# outcometick backtest

> An open-source tick-level backtester for Polymarket, in plain Python (also JavaScript/TypeScript): any binary market (elections, sports, crypto) from order-book data you recorded, or a ready-made archive of Polymarket and Predict.fun crypto Up/Down markets. Strategies are replayed against the order book event by event, fills walk the depth, markets settle at $1/$0 only on evidence of the outcome, and every fill pays the venue's taker fee. `pip install outcometick`; a free real archive day via `load_sample()`.

Key facts:
- Python 3.10+, no other dependencies: `pip install outcometick`; JavaScript/TypeScript and the `ot` CLI: `npm i -g outcometick`.
- `from outcometick.backtest import run, load_sample`; `run(StrategyClass, load_sample()).summary()`.
- No look-ahead: events arrive one at a time; a market's outcome is hidden until on_settle.
- The same engine runs locally (Python and JavaScript, identical results) and on the hosted backtester at https://outcometick.com/backtest.
- The data archive (Polymarket and Predict.fun order books, trades, Chainlink settlement streams): https://outcometick.com — https://outcometick.com/llms.txt
- AI coding tools: an MCP server ships in the npm package (`npx -y -p outcometick ot mcp`); it never submits a paid backtest.

## Docs

- [For AI agents](https://outcometick.github.io/polymarket-backtest/agents.md): The short path: what the package does, the smallest program that runs, the rules for a correct result, MCP.
- [Overview (README)](https://outcometick.github.io/polymarket-backtest/index.md): Install, the quick start, what the engine gets right, the API at a glance, your own data, limits.
- [Methodology](https://outcometick.github.io/polymarket-backtest/methodology.md): Every replay rule: decoding, the two books, orders and fills, settlement, fees, no look-ahead, the report, execution options, timers and feed latency, queue models, the recorder, spot and the locked TWAP.
- [How we verify it](https://outcometick.github.io/polymarket-backtest/verification.md): Engine-against-engine and decoder-against-decoder tests, hand-worked answers, and what is not proven.
- [API reference](https://outcometick.github.io/polymarket-backtest/api.md): run, run_template, sweep, fingerprint, load_sample, Result, SweepResult, Dataset, BacktestError, Strategy, Order.
- [简体中文 (README in Chinese)](https://outcometick.github.io/polymarket-backtest/zh.md): The overview in Chinese.

## Optional

- [Everything above in one file](https://outcometick.github.io/polymarket-backtest/llms-full.txt)
- [Python repository](https://github.com/outcometick/polymarket-backtest)
- [JavaScript/TypeScript repository](https://github.com/outcometick/polymarket-backtest-js)
- [SDK reference: ctx, events, orders](https://outcometick.com/docs/sdk)
- [Data archive and hosted backtester](https://outcometick.com/llms.txt)