A plain spoken pillar guide to how you lose money in prediction markets and event contracts: probabilities versus outcomes, fees and spread, liquidity, counterparty risk, crypto risk, and the behavioral traps. Information, not advice.
Last reviewed 28 June 2026 · Educational, not advice
You lose money in a prediction market in five main ways: the outcome goes against you, costs eat your edge, the market is too thin to exit well, the venue holding your money fails, or your own behaviour takes over. A price is an implied probability, not a promise, so a correct read and a losing result sit together comfortably.
this page is general information, not financial, investment, legal, tax, or betting advice. Prediction markets and event contracts carry a real risk of financial loss. Nothing here predicts an outcome or tells you what to trade. Trade only money you can afford to lose, never borrowed money, and never to recover an earlier loss.
You lose money on a prediction market when the outcome resolves against the share you hold, and that happens even when your price was right, because a price is a probability rather than a guarantee. On top of outcome risk sit costs that quietly raise the bar: the trading fee, the gap between the buy price and the sell price, and the drag of trading often.
Beyond your own trade there are risks attached to the venue itself. Thin markets can move against you the moment you enter or exit, the platform holding your funds can fail or freeze them, and onchain venues add wallet and smart contract risk. The largest losses, though, usually come from behaviour: chasing a loss, sizing too big, and confusing a confident feeling with a real edge.
The first way people lose money on prediction markets is by treating a probability as a promise. A contract priced at seventy cents is not a seventy cent path to a dollar. It is a position that, if the price is right, still loses its full value three times out of ten. Over many trades those losses are not bad luck, they are the maths working exactly as expected.
A correct read of the odds and a losing result are completely compatible. Understanding that is the foundation of understanding risk here.
Many venues charge to trade or settle. Every fee raises the probability the outcome needs in order for you to break even.
You usually buy a little above and sell a little below the midpoint. That gap is a cost you pay on entry and again on exit.
The more often you trade, the more times you pay the fee and the spread, so activity itself is a steady drain.
Deposit and withdrawal costs and delays can eat into what looks like a profit on screen.
If few people are trading, the price you see is not the price you get. You may move the market against yourself just by entering or exiting.
Your money sits with the venue. If the venue fails, is sanctioned, or freezes funds, your balance can be at risk regardless of how your trade went.
On onchain venues you also carry wallet, stablecoin, and smart contract risk, and a lost key or a bad approval can cost you everything.
A market can resolve in a way you did not expect, or a contested result can take time to finalise.
Many losses are not about the market at all. Chasing a loss with a bigger position, trading because it feels exciting, anchoring on a price you once saw, and confusing a confident feeling with a real edge are all common and expensive. These traps get worse when money and emotion are mixed, which is most of the time.
The defences are unglamorous: decide your limits before you trade, size positions small enough that one loss does not matter, keep records so you can see what is actually happening, and step away when it stops feeling like a free choice.
You cannot remove the risk of loss, but you can cap it. Trade only money you can afford to lose, never borrowed money, and never to recover a previous loss. Set deposit and time limits where the platform offers them. A clear regulator does not make any market a safe bet, and no page here predicts an outcome as certain.
If participating stops feeling like a free choice, it is worth stepping back. In the United States you can call or text 1-800-GAMBLER or visit ncpgambling.org for free and confidential support.
Hold one idea above all others and most of the rest follows. A market price is the crowd estimate of how likely an outcome is, expressed as a number between 0 and 1 dollar. If a YES share trades at 0.70 dollars and that price is exactly right, the outcome happens about 70 times in 100 and fails about 30 times in 100. On the 30 it fails, your share is worth nothing. You did not misread anything. The loss is the model behaving precisely as it should.
This is why a single result tells you almost nothing about whether you traded well. The only honest test is whether, across many independent trades, the prices you paid were lower than the true probabilities. Over a small number of trades, luck dominates and a good process can look bad while a reckless one looks brilliant. The grid below shows one correctly priced 70 cent contract played out ten times.
Illustrative. If 0.70 is the true probability, three losses in ten is the expected pattern, not a sign of a bad read. Outcomes in any short run will vary around this.
Costs do not feel dramatic, which is exactly why they are dangerous. Two costs apply on almost every venue. The first is the spread, the gap between the price you can buy at and the price you can sell at, so you typically pay a little above the midpoint and receive a little below it. The second is the trading or settlement fee that many venues charge. Both push up the probability the outcome must reach before you simply break even.
Because a winning share pays a fixed 1 dollar, the probability you need to break even is roughly equal to your all in cost per share. The table works that through for three prices, using an illustrative 1 cent fee and an illustrative 1 cent half spread. Real fees and spreads vary widely by platform, which is why we track them in our cross platform fees dataset rather than guessing.
| Quoted YES price | Half spread paid | Fee | All in cost | Probability needed to break even |
|---|---|---|---|---|
| 0.50 dollars | 0.01 | 0.01 | 0.52 dollars | about 52 percent |
| 0.70 dollars | 0.01 | 0.01 | 0.72 dollars | about 72 percent |
| 0.90 dollars | 0.01 | 0.01 | 0.92 dollars | about 92 percent |
Illustrative worked example. Fee and half spread set to 1 cent each for clarity. Method: a binary share pays 1 dollar if it wins, so break even probability is approximately the all in cost per share. Actual costs vary by platform and market.
The outcome now has to be about 72 percent likely, not 70 percent, just for you to end level. Trade the same position often and you pay that gap again and again.
Illustrative. The size of the fee and spread differ by venue, so the break even gap can be larger or smaller than shown.
In the United States, event contracts offered to the public are meant to trade on venues registered with the Commodity Futures Trading Commission as designated contract markets. The CFTC tells customers to confirm that a website or app belongs to a registered entity, and warns that those who trade with unregistered offshore operators may have little or no protection if something goes wrong (per the CFTC customer education page on prediction markets and event contracts, as of June 2026).
That protection is real but narrow. Registration is aimed at custody, market conduct, and disclosure: it pushes venues to safeguard customer funds, monitor for manipulation, and show you clear information about risks and costs (per the CFTC customer education page on prediction markets and event contracts, as of June 2026). It does nothing to change the odds of any individual market. A fully regulated contract can still be mispriced, and you can still lose your entire stake on it.
The CFTC also flags a specific kind of trap to ignore: promises of big payoffs, free money, celebrity endorsements, and glowing reviews are red flags of a scam, not signs of a good opportunity (per the CFTC customer education page on prediction markets and event contracts, as of June 2026). Our own house rule mirrors this. We never tip, never sell picks, and never call an outcome certain, because anyone who does is selling you something. For the longer view on the agency, see our guide on the role of the CFTC, and the US legality hub for where each platform stands.
Decentralized venues remove a custodian, which removes one risk and adds several others. When you hold your own keys, no company can freeze or lose your balance, but no company can recover it for you either. A single mistaken approval, a lost key, or a malicious transaction you sign by accident can empty a wallet, and there is no support desk that can undo it. The convenience of self directed control is paid for in personal responsibility.
There are protocol level risks too. Smart contracts can carry bugs, oracles that report the result can be wrong or delayed, governance can resolve a market in a way you dispute, and stablecoins used as collateral can lose their peg. Some onchain prediction markets also socialize a rare shortfall, meaning a winning position can in unusual cases settle below a full dollar. None of this makes onchain venues uniquely dangerous. It makes them different, with a risk profile you should understand before you fund one. Our pages on platform custody of funds and counterparty and exchange risk go deeper.
The practical takeaway is the same in every venue type. Know who holds your money, know how the market resolves, and know what it costs to get in and out, before you put money at stake rather than after.
Put together, the ways you lose fall into a short list, and each one has a plain defence. Outcome risk is unavoidable, so you manage it by sizing positions small enough that any single loss is survivable. Cost risk is reduced by trading less often and by choosing venues with tight spreads and clear fees. Liquidity risk is handled by using limit orders and avoiding thin markets you cannot exit. Counterparty risk is limited by favouring regulated, well run venues and not leaving large balances idle. Behavioural risk, the biggest of all, is met by deciding your limits in advance and stepping away when it stops feeling like a free choice.
No method removes the risk of loss. The goal is not to eliminate it but to make it small, deliberate, and affordable, so that a bad run is a disappointment rather than a disaster. If you take one thing from this page, let it be that a price is a probability, the costs are real, and the discipline is yours to keep.
Reviewed by Fredrik Filipsson, Editor, on 28 June 2026. This guide is educational and draws on the CFTC customer education materials on prediction markets and event contracts as of June 2026.
The cruelest feature of these markets is that results arrive slowly and noisily. Suppose you genuinely buy shares at prices below their true probabilities, a real and rare edge. Over ten trades you might still be down, because ten outcomes is far too few for skill to separate from luck. The same noise runs the other way. A person with no edge at all can string together a run of wins and conclude they have found a system, then give it all back and more when the average reasserts itself.
This is why we put so much weight on calibration rather than on any single call. A well calibrated trader is one whose 70 percent judgments come true about 70 percent of the time across many cases. You cannot see calibration in a single market, and you cannot feel it in a hot streak. You can only measure it by keeping records, comparing the probabilities you paid against what actually happened, and being honest about the sample being small. Confidence is not evidence, and a vivid memory of a past win is not a strategy.
The danger in the noise is that it tempts you to change your behaviour at the worst moment. A losing run feels like a signal to bet bigger and win it back. A winning run feels like permission to take more risk. Both reactions raise the chance of a large loss. The discipline that protects you is dull on purpose: keep your position sizes steady, judge your process over many trades, and let the maths do its slow work without interference.
If there is one habit that separates people who survive these markets from people who blow up, it is position sizing. The risk of ruin is the chance that a series of losses wipes out your money before any edge can show itself. It rises sharply as your bet size grows relative to your total funds. Stake a large fraction of your balance on each market and a normal, expected losing streak can end you, even if every individual price you paid was fair or good.
The defence is to make each position small enough that a loss, or several losses in a row, is merely annoying. There is no magic number, but the principle is firm: size so that the worst plausible run still leaves you in the game and clear headed. A useful test is to imagine losing the next five trades in a row, because that will happen sooner or later, and ask whether you would still be calm. If the answer is no, the position is too big.
Set these limits before you fund an account, not in the heat of a trade. Decide a maximum stake per market, a maximum amount at risk at any time, and a total you are willing to lose over a month, and use the deposit and time limits a platform offers if it has them. Treat those numbers as fixed. The whole point of deciding in advance is that the version of you mid session, chasing a loss, is exactly the version those limits exist to overrule. For more, see our guides on position sizing and bankroll and responsible play and staying in control.
By treating a probability as a promise and by underestimating costs. A contract at seventy cents still loses three times in ten if the price is right, and fees plus the spread raise the bar further. Losses can be the maths working as expected.
Both raise your break even point. You typically buy above and sell below the midpoint, and many venues add a trading or settlement fee, so the outcome must be a little more likely than the raw price for you to come out even.
It is the risk tied to the venue holding your money rather than to your trade. If the platform fails, freezes funds, or is sanctioned, your balance can be affected regardless of how your position performed.
No. Regulation can reduce some custody and conduct risks, but it does not make any market a safe bet. Prices are often wrong and you can still lose your full stake.
In the United States you can call or text 1-800-GAMBLER or visit ncpgambling.org for free and confidential support. Trade only what you can afford to lose, never borrowed money, and never to chase a loss.