The Drawdown You Did Not Choose
When people ask me how to reduce drawdown in trading, they usually expect an answer about discipline. Cut losers faster. Stop revenge trading. Be patient. All fine advice, and none of it explains the drawdowns I have watched hurt people most.
The worst ones did not come from a discipline failure. They came from a trader who was sized correctly on every single position, followed every rule, and still lost far more than planned, because the positions were never really separate. They looked like three careful decisions. The market treated them as one.
That gap between how many trades you have open and how many bets you actually hold is, in my experience, the largest uncounted risk in a retail account. So let us count it.
How to Reduce Drawdown in Trading by Counting Your Real Bets
Suppose you hold three positions of equal size. It feels like you have divided your risk three ways. Whether you have depends entirely on one number: how closely those positions move together.
The arithmetic is standard. For n positions of equal size with an average correlation r between them, the combined risk scales as the square root of n plus n times n minus one times r. And the number of genuinely independent bets you hold is n divided by one plus n minus one times r.
Run it for three positions:
At zero correlation, three positions carry 1.73 times the risk of one, and you hold three real bets. That is genuine diversification, and it is the picture most traders have in their head.
At 0.9 correlation, the same three positions carry 2.90 times the risk of one, and you hold 1.07 real bets. You have not spread your risk at all. You have taken a single position at nearly triple size, while telling yourself you were being careful.
You did not open three trades. You opened one trade, three times, and paid three spreads for the privilege.
Now ask the uncomfortable question. If you trade gold, and you are long XAU/USD, and you are also long silver, and you also have a position that is really a short-dollar expression, what do you suppose the correlation between those is on a day when the dollar moves hard? It is not zero. On the days it matters most, when everything is moving together, it tends toward one, which is exactly when correlation is least helpful to you.
The World Gold Council, whose entire remit is making the case for gold in a portfolio, still states the caveat plainly in its research on gold as a strategic asset: diversification "does not guarantee any investment returns and does not eliminate the risk of loss". If the people arguing for diversification say that, it is worth taking seriously.
Three Practical Ways to Cut It
One: set a total exposure ceiling, not just a per trade one. Most traders have a rule for a single position and no rule at all for the sum. If your per trade risk is 1 percent, decide in advance what the maximum combined risk is when several positions are open at once, and treat correlated positions as one position for that purpose. Two gold trades in the same direction are one gold trade at double size. Size them accordingly, or do not open the second one.
Two: count exposure by driver, not by ticker. The useful question is not "how many trades do I have open" but "how many separate reasons am I betting on". If every open position profits from the same thing happening, you have one reason and one bet. Writing the reason next to each trade makes this visible in about ten seconds, and it is the fastest audit in this article.
Three: reduce size when conditions are unusually unstable. Correlation rises in stress. That is the awkward property of it: the diversification you were relying on evaporates precisely when you need it. When the market is behaving erratically, false breakouts everywhere and no level holding, the honest response is smaller size or no position, not the same size with more conviction. Standing aside is a legitimate exposure decision, and I have argued that case at length in why the best trades are the ones you do not take.
It Gets Worse With More Positions, Not Better
The instinct when a drawdown hurts is often to spread wider. More positions, smaller each, surely that is safer. It is, but only if the additions are genuinely different bets, and the arithmetic punishes you sharply when they are not.
Take five equally sized positions instead of three:
- At zero correlation: combined risk 2.24 times a single position, and 5.00 independent bets. Textbook diversification.
- At 0.5 correlation: risk 3.87, and only 1.67 real bets.
- At 0.9 correlation: risk 4.80, and 1.09 real bets.
Read the last line carefully, because it is the trap in its purest form. Five positions at high correlation give you 4.80 units of risk, almost the full 5.00 you would have from simply quintupling one trade, in exchange for essentially one bet. You have taken on nearly all the risk of concentration and received almost none of the benefit of spreading, and you have paid five spreads to arrange it.
Notice also how quickly the benefit disappears. Going from correlation zero to 0.5 does not halve your diversification, it cuts five real bets down to 1.67. Correlation does not need to be extreme to do most of its damage, which is why "well, they are not exactly the same trade" is not the reassurance it sounds like.
Estimating Correlation Without Doing Statistics
You do not need to compute a correlation coefficient to use any of this. Three questions get you close enough to act on, and all three take seconds.
Would one piece of news move all of these at once? If a single dollar print, one central bank line, or one risk-off headline would hit every open position in the same direction, your correlation is high whatever the instruments are called. This is the fastest test and it catches the majority of cases.
Are they the same direction on the same underlying driver? Long gold and short dollar are frequently the same trade wearing two hats. Long gold and long silver are usually the same trade with different volatility. Neither is wrong to hold, but both should be sized as one.
Have they moved together in your own records? Your journal already contains the answer. If your losing days tend to be days when several positions lost at once rather than one at a time, you have been running correlated exposure for a while without labelling it. That pattern is visible in a few minutes of scrolling, and it is more convincing than any number I could give you.
Where the honest answer to any of these is yes, treat the group as a single position for sizing. Not because the correlation is exactly one, but because assuming it is closer to one than you would like is the error that leaves you solvent.
Why Shallow Drawdowns Matter More Than They Look
There is a second reason to care, beyond the money.
A drawdown is not only a financial event, it is a decision-making event. The deeper it gets, the worse the decisions made inside it become, and the more tempted a trader is to size up to fix it quickly. That is the mechanism that turns a bad week into a closed account, and it is why keeping the worst stretch shallow is worth more than it appears on a spreadsheet.
Which is also why this article is about exposure rather than willpower. Willpower is what you are asking of yourself at the exact moment you have least of it. An exposure ceiling written down while calm does the work for you, without needing you to be at your best. The same logic sits underneath position sizing so one trade cannot hurt you and underneath the whole protect your capital approach.
Reducing drawdown is not about predicting which trades will lose. You cannot. It is about ensuring that when several of them lose together, and they will, the total is a number you chose in advance rather than one you discover afterwards.
Frequently Asked Questions
What actually causes large drawdowns?
Usually correlated exposure rather than a single bad trade. Three equally sized positions at 0.9 correlation carry 2.90 times the risk of one position and amount to only 1.07 independent bets, so a move against you hits all of them at once. Most traders have a per trade risk rule and no rule for the combined total.
How do I know if my positions are correlated?
Write the reason next to each open trade. If every position profits from the same thing happening, they are one bet regardless of what the instruments are called. That ten second check catches most of it without any statistics.
Does trading more instruments reduce my drawdown?
Only if those instruments move independently, and in stressed conditions they typically move together. Adding positions that share a driver increases total exposure while creating the impression of spreading risk, which is the worst combination.
What is a reasonable maximum drawdown to accept?
That is your decision and it depends on circumstances I cannot see. What can be said is that recovery gets disproportionately harder as it deepens, so the practical value of a ceiling comes from setting it in writing while calm and treating it as binding rather than advisory.
Should I stop trading during a drawdown?
Reducing size during unstable conditions is a defensible exposure decision rather than an admission of defeat, particularly because correlation rises in stress and your diversification is weakest exactly then. Deciding the trigger in advance is what stops it becoming an emotional call.
Where did these numbers come from?
I calculated them in Python from the standard portfolio risk formula for equally sized positions, with the correlation assumptions stated in the text so you can change them. The quotation on diversification is from the World Gold Council research linked above.
A Word on Risk, and How to Use This
Everything above is general education about exposure and probability in leveraged markets. Nothing here is financial advice, and no entry, stop or target discussed should be treated as a signal. Trading gold and other leveraged products carries a high risk of losing money quickly.
Cut to the bone: how to reduce drawdown in trading is mostly a counting exercise. Count bets, not tickets. Cap the total, not just the single trade. Do it in writing, while calm.
If you want the risk-first companion to this way of working, it is the Black Gold Market Blueprint, a plain walk-through of protecting an account before trying to grow one. It is free, there is nothing to join, and there is no promised return anywhere in it.
Grab the Blueprint here, and for the foundation underneath all of it, start with how to protect your capital when gold gets volatile.
Protect. Master. Grow.
About the Author
Raphael, founder of Black Gold Market
Raphael runs a XAU/USD channel built on one idea: the level, the context, and the risk, stated before the trade rather than after it. More about how the channel works. It is free to follow, with an optional Kit; he does not sell certainty and does not publish profit claims.
Risk disclaimer: This article is for educational purposes only and is not financial advice, an offer, or a recommendation to buy or sell any instrument. Trading gold, CFDs and leveraged products carries a high risk of rapid loss. No entry, stop or target discussed should be treated as a signal. External figures are linked to their source, and calculated figures are shown with their assumptions so you can check them yourself.