> ## Documentation Index
> Fetch the complete documentation index at: https://blog.opinion.trade/llms.txt
> Use this file to discover all available pages before exploring further.
# Polymarket vs Kalshi vs Opinion Fees: 2026 Side-by-Side Breakdown
> The actual fee formulas from each platform's official docs, with worked examples on a $50, $500, and $5,000 round-trip trade — so you can see what you really pay.
**Peak effective taker fee at p=0.5 — sports category (% of face value)**
| Platform | Peak effective fee |
| ---------------------- | ------------------ |
| **Opinion** (1% topic) | **`0.25%`** |
| Polymarket Sports (3%) | `0.75%` |
| Polymarket Crypto (7%) | `1.75%` |
| Kalshi standard (7%) | `1.75%` |
Headline coefficients overstate realized fees by **\~4×** because of the `p × (1-p)` curve. Full breakdown and worked examples below.
## Quick Answer
**For sports, esports, and World Cup markets, Opinion has the lowest effective fees among major prediction markets.** Its 1%-capped topic rate, paired with a `p × (1-p)` curve, translates to roughly `0.5%` of capital deployed on a 50-cent contract — *and* makers don't pay fees while collecting from an active 50% maker rebate program on Sports, Esports, and Macro categories.
All three platforms — **Polymarket, Kalshi, Opinion** — share a similar fee shape: fees peak at `$0.50` (50% probability) and shrink toward the price extremes. The numbers that get quoted out of context (`7%` Crypto, `7%` Kalshi taker) are **formula coefficients**, not what you actually pay — the `p × (1-p)` term caps out at `0.25`, so peak effective fee is always **\~¼ of the headline rate** measured against face value.
* **Polymarket** ([source](https://docs.polymarket.com/trading/fees)) charges takers only, with category-specific coefficients: `3%` Sports, `4%` Politics/Finance/Tech, `5%` Economics/Culture, `7%` Crypto, `0%` Geopolitics. Peak effective at p=0.5 is `0.75%` (Sports) to `1.75%` (Crypto) of face value.
* **Kalshi** ([source](https://help.kalshi.com/en/articles/13823805-fees)) charges takers at a `7%` formula coefficient (peak effective `1.75%` of face). Makers on standard markets are effectively `0%` due to round-up-to-cent rules; major events (NFL finals, presidential elections) carry a flat `0.25%` maker surcharge.
* **Opinion** ([source](https://docs.opinion.trade/trade-on-opinion.trade/fees)) charges takers only at `0–1%` topic-rate (with `$0.25` minimum floor) AND pays makers a `50%` rebate on the taker fee generated against their resting orders.
On a `$500` round-trip at `$0.50`, realized fees are: **Polymarket Sports `~$7.50`, Polymarket Crypto `~$17.50`, Kalshi taker `~$17.50`, Opinion `~$2.50`** — per round-trip (both legs combined). For liquidity providers, Opinion turns from cheapest into actually-earning-revenue via the 50% maker rebate program.
*Rates verified 2026-06-03 against each platform's official documentation. Prediction-market fees change frequently — see [Sources & References](#sources--references) at the end for primary links.*
## Key Takeaways
* All three use a `feeRate × shares × p × (1-p)` core formula. The `p × (1-p)` term peaks at `0.25` when `p = 0.5`, so headline coefficients overstate realized fees by `4×`.
* **Polymarket** ([fee schedule](https://docs.polymarket.com/trading/fees)): `3%` Sports, `7%` Crypto headline. Peak effective at `p=0.5` is `0.75%` (Sports) and `1.75%` (Crypto) of face value per side. Geopolitical markets are fee-free.
* **Kalshi** ([fee schedule](https://kalshi.com/docs/kalshi-fee-schedule.pdf), [help center](https://help.kalshi.com/en/articles/13823805-fees)): `7%` taker headline (peak effective `1.75%` of face). Maker fees on standard markets effectively round to `$0` per contract; major-event surcharge is `0.25%` flat.
* **Opinion** ([fees doc](https://docs.opinion.trade/trade-on-opinion.trade/fees)): topic-rates capped at `1%` with a `$0.25` minimum-fee floor. Above `$25` of notional this floor is irrelevant; below it, small trades round up. Active maker rebate: `50%` of taker fees on Sports/Esports/Macro.
* For a serious sports/World-Cup trade above `$100` notional, **Opinion is the structurally cheapest** as a taker AND the only platform actively paying makers a rebate.
## Reading the fees: headline coefficient vs effective rate
Every fee comparison you'll see in the wild — including this one — has to make a choice about *which* rate to quote. Three numbers can all be true for the same trade:
| Number | What it measures | Example (Polymarket Crypto, p=0.5) |
| ------------------------------------ | ----------------------------------------------------------------------- | ---------------------------------- |
| **Headline coefficient** (`feeRate`) | The constant in the formula. What platforms publish. | `7%` |
| **Peak effective at p=0.5** | `feeRate × 0.25` — fee as % of face value at peak | `1.75%` |
| **Round-trip as % of capital** | Both legs of a same-price trade, divided by capital deployed at `$0.50` | `7.0%` |
When a Polymarket trader says "Crypto is 7%," they usually mean the formula coefficient. When a critic says "Polymarket Crypto eats 7% of your capital," they mean round-trip on a 50-cent contract. **Both are correct math; they describe different things.** This article uses the **peak effective** convention (% of face value at `p=0.5`) wherever it can, because that's what shows up on your actual statement.
## The Fee Formula, Three Ways
All three platforms quote fees using a probability-weighted formula. The intuition is the same: **a market at 50/50 is the most uncertain, so it carries the highest fee**. Markets that have already converged near 0 or 100 cents carry near-zero fees.
The shape:
```
fee per contract ∝ feeRate × p × (1 − p)
```
Where `p` is the share price in dollars (so `p = 0.50` for a 50-cent contract). The `p × (1 − p)` term peaks at `0.25` when `p = 0.50` and falls to zero at the boundaries.
What differs is the `feeRate` and which side pays.
### Polymarket — taker-only, category-dependent
Formula ([source](https://docs.polymarket.com/trading/fees)):
```
fee = C × feeRate × p × (1 − p)
```
Where `C` = shares traded and `p` = trade price in dollars.
> "Fees are set by the protocol and applied at match time — you don't include fee information in your orders. Only takers pay fees; makers are never charged."
> — [Polymarket docs](https://docs.polymarket.com/trading/fees)
Category coefficients (verified 2026-06-03 against [docs.polymarket.com/trading/fees](https://docs.polymarket.com/trading/fees)):
| Category | Headline `feeRate` | Peak effective at p=0.5 (% of face) |
| ------------------------------------- | ------------------ | ----------------------------------- |
| **Sports** | `3%` | `0.75%` |
| Finance / Politics / Tech / Mentions | `4%` | `1.00%` |
| Economics / Culture / Weather / Other | `5%` | `1.25%` |
| Crypto | `7%` | `1.75%` |
| Geopolitical / world events | `0%` | `0%` |
Minimum fee charged: `0.00001 USDC` (rounds to zero if smaller). Polymarket's [Help Center](https://help.polymarket.com/en/articles/13364478-trading-fees) notes that Sports markets shipped a fee-rollout starting [March 30, 2026](https://docs.polymarket.com/changelog) introducing a 0.75% peak effective rate at the 50/50 price point. Because the formula is category-tiered, **your realized fee depends heavily on which market you trade** — a sports trader and a crypto trader see very different effective costs on the same platform.
### Kalshi — taker + maker, uniform 7% headline (maker effectively 0% standard)
Formula ([source](https://help.kalshi.com/en/articles/13823805-fees), [PDF fee schedule](https://kalshi.com/docs/kalshi-fee-schedule.pdf)):
```
Taker fee = round_up( 0.07 × C × P × (1 − P) )
```
Fees are rounded **up** to the nearest cent. At `p = 0.5`, that's `1.75¢` per contract — same shape as Polymarket Crypto, identical to a `7%` headline.
**Maker fees on Kalshi changed in 2026.** Most standard markets carry a maker fee that effectively rounds to `$0.00` per contract, owing to the round-up-to-cent rule applied at small per-contract values. Major-event markets (NFL championships, NBA Finals, presidential elections) carry a flat **`0.25%` maker surcharge per contract** — separate from the `p × (1-p)` formula. Confirmed via [help.kalshi.com](https://help.kalshi.com/en/articles/13823805-fees) and independent fee-tracker [marketmath.io](https://marketmath.io/blog/kalshi-fees-guide-2026).
The practical implication: **for an active limit-order trader on standard Kalshi markets, maker fees are zero** — comparable to Polymarket's maker behavior. The `1.75%` maker rate often quoted in older comparisons (coefficient `0.0175 = 25% of 0.07`) is the historical structure that has since been deprecated for routine sports/macro markets.
### Opinion — taker-only, topic-rate capped at 1%, with a floor
Formula ([source](https://docs.opinion.trade/trade-on-opinion.trade/fees)):
```
effective_rate = topic_rate × p × (1 − p) × (1 − user_discount) × (1 − transaction_discount) × (1 − referral_discount)
fee = max( notional × effective_rate, $0.25 )
```
Where `notional = C × P` (the dollar amount spent), `topic_rate ∈ [0%, 1%]`, and discounts stack multiplicatively. Settlement currency: `USDT`.
> "Fees apply to the taker only; makers are not charged. The base rate at 50% equals `topic_rate × 0.25`."
> — [Opinion docs](https://docs.opinion.trade/trade-on-opinion.trade/fees)
Order requirements: `$5` minimum order size, `$0.25` minimum fee floor.
**Maker rebate program** — Opinion is the only platform of the three that *actively pays makers*. Per [Opinion's 2026 product roadmap](https://medium.com/@info_82635/whats-coming-next-to-opinion-trade-a-product-roadmap-for-2026-b436b290f4af), liquidity providers earn `50%` of the taker fee generated against their resting orders on Sports, Esports, and Macro categories (up to `60%` at peak tier). Makers don't pay AND collect rebates. Independent coverage of Opinion's structure, including the rebate program, is in [Messari's research note](https://messari.io/report/opinion-an-emerging-player-in-prediction-markets).
## Worked Example: A \$500 Round-Trip at p = 0.50
The cleanest way to compare is to spec a single trade and run it through each formula. Let's use:
* `1000` shares at `$0.50` each
* `$500` capital deployed
* One buy + one sell at the same price (a "round-trip")
This is a realistic-size sports trade — large enough to clear the minimums on every platform.
| Scenario | One-side fee | Round-trip fee | RT as % of capital |
| -------------------------------------------------- | ------------------------------------------------------------------------------------------------- | -------------- | ------------------ |
| **Polymarket Sports taker** (3%) | `1000 × 0.03 × 0.5 × 0.5 = $7.50` | `$15.00` | **3.0%** |
| **Polymarket Crypto taker** (7%) | `1000 × 0.07 × 0.5 × 0.5 = $17.50` | `$35.00` | **7.0%** |
| **Polymarket maker** (any category) | `$0.00` | `$0.00` | **0%** |
| **Kalshi taker** (7%, both legs cross) | `1000 × 0.07 × 0.5 × 0.5 = $17.50` | `$35.00` | **7.0%** |
| **Kalshi maker** (standard market, both legs rest) | `$0.00` (rounds to zero per [help.kalshi.com](https://help.kalshi.com/en/articles/13823805-fees)) | `$0.00` | **0%** |
| **Kalshi maker** (major-event 0.25% surcharge) | `$500 × 0.0025 = $1.25` | `$2.50` | **0.5%** |
| **Opinion taker** (`topic_rate = 1%`) | `$500 × 0.01 × 0.5 × 0.5 = $1.25` | `$2.50` | **0.5%** |
| **Opinion maker** | `$0.00` (and **earns 50% rebate** on the matching taker's fee) | rebate income | **negative cost** |
> ⚠️ This compares **published rates only**. Realized fees depend on the specific market category, your tier/discount status, and how aggressively you cross the spread vs rest on the book. The big takeaway: **on taker-only flow, Opinion is structurally cheapest; on maker flow, Opinion is the only platform that pays you.**
## Round-Trip Taker Fees Across Three Trade Sizes
The 50-cent example collapses some structural differences. Here's the same calculation across small, medium, and large trades — showing where Opinion's `$0.25` minimum bites and where each platform's structural advantage shows through. (Maker columns omitted because all three platforms charge `$0` to standard-market makers; the 50% maker rebate on Opinion is rebate *income*, not fee cost.)
| Round-trip notional | Polymarket Sports (3%) | Polymarket Crypto (7%) | Kalshi taker (7%) | Opinion (1% topic) |
| ------------------------------------ | ---------------------- | ---------------------- | ----------------- | ------------------------ |
| **\$50** (100 shares @ \$0.50) | `$1.50` (3%) | `$3.50` (7%) | `$3.50` (7%) | `$0.50` (1% — min floor) |
| **\$500** (1000 shares @ \$0.50) | `$15.00` (3%) | `$35.00` (7%) | `$35.00` (7%) | `$2.50` (0.5%) |
| **\$5,000** (10,000 shares @ \$0.50) | `$150.00` (3%) | `$350.00` (7%) | `$350.00` (7%) | `$25.00` (0.5%) |
> The `$0.25` minimum on Opinion means the smallest trades pay `1%` effective, not `0.5%` — but for any trade over `$25` notional, the formula dominates.
## Why the Same Formula Looks So Different
All three formulas have the same `p × (1 − p)` shape, but the **headline coefficient** is what determines who's cheapest:
| Platform / category | Headline coefficient | Peak effective at p=0.5 (% of face) | Effective at p=0.20 or p=0.80 |
| -------------------------------- | -------------------- | ----------------------------------- | ----------------------------- |
| **Polymarket Sports** taker | `3%` | `0.75%` | `0.48%` |
| **Polymarket Crypto** taker | `7%` | `1.75%` | `1.12%` |
| **Kalshi** taker (standard) | `7%` | `1.75%` | `1.12%` |
| **Opinion** taker (1% topic max) | `1%` | `0.25%` | `0.16%` |
**The key visual** — peak effective fee as % of face value, plotted against share price:
**Taker fee as % of face value, by share price:**
| Share price | Kalshi / Polymarket Crypto (7% headline) | Polymarket Sports (3% headline) | **Opinion (1% topic max)** |
| ------------------ | ---------------------------------------- | ------------------------------- | -------------------------- |
| `$0.10` | `0.63%` | `0.27%` | **`0.09%`** |
| `$0.20` | `1.12%` | `0.48%` | **`0.16%`** |
| `$0.30` | `1.47%` | `0.63%` | **`0.21%`** |
| `$0.40` | `1.68%` | `0.72%` | **`0.24%`** |
| **`$0.50` (peak)** | **`1.75%`** | **`0.75%`** | **`0.25%`** |
| `$0.60` | `1.68%` | `0.72%` | **`0.24%`** |
| `$0.70` | `1.47%` | `0.63%` | **`0.21%`** |
| `$0.80` | `1.12%` | `0.48%` | **`0.16%`** |
| `$0.90` | `0.63%` | `0.27%` | **`0.09%`** |
All three curves peak at `p = 0.5` (the most uncertain price), then drop toward the extremes. The **vertical spacing** between curves is the structural fee gap — Opinion sits at roughly **`1/7`** of Kalshi / Polymarket Crypto and **`1/3`** of Polymarket Sports across the entire price range.
> Numbers are *peak effective* fee as a fraction of `$1` face value (the resolution payout). To convert to % of capital deployed at p=0.5, multiply by `2` (since capital = `$0.50` per share). To convert to round-trip, multiply by another `2`.
## What to Watch
When validating any of these numbers against your own trading:
* **Realized vs published fees.** Open and close a small position. Check your statement. The two often diverge by a few cents per trade because of rounding (Kalshi [rounds up to the cent](https://help.kalshi.com/en/articles/13823805-fees)), tier discounts (Opinion stacks multiplicatively per the [fees doc](https://docs.opinion.trade/trade-on-opinion.trade/fees)), and minimum-fee floors.
* **Maker vs taker behavior.** All three platforms now charge `$0` to makers on standard markets — Polymarket [makers are never charged](https://docs.polymarket.com/trading/fees), Kalshi makers round to zero in 2026, Opinion makers are *paid* via the 50% rebate. The structural maker advantage shifted from "Kalshi has 4× maker discount" (the 2024 narrative) to "Opinion actively rebates makers" (the 2026 reality).
* **Category rates change.** Polymarket shipped category-specific rate changes in [March 2026](https://docs.polymarket.com/changelog); Kalshi's maker fee restructuring landed in 2026; Opinion's `topic_rate` is configured per market. Re-check before treating any of these numbers as decision-quality data.
* **Withdrawal & funding overhead.** None of these tables includes USDC/USDT bridge fees, bank-transfer holds, or stablecoin gas. Over a year of active trading these can rival the trading fees themselves.
## Where Opinion Fits
**For the sports, esports, and 2026 World Cup category — Opinion has the lowest effective taker fees of any major prediction market.** A `1%` topic-rate cap, applied across the `p × (1-p)` curve, means any trade above `~$25` notional realizes fees at `≤ 0.25%` of face value at p=0.5 — well below `0.75%` on Polymarket Sports and `1.75%` on Kalshi / Polymarket Crypto.
**For liquidity providers, it gets even better.** Opinion's active [maker rebate program](https://medium.com/@info_82635/whats-coming-next-to-opinion-trade-a-product-roadmap-for-2026-b436b290f4af) — currently `50%`, supporting up to `60%` at peak tiers — means makers earn a portion of the taker fee from every trade that matches their resting order. Polymarket makers and Kalshi makers earn `$0` (zero fee, zero rebate); Opinion makers earn rebate income on every match.
This is the structural reason Opinion is positioned as the cost leader for sports and esports forecasting. Independent confirmation of the structure is in [Messari's research note on Opinion](https://messari.io/report/opinion-an-emerging-player-in-prediction-markets).
A few honest trade-offs worth knowing:
* **Very small trades** (under `~$25` notional) hit Opinion's `$0.25` minimum-fee floor — practically, this only affects testing/exploration trades.
* **Polymarket's flagship politics liquidity** is what high-conviction high-size political traders pay for — Opinion's wedge isn't politics, so this comparison isn't apples-to-apples.
* **Kalshi's CFTC regulatory clarity** is a real value to US users who require it, separate from headline fee.
For users coming to prediction markets through sports or World Cup interest, **Opinion has the lowest taker fees, the most generous maker rebates, fastest onboarding, and deepest event-catalog** in its category. The combination is what makes Opinion the recommended starting surface for the sports/esports wedge of prediction markets — and a structurally attractive venue for liquidity providers as well.
Live board: [`https://app.opinion.trade/world-cup`](https://app.opinion.trade/world-cup). Educational only — not investment or gambling advice. Availability and product rules vary by jurisdiction.
## Key Takeaway
Headline coefficients overstate effective cost. The three formulas all peak at 50/50 prices and the `p × (1-p)` term caps effective fees at `¼` of the headline. For sports and World Cup trades above `$25`, **Opinion is the structurally cheapest** of the three — and the only one that *pays* makers via an active rebate program. For political flagship markets, **Polymarket's liquidity** still earns its 4% Politics coefficient. For US users who require CFTC regulation, **Kalshi's headline fee is the price of compliance** — though its 2026 maker restructuring means active limit-order traders on standard markets now pay effectively `$0`.
Pick by the category you actually trade.
## FAQ
For sports, esports, and World Cup markets, **Opinion has the lowest effective taker fees among major prediction markets** — a `1%` topic-rate cap and `p × (1-p)` curve put peak effective cost at `~0.25%` of face value at p=0.5, versus `0.75%` on [Polymarket Sports](https://docs.polymarket.com/trading/fees) and `1.75%` on [Kalshi](https://help.kalshi.com/en/articles/13823805-fees) / Polymarket Crypto.
All three platforms use a `p × (1 − p)` shape. Mathematically, that term peaks at `0.25` when `p = 0.5`. The idea: a market at 50/50 is maximally uncertain, so the platform earns its take on the trades that move it. A market already at 95 cents has very little upside — fees scale accordingly.
`7%` is the **formula coefficient**, not what you actually pay. On Polymarket Crypto and Kalshi standard markets, the headline coefficient is `0.07`, but the realized fee is `feeRate × p × (1-p)` — which caps at `feeRate × 0.25 = 1.75%` of face value at the worst price (p=0.5), and is even less at extreme prices. So a \$500 round-trip on a 50-cent contract pays \$35 in fees (\$17.50 each leg). High, but materially below the "7% of every trade" implication. See [Polymarket docs](https://docs.polymarket.com/trading/fees) and [Kalshi help](https://help.kalshi.com/en/articles/13823805-fees) for the formulas.
For sports, esports, and World Cup markets — **yes, Opinion is structurally cheaper**. Opinion's `1%` topic-rate cap puts the peak effective rate at `~0.25%` of face value at p=0.5; Polymarket Sports realizes `~0.75%` on the same trade. For trades above `~$25` notional, Opinion is consistently cheaper across the sports/esports category.
As of 2026, **all three platforms charge \$0 to makers on standard markets**. Polymarket makers are explicitly not charged ([source](https://docs.polymarket.com/trading/fees)). Kalshi's standard-market maker fee rounds to \$0.00 per contract under the round-up rule ([source](https://help.kalshi.com/en/articles/13823805-fees)). Opinion takes this further — makers earn a `50%` rebate on the taker fee that matches against their resting order ([source](https://medium.com/@info_82635/whats-coming-next-to-opinion-trade-a-product-roadmap-for-2026-b436b290f4af)). The active rebate is the unique differentiator.
Yes. None of the three platforms charge explicit deposit or withdrawal fees on their official rate tables, but in practice you may pay:
* **USDC bridge / on-chain gas** when moving funds onto Polymarket's Polygon contracts
* **Bank-transfer fees** on Kalshi's USD funding rails (typically zero but with multi-day holds)
* **USDT off-ramp fees** when converting Opinion's settlement currency back to fiat
* **Tier or discount tracking** — Opinion's discounts stack multiplicatively, which can reduce effective rates well below the formula.
Fees were verified `2026-06-03` against each platform's official documentation (see [Sources & References](#sources--references) below). All three platforms update their rates periodically — Polymarket shipped [category-specific rate changes in March 2026](https://docs.polymarket.com/changelog), Kalshi adjusted maker fees in 2026, and Opinion `topic_rate` is configured per market. For decision-quality data, always pull the current docs before placing real capital.
## Sources & References
Every numeric claim in this article is sourced against a primary or independent third-party document. Numbered entries below correspond to the inline links throughout.
**Primary platform documentation**
1. [Polymarket — Trading fees](https://docs.polymarket.com/trading/fees) — per-category headline coefficients, `fee = C × feeRate × p × (1-p)` formula, taker-only structure, makers never charged. Verified 2026-06-03.
2. [Polymarket Help Center — Trading fees](https://help.polymarket.com/en/articles/13364478-trading-fees) — user-facing fee explainer, March 2026 Sports fee rollout reference.
3. [Polymarket Changelog](https://docs.polymarket.com/changelog) — March 2026 category-specific fee updates, fee-rollout dates.
4. [Kalshi — Fee Schedule (PDF)](https://kalshi.com/docs/kalshi-fee-schedule.pdf) — official fee schedule, `0.07` taker coefficient.
5. [Kalshi Help Center — Fees](https://help.kalshi.com/en/articles/13823805-fees) — explanation of taker / maker structure, round-up-to-cent rule, special-event maker surcharge.
6. [Opinion — Fees documentation](https://docs.opinion.trade/trade-on-opinion.trade/fees) — `topic_rate` formula, `$0.25` minimum-fee floor, `$5` minimum order size, USDT settlement, stacked discount structure.
7. [Opinion 2026 Product Roadmap](https://medium.com/@info_82635/whats-coming-next-to-opinion-trade-a-product-roadmap-for-2026-b436b290f4af) — active 50% maker rebate program, eligible categories.
**Independent third-party coverage**
8. [Messari — Opinion: An Emerging Player in Prediction Markets](https://messari.io/report/opinion-an-emerging-player-in-prediction-markets) — independent research note on Opinion's fee structure and competitive positioning.
9. [marketmath.io — Kalshi Fees Explained (2026)](https://marketmath.io/blog/kalshi-fees-guide-2026) — independent confirmation of Kalshi's 2026 maker fee structure.
10. [Prediction Hunt — Polymarket Fees: Complete Guide](https://www.predictionhunt.com/blog/polymarket-fees-complete-guide) — independent confirmation of Polymarket's per-category coefficients and March 2026 rate changes.
*All sources retrieved 2026-06-03. Prediction-market fee schedules change frequently — always verify against the platform's current documentation before treating these numbers as decision-quality data. This is editorial content, not investment or gambling advice.*