Prediction market strategy: edge, fees and sizing explained

A sound prediction market strategy comes down to one rule: only trade when your probability estimate differs from the price by more than your costs: fees, the bid/ask spread and the risk of misreading the contract. Everything else is about finding that gap honestly and sizing positions so you survive being wrong. This guide explains the maths with each platform’s published fees, as of October 2026. It’s education, not tips: we don’t recommend trades, and most traders lose money after costs.

1. Think in edge, not in “who will win”

A contract at 40¢ pays $1 if it happens. If your honest estimate is 40%, you have no edge, whatever you think will happen. If you believe it’s 50%, your expected profit per contract is 50¢ − 40¢ = 10¢ before costs. Kalshi itself describes prices as “the consensus beliefs of market participants.” Beating a consensus needs information or analysis the market hasn’t priced. Our guide to implied probability covers the basics.

2. Know your break-even after fees

Fees raise the probability you need to break even. For a taker on Kalshi, the break-even probability is the price plus 0.07 × price × (1 − price):

At 10¢, FanDuel’s flat 2¢ fee needs a 12% hit rate to break even, which is 20% more than the price implies. That’s why fee structure is part of your strategy, not an afterthought. See our Kalshi fees guide for every platform’s formula.

3. Don’t pay the spread if you don’t have to

“When you buy with a market order, you pay the ask price,” Polymarket US’s docs note, and “when you sell, you receive the bid price.” Crossing a 3¢ spread costs as much as a fee. A limit order that rests on the book avoids the spread, and on many platforms it costs less: Novig and Polymarket US charge makers nothing, while Kalshi charges makers a smaller 0.0175 fee on some markets. The trade-off is that your order may never fill. Patience is a strategy.

4. Read the resolution rules before the headline

Many apparent edges are really misread contracts. Kalshi’s Fed markets, for example, ask whether the rate’s upper bound will be “above” a level after a specific meeting. A similar-sounding Polymarket market asks about “greater than or equal to” a level on a fixed date. Before trading, check:

• The exact threshold and date in the rules.
• The settlement source (league, government agency or named outlet).
• Edge cases: ties, postponements, did-not-play rules in combos.

5. Size positions so you survive being wrong

Even a genuine edge loses often. A 60% shot loses four times in ten. The classic sizing rule is the Kelly criterion, from J. L. Kelly Jr.’s 1956 Bell System paper. For a $1 binary contract bought at price p, when you believe the true probability is q, the Kelly fraction of your bankroll is (q − p) ÷ (1 − p). Example: you believe 50% on a 40¢ contract, so (0.50 − 0.40) ÷ 0.60 ≈ 17% of your bankroll. That’s aggressive, and it assumes your estimate is right. Many disciplined traders bet a fraction of Kelly, or a small fixed percentage, because overestimating your edge with full Kelly can drain an account quickly.

6. Treat “arbitrage” with care

You’ll sometimes see the same event priced differently on Kalshi and Polymarket. Real arbitrage needs identical settlement terms (event, source and date) and a gap bigger than both platforms’ fees and spreads combined. Usually the wording differs, so you’d hold two different bets, not a risk-free position. Our Kalshi odds guide explains why prices differ.

Our checklist before any trade

Prediction markets carry real risk of loss. If trading stops being fun or starts costing more than you can afford, stop, and call or text 1-800-MY-RESET for free, confidential help.

18+ (21+ where required). Prediction markets involve risk of loss — only trade what you can afford to lose. Gambling problem? Call or text 1-800-MY-RESET (National Problem Gambling Helpline, 24/7).

Last updated 2026-10-17. Not financial or betting advice.