A Polymarket expected-value (EV) bot compares its own probability estimate with the price it can actually trade at, then subtracts fees and execution costs. For a YES share held to resolution, gross EV per share is q − p: your estimated probability of YES resolving true (q) minus the purchase price (p). That calculation describes an edge only if the estimate is sound and the market resolves as the bot expects; positive calculated EV is not a promise of profit.
How do you calculate expected value on Polymarket?
Polymarket outcome shares trade between $0 and $1. A winning share pays $1 USDC at resolution; a losing share is worth $0. A share can also be sold before resolution at the then-current market price. Polymarket describes prices as probabilities, but the market price reflects what users are currently willing to buy and sell at—it is not a guarantee of the outcome or your own probability estimate.
YES shares
Let p be the executable price for one YES share and q your estimate that YES will resolve true. If you hold through resolution, gross expected profit per share is:
EV = q × $1 + (1 − q) × $0 − p = q − p
For example, if a bot estimates q = 0.58 and can buy at p = $0.52, its estimated gross EV is $0.06 per share before fees and other costs. This is an illustrative calculation, not a forecast or measured result.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →#1 Best Overall
- Language: english
- Book - trading: technical analysis masterclass: master the financial markets
- It is made up of premium quality material.
NO shares
Use the bot’s estimated probability that NO resolves true and the executable NO-share price. If that probability is qNO and the purchase price is pNO, gross EV per share held to resolution is qNO − pNO. Do not assume that the displayed YES price gives the exact price at which a NO share can be bought; use the relevant executable quote.
Use a tradeable price, not a convenient price
A last trade or midpoint may no longer be available. Compare the estimate with the price the bot can execute against on the relevant side of the order book, and account for how much quantity is available there. A theoretical edge can disappear if the bot’s order moves through the book or only part of it fills.
What costs belong in a bot’s EV decision?
The basic q − p result is gross EV for a YES purchase held to resolution. A practical decision subtracts expected trading fees and execution costs:
Rank #2
- As a day trader, you can live and work anywhere in the world. You can decide when to work and when not to work.
- You only answer to yourself. That is the life of the successful day trader. Many people aspire to it, but very few succeed. Day trading is not gambling or an online poker game.
- To be successful at day trading you need the right tools and you need to be motivated, to work hard, and to persevere.
Net EV = estimated gross EV − expected fees − expected execution costs
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallCrashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteExecution costs can include the effect of buying at a worse price than the quote used in the calculation, as well as costs associated with an exit if the bot plans to sell before resolution. A bot should calculate expected value for its intended entry, position size, and exit plan rather than treating a displayed price as a guaranteed fill.
Check the current fee schedule
Polymarket Help Center’s July 10, 2026 fee article says makers are not charged fees and takers pay fees in certain market categories. It gives the formula fee = C × feeRate × p × (1 − p), where C is shares and p is share price. Its listed category rates are:
| Market category | Fee rate in the July 10, 2026 help article |
|---|---|
| Crypto | 0.07 |
| Sports, economics, culture, weather, and general | 0.05 |
| Finance, politics, mentions, and tech | 0.04 |
| Geopolitics | 0 |
These are the rates published in that dated article, not a guarantee of the fee on a particular order today. Fees can change; check the live market settings and current schedule before trading. Whether an order is treated as maker or taker also matters to the fee calculation.
How would you build a Polymarket EV bot?
Build the bot as a chain of checks: discover eligible markets, collect current prices and market details, generate a probability estimate, calculate net EV, validate the resolution rules, and only then decide whether an order fits the strategy. The Polymarket Institute’s official research-data page documents the Gamma API for market and event records—including outcomes, prices, volume, status, fee fields, and token IDs—the CLOB API for requests keyed by outcome token_id, including price and historical-price requests, and the Data API for user trade history and closed positions.
- Define the market universe. Use market and event records to identify active markets that match the bot’s subject, liquidity requirements, and settlement criteria. Exclude markets whose wording or resolution source the model cannot interpret reliably.
- Retrieve the relevant outcome data. Map each candidate outcome to its
token_id. Monitor current prices and, where useful, retrieve historical prices for analysis. Confirm the current API documentation for endpoint behavior, authentication, rate limits, and order execution before implementation; those details may change. - Generate a probability estimate. Keep the model’s estimate separate from the market-implied price. Record the inputs, timestamp, market identifier, and predicted probability so that later evaluation can compare predictions with resolved outcomes.
- Calculate net EV against an executable quote. Apply the YES or NO formula to the price available for the intended order, then account for the applicable fee and expected execution costs. Require a margin above zero large enough for the model’s uncertainty and the bot’s operational constraints; the sources do not establish a universal threshold.
- Validate the settlement terms. Parse the full market rules and specified resolution source before placing an order. A matching headline or an outside source’s apparent result is not a substitute for the market’s own settlement criteria.
- Track fills and outcomes. Record intended and actual order prices, quantities, fees, partial fills, exits, and final resolutions. Trade history and closed-position data can support record keeping and model evaluation.
Access to APIs makes data collection and monitoring possible; it does not by itself establish that a probability model has an edge or that an order can be executed profitably. Polymarket’s published interface descriptions should be paired with current API and market documentation before a bot is connected to live trading.
How can you tell whether the probability model has an edge?
The difficult input in EV is q, not the arithmetic. A model that systematically overestimates the chance of its chosen outcome can report positive EV on losing trades. Evaluate probability forecasts against resolved outcomes and test them on data not used to develop the model. Historical price data can help examine market behavior, but a backtest does not guarantee that future prices, liquidity, fees, or fills will match.
When comparing candidate trades, assess them across the factors that determine whether a calculated edge is meaningful:
- Probability versus executable price: compare the bot’s estimate with the available bid or ask for the intended trade, not only a midpoint or last trade.
- Net EV: include the market’s fee treatment and realistic execution costs.
- Depth and liquidity: check whether the intended order size can be filled without consuming prices that erase the edge.
- Model quality: review calibration and out-of-sample performance rather than relying on a single favorable estimate.
- Settlement clarity: prefer rules and resolution sources the model can interpret with confidence.
- Exposure and time: consider how much capital is committed and how long it may remain exposed before resolution.
These are evaluation criteria, not a claim that any one setting or market category produces profitable trades.
Recommended Free Tools
Why can an apparently positive EV trade still lose?
The probability estimate can be wrong
EV is conditional on the quality of the estimated probability. A small numerical gap between q and p may be less meaningful than the uncertainty in the model itself. Backtesting and calibration can help reveal errors, but they cannot remove uncertainty about future events.
Rules and resolution can differ from a headline
Polymarket says markets resolve according to their predefined rules and describes its resolution mechanism as the UMA Optimistic Oracle. Its Help Center account describes a proposal bond and a two-hour challenge period; operational details should be checked against current platform information. Ambiguous wording, a delayed resolution, or a disputed outcome can affect both whether the bot’s interpretation matches settlement and when capital is released. The market’s complete resolution terms and named source therefore matter as much as the short title.
Execution and fees can erase a thin edge
A trade that looks favorable at a stale price can become unattractive at the executable price, after a partial fill, or once applicable fees are included. The bot should recalculate using the actual order conditions rather than assuming the initial quote remains available.
What does published arbitrage evidence show—and not show?
The 2025 paper “Unravelling the Probabilistic Forest: Arbitrage in Prediction Markets” by Oriol Saguillo, Vahid Ghafouri, Lucianna Kiffer, and Guillermo Suarez-Tangil distinguishes rebalancing arbitrage within a market from combinatorial arbitrage across related markets. The authors estimate that $40 million in realized profit was extracted in their analysis. That is a historical, study-specific estimate; it does not establish that a new bot can find the same opportunities now, execute them after costs, or reproduce those returns.
Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Repair Windows errors before they cause bigger problems3Fix the driver behind crashes, sound loss and screen glitchesThe paper also describes a possible inconsistency across exhaustive, mutually exclusive outcomes: their combined probabilities should equal 1, so conflicting prices can indicate an apparent arbitrage. In practice, a bot still needs to verify that the markets have compatible settlement definitions, that the relevant orders can be executed, and that fees and other costs do not remove the apparent opportunity.
Quick Recap
What should you verify before running a bot?
- Confirm the live fee schedule, market-specific fee fields, and whether the intended orders are maker or taker orders.
- Confirm current API documentation, access requirements, rate limits, and order-submission behavior.
- Inspect the complete resolution rules and source for each market rather than relying on the title alone.
- Test probability calibration and trading logic out of sample, including the effects of realistic fill prices and fees.
- Check whether platform access and automated trading are permitted where you are located; legality and access restrictions are jurisdiction-specific and are not established universally by the platform materials cited here.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
