Is forex a zero-sum game?
The accounting is, trade by trade. The intent is not. 42% of volume is funding swaps and 2025’s growth came from hedging — but among those who speculate, once costs are in, the game is negative-sum.
No, and the mistake is conflating two different things. The accounting is zero-sum trade by trade: what one side gains, the other loses. The intent is not. 42% of all volume is funding swaps and the growth of 2025 came from hedging — people who are not trying to guess the price. And among those who do speculate, once costs are in, the game is negative-sum.
What does "zero-sum" actually mean?
A zero-sum game is one where a participant's gain comes exactly at another's expense. The CME defines it that way in its glossary, and in a futures contract it is literal: if I buy a contract and you sell it, at settlement I gain X and you lose X. Sum: zero.
That identity is true and not up for debate. The problem is what people infer from it:
"For you to win, somebody has to lose. Therefore the market is a war of all against all, and you are the mark."
The first sentence is accounting. The second is a story about other people's intentions. And the data does not support that story.
Why doesn't forex fit that mould?
| Futures | Spot forex | |
|---|---|---|
| Structure | centralised market | OTC and fragmented |
| Who is opposite you | a long and a short, matched | a dealer against inventory |
| After your trade | nothing left to do | netted against twenty other clients |
| The accounting identity | clean and visible | true, but spread out |
There is no common book where your order is matched against another individual's. There is no "other side" in the singular.
Then there is size. According to the BIS Triennial Survey, the FX market turned over roughly $9.6 trillion a day in April 2025, 28% more than in 2022. No other financial market comes close. Asking who is inside that number is more useful than assuming.
Who is really on the other side of your trade?
Here the data delivers a surprise, and not the one usually told.
The standard argument against zero-sum is that "companies who need to change currency are not speculating". True, but weak — and the table shows why:
| Who trades | Daily volume ($tn) | Share |
|---|---|---|
| Between dealers | 4.4 | 46% |
| Dealers vs other financials | 4.8 | 50% |
| Non-financial customers | 0.48 | 5% |
(Does not sum to exactly 100 because of rounding.) Importers, exporters and multinationals paying invoices are 5%. The example of the American company buying euros is real, and marginal.
Careful with that 50%, though: it is a counterparty category, not an intention. Inside it sit non-reporting banks, institutional investors and also hedge funds and proprietary trading firms — 8% of global turnover — who are indeed speculating.
The strong argument is in the other 95%, and above all in one figure that cuts across the whole split: 42% of all volume is FX swaps. A swap is not a directional bet, it is funding. Two parties exchange currencies today and unwind later at an agreed rate. Nobody is predicting anything.
And the engine of 2025's growth was explicitly hedging. After April's tariff shock, when the usual dollar-equity correlation flipped, institutional investors rushed to re-cover their exposure:
| Instrument | Change versus 2022 |
|---|---|
| Options | +108% |
| Forwards | +51% |
| FX swaps | +5% |
The BIS attributes around $1.5 trillion of the jump to that episode. Translation: most forex volume is not people trying to guess where price is going. It is funding, and it is insurance. Knowing where and when liquidity concentrates is more useful than imagining who wants to take yours.
Example: the fund that loses 15 million on purpose
A US pension fund holds $500 million in European equities. It does not want a bet on the euro — it wants the shares. So it hedges the exposure by selling euros forward. The euro rises 3%.
| Item | Result |
|---|---|
| Loss on the hedge | −$15M |
| Gain translating the shares into dollars | +$15M |
| Net currency result | ≈ 0 |
Exactly what it wanted. And if you were on the other side of that forward, you made money. But you did not take it from anyone trying to beat you: you took it from someone who paid for insurance and did not need it. They are content with the outcome. That is the point of insurance.
So where does the money you win come from?
The honest answer is not comforting, but it is the only defensible one: the identity still holds. If you make $200 on EUR/USD, that $200 came from somewhere. What is false is inferring that it came from the pocket of someone trying to beat you who failed.
| Where it can come from | Did someone lose to you? |
|---|---|
| A dealer's inventory, hedged worse than expected | No: that is their business risk |
| Someone who bought insurance and did not need it | No: they got exactly what they paid for |
| Another speculator who got the direction wrong | Yes — and it is the only one of the three |
Only the third is the war the meme describes. And here is the part that matters: inside that third group the game is worse than zero-sum. Before costs it sums to zero; after spread, commissions, overnight swap and slippage, retail speculators as a group lose money in aggregate, and that money goes to intermediaries.
So the correct formulation is not "forex is zero-sum". It is this: zero-sum before costs, negative-sum after — and only within the subset that speculates. You are in that subset. It is the same mechanism that makes almost nobody make money day trading, seen from market structure instead of from the account.
What does this change in your trading?
Stop optimising against an imaginary enemy. The "they are hunting my stop" mental model is expensive because it makes you design rules against a ghost instead of against your own numbers. We already saw that smart money and liquidity carry far less mystery than advertised, and that you are not competing against high-frequency trading in the way you are told.
The one counterparty that gets paid whether you win or lose is the intermediary. Not because they are the villain, but because they charge on every trade. Theirs is the only profitability that does not depend on being right. Which is why your broker being regulated matters more than any zero-sum debate.
The market's scale is not your scale. $9.6 trillion a day does not mean your pair can be traded at any hour at the same cost. Aggregate volume and the liquidity you personally find at three in the morning are different things.
Three ways this debate goes wrong
Confusing accounting with intent. That every trade nets to zero says nothing about why the other side is there. A funding swap and a directional bet net identically in the book and have nothing in common.
Applying the futures model to spot. It is what makes the Market Maker Model so often misread as a conspiracy rather than as inventory management.
Believing that "not zero-sum" means "there is money for everyone". That is the opposite error and it costs just as much. A market richer than a duel does not make retail speculation a favourable game. It stays negative after costs, and market structure does not fix that part.
The market is not a war. It is a toll.
