Trader Playbook: Cross-Platform Prediction Arbitrage

# Trader Playbook: Cross-Platform Prediction Arbitrage **Cross-platform prediction arbitrage** is the practice of exploiting price discrepancies for the same event across different prediction markets — buying "Yes" on one platform where the price is low and simultaneously selling "Yes" (or buying "No") on another where the price is higher. When executed correctly, this strategy can generate consistent, low-risk returns that are largely independent of the actual event outcome. This playbook covers every layer of the strategy, from identifying opportunities to managing execution risk, with real-world examples that show exactly how the math works. --- ## What Is Cross-Platform Prediction Arbitrage? Prediction markets let participants buy and sell contracts on the probability of real-world events — elections, economic data releases, sports outcomes, and more. Platforms like **Polymarket**, **Kalshi**, **Manifold**, and **PredEx** often list the same or closely related events simultaneously. Because each platform has its own liquidity pool, market makers, and user base, prices for the same contract frequently diverge. **Arbitrage** exploits those gaps. If Polymarket prices "Federal Reserve cuts rates in September" at 42 cents and Kalshi prices the same event at 51 cents, a trader who buys on Polymarket and sells on Kalshi locks in a theoretical 9-cent spread before fees. At scale, these small spreads compound into meaningful returns. This is fundamentally different from speculative trading. You're not betting on whether the Fed cuts rates — you're betting that the market will price it consistently, and you're capturing the gap when it doesn't. --- ## The Major Platforms and Their Characteristics Understanding each platform's mechanics is step one. Here's a quick breakdown: | Platform | Asset Type | Fees | Liquidity | Best For | |---|---|---|---|---| | **Polymarket** | USDC (crypto) | ~2% taker fee | High (top markets) | Fast-moving political/macro events | | **Kalshi** | USD (regulated) | 7% on profits | Medium | Economic data, regulated contracts | | **Manifold** | Play money (Mana) | None | Low | Practice/calibration only | | **PredEx** | Crypto | Variable | Low-medium | Niche markets | | **Metaculus** | Points-based | None | Very low | Long-horizon forecasting | **Key insight:** Polymarket and Kalshi are the two highest-liquidity real-money platforms and represent the most actionable arbitrage pairing. Manifold can be used to calibrate your probability estimates but shouldn't be included in real-money arb calculations. --- ## How to Identify Arbitrage Opportunities (Step-by-Step) Finding arb opportunities manually is slow and error-prone. Here's the systematic process professionals use: 1. **Build or subscribe to a price aggregator.** Tools like [PredictEngine](/) automatically scan multiple platforms and flag price discrepancies above a defined threshold in real time. 2. **Normalize contract definitions.** Confirm both contracts resolve on the same event, the same date, and with the same conditions. "Fed cuts by 25bps in September" and "Fed cuts rates in September" are *not* the same contract. 3. **Calculate gross spread.** Subtract the lower price from the higher price. Example: 51 - 42 = 9 cents. 4. **Subtract all fees.** Polymarket charges ~2%; Kalshi charges 7% on profits. On a $1,000 position, fees could absorb 4-6 cents of that 9-cent spread. 5. **Assess liquidity depth.** Check the order books on both platforms. If Kalshi only has $200 of liquidity at 51 cents, your position is capped and the price will move against you as you fill. 6. **Check timing risk.** How long will it take to fill both legs? A gap that exists for 30 seconds in a fast-moving market may close before you finish executing. 7. **Execute simultaneously where possible.** Use API access or an [AI trading bot](/ai-trading-bot) to fire both orders at the same time. 8. **Monitor for resolution risk.** Even perfectly hedged arbitrage can lose money if one platform resolves differently than another (rare but real). --- ## Real Arbitrage Examples with the Math ### Example 1: The Fed Rate Cut Play (Polymarket vs. Kalshi) In August 2024, the "Fed cuts rates at September FOMC" contract showed the following prices: - **Polymarket:** 42 cents (42% implied probability) - **Kalshi:** 51 cents (51% implied probability) A trader buys $1,000 of Yes on Polymarket at 42 cents and sells $1,000 of Yes on Kalshi at 51 cents (i.e., buys No at 49 cents). **Scenario A: Fed cuts (contract resolves Yes)** - Polymarket pays out $1,000 × (1/0.42) = $2,381 gross → profit of $1,381 - Kalshi Yes pays out (trader is short Yes, so loses) → loss of ~$1,000 on the Kalshi leg - Net before fees: approximately +$381 minus platform fees **Scenario B: Fed does not cut (contract resolves No)** - Polymarket: lose $1,000 - Kalshi No pays out: profit of approximately $490 ($1,000 × 0.49/1) - Net before fees: approximately -$510 on Polymarket + $490 on Kalshi = -$20 Wait — that's a loss in Scenario B. This illustrates why **position sizing matters**. The trader needs to size the two legs so they're dollar-neutral at resolution, not dollar-neutral in cost. Proper hedging math: > Buy X contracts of Yes on Polymarket (at $0.42 each), sell Y contracts of Yes on Kalshi (at $0.51 each), where X × $0.58 (payout if No) = Y × $0.49 (payout if No on Kalshi short). Getting this ratio right is non-trivial, which is why platforms like [PredictEngine](/) that automate hedge ratio calculations are so valuable for active arb traders. ### Example 2: The 2026 Midterms Congressional Control Around October 2026, "Republicans retain House majority" was priced at: - **Polymarket:** 61 cents - **Kalshi:** 67 cents This 6-cent spread on a high-volume contract attracted significant arb activity. Because the Kalshi fee structure bites harder on winning contracts, the actual post-fee spread was closer to 3.5 cents. On a $10,000 position, that's still $350 in near-riskless profit — earned in days, not months. For context on how traders approached that election cycle, our piece on [algorithmic trading strategies for Supreme Court ruling markets](/blog/algorithmic-trading-strategies-for-supreme-court-ruling-markets) covers similar event-driven setups in detail. ### Example 3: Sports Event Arbitrage Cross-platform arb isn't limited to politics. NBA playoff game winner contracts frequently show 4-8 cent spreads between Polymarket and niche sports prediction platforms. The key challenge here is **time decay** — sports contracts resolve within hours, which means the window to execute is tight and timing risk is highest. Our analysis on [election outcome trading during NBA playoffs](/blog/election-outcome-trading-during-nba-playoffs-risk-analysis) explores how to manage this compressed time horizon effectively. --- ## Tools and Technology for Scaling Your Arb Strategy Manual arbitrage at meaningful scale is nearly impossible. You need infrastructure. ### Price Monitoring Dedicated aggregators scan platforms every few seconds and alert you to spreads above your minimum threshold. [PredictEngine](/) includes this as part of its core dashboard, showing live cross-platform spreads with fee-adjusted net estimates. ### Order Book Analysis Understanding *where* liquidity sits in the book tells you whether you can fill your target size without moving the market against yourself. The guide on [AI-powered prediction market order book analysis](/blog/ai-powered-prediction-market-order-book-analysis-simplified) explains how to read depth charts specifically for prediction markets, which behave differently from equity order books. ### API Access Both Polymarket and Kalshi offer APIs. Automating your order placement means you can execute both legs within milliseconds of each other — critical for tight spreads. If you're new to building on these APIs, the [beginner's guide to LLM-powered trade signals](/blog/beginners-guide-to-llm-powered-trade-signals-this-may) offers a practical starting point. ### Limit Orders vs. Market Orders Never use market orders for arbitrage on thin books. A market order on a $500-liquidity contract could fill you at 10 cents worse than the quoted price. For deep dives on limit order strategy specifically, the article on [maximizing Kalshi returns with limit orders](/blog/maximize-kalshi-returns-mastering-limit-orders-for-profit) is required reading. --- ## Risk Management: What Can Go Wrong Even "riskless" arbitrage carries real risks that traders underestimate: | Risk Type | Description | Mitigation | |---|---|---| | **Execution risk** | One leg fills, the other doesn't | Use simultaneous API execution | | **Liquidity risk** | Market moves before full fill | Cap position to available depth | | **Resolution risk** | Platforms interpret event differently | Read contract specs carefully | | **Counterparty/platform risk** | Platform insolvency or withdrawal freeze | Diversify capital across platforms | | **Regulatory risk** | Platform restricts access mid-position | Maintain withdrawal access proactively | | **Fee miscalculation** | Underestimating platform costs | Build fee calculator into workflow | The most common mistake new arb traders make is ignoring **resolution risk**. In 2023, a high-profile case saw Polymarket and a competing platform resolve a contract on an ambiguous news event differently — traders who thought they were hedged ended up with two losing positions. Always read the full resolution criteria, not just the headline. --- ## Sizing and Portfolio Management Successful arb traders treat their operation like a portfolio, not a series of one-off bets. **Core principles:** - Allocate no more than **15-20% of capital** to any single arb pair, regardless of how tight the spread looks - Maintain a **cash reserve of 20-30%** at all times to cover unexpected margin calls or to capitalize on sudden large spreads - Track **return on capital deployed**, not just gross profit — a 3% return on $5,000 deployed for 48 hours is a 547% annualized return - Log every trade with entry/exit prices, fees, and net outcome to identify which platforms and event types generate the best spreads Traders scaling beyond $50,000 in arb capital should consider [reinforcement learning approaches to portfolio management](/blog/scale-a-10k-portfolio-using-reinforcement-learning-trading), which can optimize position sizing dynamically based on historical spread data. --- ## Common Mistakes and How to Avoid Them - **Treating similar contracts as identical.** "Democrats win Senate" and "Democrats win Senate majority" may resolve differently. Always compare resolution criteria word for word. - **Ignoring slippage.** Quoted prices in thin books are not fill prices. Test with small orders first. - **Overcomplicating with too many platforms.** Start with Polymarket/Kalshi pairs. Add platforms only when you have monitoring infrastructure in place. - **Missing the withdrawal timeline.** Kalshi withdrawals can take 1-3 business days. If your capital is locked up post-resolution, you miss the next opportunity. - **Chasing tiny spreads.** A 1-cent spread after fees is rarely worth the execution complexity. Focus on spreads of 4 cents or more net of fees. --- ## Frequently Asked Questions ## What is cross-platform prediction arbitrage? **Cross-platform prediction arbitrage** is the practice of buying and selling contracts on the same real-world event across two or more prediction market platforms to profit from price discrepancies. By simultaneously holding opposing positions at different prices, traders can lock in a profit regardless of the event's actual outcome. The key is ensuring both contracts have identical resolution criteria and executing both legs before the spread closes. ## How much money can you realistically make from prediction market arbitrage? Returns vary widely based on capital deployed, market conditions, and execution speed. Experienced traders report net returns of **1-4% per arb trade** after fees, with trades lasting anywhere from hours to weeks. At scale ($50,000+ deployed), consistent arb activity can generate $2,000-$8,000 monthly, though opportunities fluctuate significantly around major events. ## Is prediction market arbitrage legal? Yes, in most jurisdictions where prediction market trading itself is legal. **Kalshi** is a CFTC-regulated exchange in the United States, making trading on it fully legal for U.S. residents. Polymarket restricts U.S. users due to regulatory constraints, though enforcement is inconsistent. Always consult current platform terms of service and local regulations before trading. ## What are the biggest risks in cross-platform arbitrage? The biggest risks are **execution risk** (one leg fills but not the other), **resolution risk** (platforms interpret the event differently), and **liquidity risk** (insufficient depth to fill at the quoted price). Less obvious but equally important is platform risk — if a platform freezes withdrawals or becomes insolvent while you have capital locked in a position, you could face significant losses. ## Do I need coding skills to do prediction market arbitrage? Not necessarily, but having some technical ability dramatically improves your results. Platforms like [PredictEngine](/) abstract much of the complexity, offering pre-built dashboards that flag cross-platform spreads without requiring custom code. However, traders who can access APIs and automate execution will consistently outperform those relying on manual monitoring, especially on fast-moving contracts. ## How do I choose which platforms to monitor for arbitrage? Start with **Polymarket and Kalshi** — they have the highest liquidity, the most overlapping contract coverage, and the most transparent fee structures. Once you're consistently profitable on that pair, expand to other platforms. Prioritize platforms that cover the same event categories you're already tracking, and always verify that contract definitions match before assuming a spread is real. --- ## Start Trading Smarter with PredictEngine Cross-platform prediction arbitrage is one of the most systematic, repeatable edges available in modern financial markets — but it rewards preparation and infrastructure over gut instinct. The traders who profit consistently are those who've automated their monitoring, built disciplined position sizing rules, and invested in tools that do the heavy lifting. [PredictEngine](/) was built specifically for this kind of sophisticated prediction market trading. From real-time cross-platform spread alerts to AI-assisted order book analysis and automated execution tools, it gives you everything in this playbook in one place. Whether you're running your first arb trade or managing a six-figure prediction market portfolio, [explore PredictEngine's full feature set and pricing](/pricing) to see how it can accelerate your edge — and start capturing spreads before someone else does.