How much money should I put on Kalshi? Only money that can go to zero without changing your life, and less of it than you think: a total account no bigger than a few percent of your liquid savings, with no single market ever holding more than 1% to 5% of that account. Those two numbers do more to decide whether you're still trading in six months than any single position you'll ever take. Kalshi doesn't blow up accounts. Sizing does. And the mechanics of a binary contract, where the losing side of every question settles at exactly $0, punish oversizing with a speed a weekend of sportsbook bets never matches: another expiration is always minutes away.
The Quick Answer
Fund your Kalshi account only with risk capital — a defensible ceiling is 1% to 5% of your liquid savings, and never money earmarked for rent, debt, or emergencies. Inside the account, cap each market at 1% to 5% of your balance, with correlated positions counted together as one. The worked example below runs a $500 bankroll contract by contract, including what a losing streak actually does to it.
Why "How Much" Is The Question That Decides Everything
The reason this question needs a real answer is visible in the wreckage. "FML just lost my life savings" is a verbatim user post surfaced in our research into how prediction-market traders actually lose money — and it describes a sizing failure, not a prediction failure. Whoever wrote it may even have been right more often than not. Right 70% of the time at the wrong size still ends in that post.
The aggregate numbers say the same thing quietly. Sportico's analysis of Kalshi's public trade data found retail bettors lost about $117 million on parlays from January 1 through April 30, 2026, roughly 15% of the $800 million they risked on them: for every $100 wagered on multi-leg combos, about $15 gone (Sportico, May 2026). Losing 15% of cost basis is survivable if a parlay is 2% of your bankroll. It is a life-savings post if it's 80% of one.
So the honest framing isn't "how much can I make." It's "how much can I watch settle at zero and keep functioning." Everything below builds from that, and I'll make one promise now: by the time you reach the worked example, you'll be able to size any contract on the board in about ten seconds.
What A Kalshi Contract Actually Risks
Sizing rules only make sense once you're clear on what the instrument does. Every yes/no market on Kalshi is a binary contract that settles at $1.00 or $0. The exchange's CFTC-filed rulebook states it plainly in Rule 6.3(a): when a contract expires in your favor it "will pay the Settlement Value for such Contracts (e.g. $1.00)" to your side; when it expires against you, the full value goes "to the holders of short positions" instead (KalshiEX Rulebook, Rule 6.3). There is no partial credit. A contract bought at 60¢ either resolves YES and pays $1.00, or it settles on expiration at $0 and the whole 60¢ is gone.
Total loss as the routine outcome, on repeat, is what sizing has to survive here: there are no pushes and no partial grades, and a new expiration is never far away. If the exchange structure itself is new to you, our explainer on why a prediction market is not a sportsbook covers the mechanics, and how Kalshi settlement works walks the payout timeline.
Fees compound the math slightly. Per Kalshi's fee schedule as filed with the CFTC (fetched August 12, 2026), general trading fees follow the formula round up(0.07 × C × P × (1 − P)) — 7% of expected earnings, peaking at mid-range prices: 100 contracts at 50¢ costs $1.75 in fees, while the same 100 contracts at 10¢ or 90¢ cost $0.63. The same 2022-filed schedule states "There is no settlement fee," and Kalshi's help center (fetched August 12, 2026) confirms the model: "Kalshi makes money by charging a transaction fee on the expected earnings on the contract," with maker fees applying only in some markets. Rates can change by filing, so check the linked schedule for current numbers. See our full breakdown of what Kalshi fees really cost at every price point.
Rule One: Size The Deposit Like Risk Capital
The deposit decision comes before any trade and matters more than all of them. The test: if this entire balance settled at $0 this month, would anything in your life change? If yes, the deposit is too big.
Stated in percentages, because dollar figures mean different things to different people:
- A Ceiling Of 1% To 5% Of Liquid Savings is a defensible range for a total event-contract bankroll. Closer to 1% if you're new, if your income is variable, or if this money took years to save.
- 0% Of Anything With A Job. Rent, tuition, emergency fund, debt payments: money with an assignment never goes on an exchange. That's the line the "life savings" post crossed.
- Top-Ups Are A Decision, Not A Reflex. Redepositing after a bust is where casual trading turns into chasing. If you wouldn't have made the deposit before the loss, don't make it after.
None of this is unique to prediction markets; it's the same discipline behind bankroll management for sports betting, applied to an instrument that hits zero faster.
Rule Two: Cap Every Market At 1% To 5% Of The Bankroll
Inside the account, the per-market cap is the rule that actually saves you, because binaries tempt you into oversizing in a specific way: high-probability contracts feel safe. A market at 95¢ reads like a formality, but buying it risks 95¢ to win a nickel, and when one of those settles NO you need nineteen winners just to get back to even. Even 95¢ favorites go to zero, and the price being high does not change what a full loss does to an oversized position.
So the cap is unconditional: 1% to 5% of your current balance per market, regardless of how obvious the outcome feels. Two refinements make it sharper:
- Count Correlated Markets As One Position. Five contracts that all need the same election result, the same rate cut, or the same team's win are one bet wearing five tickers. The 15% parlay-hold number from earlier is what correlation-stacking costs in aggregate — cap the cluster, not just each leg.
- Let Conviction Move You Inside The Range, Not Outside It. If you want a formula, the binary version of the Kelly criterion says the edge-justified fraction is (p − price) / (1 − price), where p is your true probability. Believe a 50¢ contract is really 60% to happen and Kelly allows (0.60 − 0.50) / 0.50 = 20% of bankroll. But full Kelly assumes your probability estimate is right, which is exactly what you shouldn't assume. Quarter-Kelly lands at 5%, the top of the range. That's not a coincidence; it's why the range ends there.
Note what these caps are not: Kalshi's own position limits. The exchange files per-market ceilings with the CFTC for compliance reasons; they're set for whales, not for you, and hitting one is a symptom, not a strategy.
Worked Example: A $500 Bankroll, Contract By Contract
Take a $500 account and a 3% per-market cap: $15 per market. Here's what that buys at three prices, fees computed with the filed formula:
| Contract Price | Contracts for ≈$15 | Cost | Fee | If it resolves YES | If it settles at $0 |
|---|---|---|---|---|---|
| 30¢ | 50 | $15.00 | $0.74 | +$34.26 | −$15.74 |
| 60¢ | 25 | $15.00 | $0.42 | +$9.58 | −$15.42 |
| 90¢ | 16 | $14.40 | $0.11 | +$1.49 | −$14.51 |
The row worth staring at is the last one. The 90¢ position risks $14.51 to make $1.49 — nearly 10-to-1 the wrong way, yet it's the trade that feels safest at the ticket screen, and it's the shape most oversized positions take. The 30¢ row inverts that: same $15 at risk, but the payoff justifies the likely loss. The cap keeps every row survivable; the price decides whether the risk is worth taking at all.
Now the part sizing is really for: the losing streak. The cap is recomputed on your current balance before every trade, so position size shrinks as the account does — that recalculation is the self-braking feature of percentage sizing. Ten straight losses at 3% of the shrinking balance leave this account around $370 — a 26% drawdown that stings and teaches. Ten straight losses at 25% each leaves about $28. Same trader, same markets, same bad stretch; one sizing rule is the entire difference between a drawdown and a funeral.
The Blow-Up Patterns That Empty Accounts
The caps above fail in practice through three repeatable behaviors:
- Chasing velocity. Fast-cycle markets let a loss be "fixed" within the hour, which is exactly how losses metastasize. Kalshi's 15-minute markets are the sharpest version: a new expiration every quarter hour means a tilted trader never has to wait to redeposit the pain.
- Averaging down on a thesis. Run it through the worked example: your 25 contracts at 60¢ are trading at 35¢, so you buy 40 more "to lower your basis" — and a $15 position becomes a $29 one, nearly double the cap, in the market that's already going against you. Re-evaluate at the new price as if you held nothing; add only if the combined position would still pass the cap fresh.
- Promoting "can't-miss" trades above the cap. The 95¢ trap in its most expensive form: conviction so high the rules feel unnecessary. The rules exist precisely for the trades that feel like exceptions.
One more honest note: Kalshi being a CFTC-regulated exchange (covered in our look at whether Kalshi is legit) protects the integrity of settlement, not the quality of your decisions. Regulation means the 95¢ contract that goes to zero pays out correctly. It does not mean it was a good buy.
FAQ
Is there a minimum deposit on Kalshi? Deposit minimums and funding methods change; check Kalshi's help center for current terms before funding. The bankroll question is independent of the minimum; the right deposit is set by your finances, not by the smallest number the platform accepts.
Should I put my whole bankroll into one high-confidence market? No. That's the exact pattern behind the life-savings post above. A 90¢ contract still settles at $0 when the event breaks the other way, and a single position holding 100% of the account converts one wrong call into a zeroed balance.
How is Kalshi bankroll management different from sportsbook bankroll management? The unit-sizing logic transfers almost directly, but binaries change two things: the losing side always settles at exactly $0 (no partial grading, no push except per contract terms), and high-priced contracts invert your risk-reward: at 90¢ you risk nine times what you can win. Sizing has to respect the price, not just the stake.
Can I lose more than I put in on Kalshi? Buying YES or NO contracts, your maximum loss is the price paid plus trading fees; a bought contract can't go below $0. That containment is per position; it does not stop an account from bleeding out through many oversized positions in a row.
In Summary
The answer to "how much money should I put on Kalshi" was never going to be a dollar figure. It's a pair of percentages: an account that's a small slice of money you can truly lose, and a 1% to 5% cap on every market inside it, correlated clusters counted as one. The person behind the life-savings post didn't need better predictions — they needed a rule that made their worst day a 3% day. Ten seconds of arithmetic per ticket is the whole system: balance, times cap, divided by price, rounded down.
Research belongs inside those caps, never as a reason to stretch them. If you want to build the habit of checking a read against sharper information before you click buy, our free expert picks are a no-cost place to practice — with the sizing rules above deciding how much any conviction is allowed to cost you.
