Polymarket vs Kalshi Fees: Which Platform Takes Less of Your Winnings?

---

Industry Context

It's not just about fees — the whole industry is shifting. A few years ago, prediction markets were niche, almost underground. Fast-forward to 2026, and they're fully mainstream.

The big headline: Polymarket, after years of being blocked in the U.S., came back in a legal, regulated form. They acquired a regulated exchange and got a green light from the CFTC to operate under U.S. law. That means the biggest, most liquid crypto-driven prediction market is available to American users without a VPN workaround. This move doesn't just legitimize Polymarket; it signals that regulators are treating prediction markets as a financial instrument, not just a novelty.

Meanwhile, Kalshi has leaned hard into legitimacy by pitching itself not just as a political or economic forecasting tool, but also as a way to trade sports event contracts. Want to buy shares on whether a team makes the playoffs, or if a star QB throws for 300 yards? Kalshi has pushed to make these legal under CFTC oversight. This is where the line between "finance" and "sports betting" gets blurry — some state regulators argue it looks like gambling, while Kalshi frames it as a new category of tradable event derivatives. The debate is still ongoing, but the fact that Kalshi is even allowed to test these waters shows how much the landscape has shifted.

PredictIt, still running under its older "no-action" letter with caps and restrictions, feels stuck in the past compared to these two. With this much capital flowing in, regulators are also paying closer attention to insider trading and market manipulation risks — an evolving area that could reshape the competitive landscape.

---

2026 Fee Comparison Table

Before we dive into the details, here's the at-a-glance breakdown:

Exchange	Trading Fee	Withdrawal Fee	The Catch
Polymarket	Category-based, 0%–1.8% peak	None	Varies by market category; some categories exempt.
Kalshi	0.07% to 1.75%	Free (ACH)	Formula-based; peaks at 50/50 odds.
PredictIt	10% of gross profit	5%	You lose 15%+ on winning trades.
Opinion	~1% peak, $0.25 minimum	None	Minimum fee is punishing on small contracts.
ProphetX	2% of profits	None	Thinner liquidity on most markets.

Tired of doing the math yourself? Use Prediction Hunt's free fee calculator to see your exact net payout before you place a trade.

---

Fee Differences Explained

Polymarket recently introduced category-based trading fees. Fees use the same P×(1-P) curve as Kalshi, but the multiplier varies by market category. Politics markets carry a 4% multiplier (~1.0% peak at 50¢), crypto markets are higher at 7.2% (~1.8% peak), sports markets are lower at 3% (~0.8% peak), and geopolitical markets remain fee-exempt. Maker orders (limit orders) are still free. This is a significant change from Polymarket's previous near-zero fee structure, but fees are still competitive with Kalshi on most market types.

Kalshi takes a more "financial exchange" approach. Kalshi fees come from a small transaction charge on your expected earnings as defined here. Depending on whether you're a market maker / taker and what the contract price is, your fees can range from close to 0% to a maximum of 1.75%.

PredictIt is the most old-school (and arguably the priciest with its fees). They lop off 10% of your profits whenever you cash out a winning trade, plus another 5% fee if you want to withdraw your money. If you lose a bet, there's no fee — but obviously you're still out the cash. For casual traders, this model is punishing, and it makes short-term trades particularly unattractive.

Opinion uses a formula-based fee similar to Kalshi and Polymarket, with a 4% multiplier (~1.0% peak). The key difference is a $0.25 minimum fee per contract, which makes Opinion significantly more expensive on low-priced contracts. On a 10¢ contract, you're paying 25¢ in fees — more than double the contract price. Opinion also requires a minimum $5 purchase. For higher-priced contracts the minimum is less impactful, but it's something to factor in when evaluating opportunities on the platform.

ProphetX sits in an interesting middle ground. They charge a flat 2% fee on profits — no withdrawal fee, no complicated formula. On a winning trade, you keep 98% of your gains. The tradeoff is liquidity: ProphetX markets tend to be thinner than Kalshi or Polymarket, so you may face wider spreads that effectively cost you more than the 2% fee saves.

---

Why Fees Matter

Fees also play into how "efficient" markets are. With lower costs, arbitrageurs jump in quickly to correct mispricings. With higher fees, some inefficiencies just linger, which means the market's "implied probabilities" might not reflect reality as well. By contrast, PredictIt's structure almost disincentivizes active trading, which can leave its markets stale or distorted.

The introduction of category-based fees on Polymarket is a notable shift. It suggests that as prediction markets mature and attract more volume, platforms are moving toward sustainable revenue models. Kalshi's fee structure has remained relatively stable, while Polymarket's move brings its fee model closer to what Kalshi has offered for years. For traders, the key takeaway is that fee-shopping across platforms now matters more than ever — a trade that's cheapest on Polymarket for a politics market might be cheaper on Kalshi for a crypto market.

Want to see exactly how fees eat into your returns? Use Prediction Hunt's free fee calculator to compare your net payout across platforms before you trade.

---

How We Incorporate Fees Into Our Estimated Returns

When you see estimated returns in the "Smart Bets" tab on Prediction Hunt, those numbers already factor in platform-specific fees.

On PredictIt, we account for their 10% fee on profits. That means if a trader wins $100 on a contract, they only keep $90 after fees. PredictIt also charges a 5% withdrawal fee, but we don't incorporate that one since it varies depending on how often users cash out. The 10% profit fee, though, directly impacts returns and must be included to make comparisons realistic.

For Polymarket, we now account for their category-based taker fees. The fee uses the same P×(1-P) curve as Kalshi, but with a category-specific multiplier — for example, 4% for politics (~1.0% peak) and 7.2% for crypto (~1.8% peak). This means Polymarket is no longer free for takers, and our return estimates reflect these costs.

Kalshi operates more like a regulated financial exchange. They charge both maker and taker fees, depending on your order type. To stay conservative, we use the taker fee formula, which is applied to the full contract value:

Fee = 0.07 x P x (1 - P)

where P is the contract price in dollars (so a 50¢ contract means P = 0.5). This effectively scales the fee with the liquidity and risk of the contract — and we round up to the nearest cent to mirror real-world execution.

For Opinion, we apply their fee formula (4% multiplier on the P×(1-P) curve) with a $0.25 minimum fee per contract. The minimum fee is the dominant cost on most trades — for any contract under ~$0.90, you're paying the $0.25 floor. This makes Opinion significantly more expensive than other platforms for small or low-priced contracts, and our return estimates reflect that.

For ProphetX, we apply their 2% profit fee to winning positions. The fee is straightforward — 2% of your net profit on the trade. No fee on losing positions.

These adjustments ensure that when you see a "risk-free return" on Prediction Hunt, it already reflects the fees you'd realistically pay. Arbitrage looks great in theory — but after fees, some edges shrink fast. Our goal is to make those returns as close to real-world performance as possible.