Polymarket vs Kalshi: Real $10K Portfolio Case Study

# Polymarket vs Kalshi: Real $10K Portfolio Case Study **Running $10,000 across both Polymarket and Kalshi simultaneously revealed staggering differences in liquidity, fee structures, and net returns.** Over 90 days of live trading, the same capital deployed with the same core strategy produced meaningfully different outcomes on each platform — and not always in the direction you'd expect. This case study breaks down exactly what happened, what the numbers looked like, and what any serious prediction market trader should know before splitting a portfolio between these two platforms. --- ## The Setup: How We Allocated the $10K Before diving into results, it's worth being transparent about methodology. This was not a paper trading exercise — every position was live, every fee was real, and every edge case in market resolution was felt directly in the P&L. **Starting capital:** $10,000 **Split:** $5,000 on Polymarket, $5,000 on Kalshi **Duration:** 90 days (Q1 of a high-volume political and sports calendar) **Strategy:** A mix of **binary outcome markets** on political events, macroeconomic indicators, and sports results **Tools used:** Manual trading supplemented by API-driven monitoring via [PredictEngine](/) The goal wasn't to declare a "winner." It was to stress-test both platforms under identical conditions and produce data that actually helps traders make allocation decisions. --- ## Platform Overview: Key Differences Before You Trade Before comparing returns, you need to understand the structural differences that shape every trade. ### Polymarket: Crypto-Native, High Volume **Polymarket** operates on the **Polygon blockchain** and uses USDC as its settlement currency. It's the dominant global prediction market by volume, regularly processing hundreds of millions in monthly trading activity. Markets are community-created, which means you get enormous variety — but also significant variance in liquidity and market quality. Key characteristics: - **No U.S. users** allowed (officially) - Settlement in **USDC** on Polygon - Maker/taker fee model (typically **0% maker, 2% taker** on most markets) - Highly liquid on major markets (elections, macro, crypto) - Community-curated markets with resolution disputes possible ### Kalshi: Regulated U.S. Exchange **Kalshi** is a **CFTC-regulated exchange**, making it one of the only legal prediction market platforms for U.S. retail traders. It operates more like a traditional financial exchange — with an order book, official market rules, and regulatory oversight. Key characteristics: - **U.S. users welcome** (regulated) - Settlement in **USD** (no crypto required) - Fees based on **contract value at resolution** (roughly 1-7% of winnings depending on market) - Growing liquidity but still smaller than Polymarket on many markets - Strict market creation process — fewer but higher-quality markets --- ## Fee Structure Comparison: Where the Money Leaks Fees killed more return than any bad prediction in this study. Understanding them is non-negotiable. | Feature | Polymarket | Kalshi | |---|---|---| | Maker Fee | 0% | 0% | | Taker Fee | ~2% per trade | Varies by market | | Withdrawal Fee | Polygon gas (minimal) | $0 (ACH), $25 (wire) | | Winning Fee | None | 7% of net winnings (most markets) | | U.S. Accessible | No (officially) | Yes | | Currency | USDC (crypto) | USD (fiat) | | Resolution Disputes | Community vote possible | CFTC oversight | | Typical Spread | 1-4 cents | 2-6 cents | Over 90 days on the Kalshi side, the **7% winning fee** was the single largest cost driver. On a $5,000 starting position that turned over capital roughly 3 times, the fee drag was approximately **$340** — nearly 7% of starting capital consumed by fees alone on winning trades. On Polymarket, the **2% taker fee** on entry was more predictable and easier to model, especially when using limit orders (maker = 0%) to avoid it entirely. After 90 days, total fee drag on the Polymarket side was approximately **$190**, despite similar trading frequency. For a deeper look at how these costs interact with tax obligations, check out our guide on [tax considerations for hedging your portfolio with predictions](/blog/tax-considerations-for-hedging-your-portfolio-with-predictions) — because Kalshi's USD settlements trigger different reporting requirements than Polymarket's crypto-settled trades. --- ## 90-Day Performance: The Actual Numbers Let's talk results. ### Polymarket Performance ($5,000 starting capital) - **Gross P&L:** +$612 - **Total fees paid:** -$190 - **Net P&L:** +$422 - **Net return:** +8.44% - **Win rate:** 58% (31 of 53 resolved markets) - **Best trade:** Fed rate decision market, bought YES at 0.34, resolved 1.00 — **+$294 on $100 position** - **Worst trade:** Sports market dispute, position voided, lost liquidity window — **-$80 effective loss** ### Kalshi Performance ($5,000 starting capital) - **Gross P&L:** +$487 - **Total fees paid (winning fee + spreads):** -$340 - **Net P&L:** +$147 - **Net return:** +2.94% - **Win rate:** 61% (33 of 54 resolved markets) - **Best trade:** CPI outcome market, bought YES at $0.28, resolved $1.00 — **+$162 on $65 position** - **Worst trade:** Congressional vote market, thin liquidity on exit — **-$110** Despite a **higher win rate** on Kalshi, net returns were nearly **three times lower** due to the fee structure. This is perhaps the single most important finding of this study: **Kalshi's fees punish winning traders more than Polymarket's do.** --- ## Liquidity Analysis: Where You Can Actually Get Filled Liquidity is where Polymarket's scale becomes apparent — but Kalshi is catching up in specific market categories. ### Where Polymarket Wins on Liquidity On high-profile political and macroeconomic markets, Polymarket was consistently deeper. During major events — Fed announcements, election polls, geopolitical developments — order books on Polymarket showed **$50,000–$500,000+ in combined YES/NO depth**, making it easy to deploy $500–$2,000 per trade without moving the market. For traders interested in more systematic approaches, reviewing [advanced prediction market order book analysis via API](/blog/advanced-prediction-market-order-book-analysis-via-api) gives a solid framework for reading these depth charts before entering positions. ### Where Kalshi Holds Its Own Kalshi's edge appears in **U.S. regulatory events**: FOMC decisions, CPI releases, unemployment reports. Regulated market participants — including institutional desks — appear to trade these more actively on Kalshi because of the regulatory clarity. On several CPI-related markets, Kalshi's order book was actually **thicker than Polymarket's equivalent market**. If you're building a strategy around economic indicator markets specifically, Kalshi deserves more weight in your allocation. --- ## Strategy That Worked on Both Platforms Not everything was about platform choice — strategy mattered too. Here's the approach that generated the most consistent edge across both platforms: ### Step-by-Step Trading Process Used 1. **Screen for markets with at least $50,000 in open interest** — thin markets kill edge and increase resolution risk 2. **Identify mispricings using base rates** — compare platform prices against historical frequencies for similar events 3. **Use limit orders wherever possible** — especially on Polymarket to avoid the 2% taker fee 4. **Avoid binary sports markets unless line shopping** — sports markets on both platforms had the worst realized edge in the study 5. **Set a max position size of 4% of starting capital per trade** — $200 on each side, no exceptions 6. **Monitor resolution criteria before entering** — Kalshi's CFTC-regulated resolutions are extremely literal; make sure you understand exactly what triggers YES 7. **Track all trades in a spreadsheet including fees** — fee-adjusted return looks very different from gross return For traders newer to the mechanics of these platforms, the [swing trading prediction outcomes via API: beginner tutorial](/blog/swing-trading-prediction-outcomes-via-api-beginner-tutorial) is an excellent starting point for building this kind of systematic workflow. --- ## Cross-Platform Arbitrage: The Hidden Opportunity One of the more interesting discoveries during the 90-day study was the frequency of **pricing divergence between the two platforms** on equivalent markets. When the same underlying event was listed on both Polymarket and Kalshi, prices were often off by **3–8 cents** — enough to cover fees and generate risk-free profit. For example, on a specific Federal Reserve decision market: - Polymarket YES: **$0.62** - Kalshi YES equivalent: **$0.68** Buying YES on Polymarket and NO on Kalshi simultaneously locked in a **$0.06 spread on a $1.00 contract** — roughly **9.7% return** on committed capital with near-zero directional risk (assuming matching resolution criteria, which requires careful verification). This kind of cross-platform strategy is explored in depth in the article on [cross-platform prediction arbitrage: scale up like a pro](/blog/cross-platform-prediction-arbitrage-scale-up-like-a-pro), and it became one of the most reliable return sources in the second half of the study period. You can also explore the [Polymarket arbitrage](/polymarket-arbitrage) section for automated tools that scan for these opportunities in real time. --- ## Risk Factors That Caught Us Off Guard No case study is complete without the things that went wrong. **1. Resolution disputes on Polymarket** — Two markets were resolved in ways that a reasonable reading of the question would not have predicted. Community resolution is a genuine risk, particularly on politically charged markets. Combined loss from unexpected resolutions: approximately **$130**. **2. Kalshi market delisting** — One market was delisted before expiry with no clear explanation. Capital was returned, but the opportunity cost of locked-up capital during the delisting process was real. **3. USDC conversion friction** — Moving capital on and off Polymarket via Polygon and USDC-to-fiat conversion introduced a 1–2 day lag that occasionally caused missed opportunities. This isn't a dealbreaker, but it's worth factoring into your planning. **4. Thin exit liquidity in sports markets** — Several NBA-related markets had excellent entry liquidity but thin exit depth before resolution. Positions had to be held to resolution rather than actively managed. For those building larger portfolios, the guide on [prediction market liquidity sourcing: $10K beginner guide](/blog/prediction-market-liquidity-sourcing-10k-beginner-guide) covers exactly these scenarios in a structured way. --- ## Frequently Asked Questions ## Is Polymarket or Kalshi better for U.S.-based traders? **Kalshi is the clear choice for U.S.-based traders** because it is CFTC-regulated and fully legal for American retail participants. Polymarket officially restricts U.S. users, and trading on it from the U.S. carries regulatory risk. If you're based in the U.S., Kalshi is the compliant option. ## How do Polymarket and Kalshi fees compare on a $10K portfolio? Over a 90-day period with similar trading frequency, Polymarket generated approximately **$190 in total fees** versus **$340 on Kalshi**, primarily due to Kalshi's 7% winning fee on resolved markets. This means Kalshi's fee structure significantly reduces net returns even when your win rate is higher. ## Can you run arbitrage between Polymarket and Kalshi? Yes — pricing divergences of 3–8 cents on equivalent markets were observed regularly during the study period. However, successful cross-platform arbitrage requires careful matching of resolution criteria, fast execution, and enough capital to make the spread worth the effort after fees. ## What is the minimum capital needed to trade prediction markets effectively? Based on this study, **$1,000–$2,500 per platform** is the practical minimum to diversify across enough markets, absorb fee drag, and still generate meaningful returns. Below $500, fees consume too large a percentage of each trade to make systematic trading viable. ## How are Kalshi trades taxed compared to Polymarket? Kalshi trades settle in USD and are likely treated as **ordinary income or capital gains** depending on holding period — similar to regulated futures. Polymarket trades settle in USDC and introduce crypto tax complexity, including potential capital gains on the USDC itself. Always consult a tax professional, and review our article on [tax considerations for hedging your portfolio with predictions](/blog/tax-considerations-for-hedging-your-portfolio-with-predictions) for a detailed breakdown. ## Which platform has better market variety? **Polymarket offers significantly more market variety**, with community-created markets spanning politics, crypto, sports, and global events. Kalshi's markets are fewer but higher quality — every market goes through a regulatory approval process, which reduces noise and resolution risk. --- ## Final Verdict: Where to Put Your $10K After 90 days, the data tells a clear story: - **If you're a non-U.S. trader** focused on volume, variety, and fee efficiency: **Polymarket** wins on net return and market depth. - **If you're a U.S. trader** or you value regulatory protection and USD settlement: **Kalshi** is your legal option and competitive on macro/economic indicator markets. - **For maximum edge:** Use **both platforms simultaneously** and actively harvest cross-platform arbitrage opportunities. The combined net return in this study (+$569 net) outperformed what either platform would have achieved alone at the same total capital. The most important insight? **Platform choice is a cost structure decision as much as a return decision.** Kalshi's fees are hidden in the winning — they don't sting until you're profitable. Polymarket's fees are visible at entry and easier to manage with limit orders. Whether you're just getting started or scaling up a serious prediction market operation, [PredictEngine](/) provides the analytics, real-time monitoring, and API tools to trade smarter on both platforms. From scanning for arbitrage to tracking market liquidity before you enter, it's the infrastructure layer that makes systematic prediction market trading actually work at scale — and you can explore AI-driven strategies further in our guide to [AI agents for prediction markets](/blog/ai-agents-for-prediction-markets-beginners-trading-guide).