance is public. Counting money is uncomfortable because money is yours. More about how I run the channel.
Today I run the REX Trading Signal channel, around 11,900 people, on three rules I don't break: every setup carries a stop loss; I post the losing trades, not only the winners; and I never promise profit, no "guaranteed," no "fixed," no "risk-free." The pip results we post describe a move, and nothing more. What that move is worth is decided by your size and your risk, which is your side of the desk and nobody else's. Nothing here is financial advice, and no entry, stop or target discussed should be treated as a signal.