Client‑side Monte‑Carlo Lineup Simulator with Sleeper API

Описание и применение

Веб‑приложение, полностью работающее в браузере, использует API Sleeper для получения данных о лиге и реализует 10 000‑итерационный Монте‑Карло‑симулятор с лог‑нормальным распределением, чтобы оценивать вероятности победы без серверной инфраструктуры.

Задача

Fantasy platforms use single deterministic point estimates, ignoring the asymmetric, high‑variance nature of player scoring, which hampers optimal lineup decisions.

Результат

The app runs entirely in the browser, incurs zero hosting costs, and provides realistic probabilistic lineup evaluations with smooth UI performance.

Рабочий процесс

• 1. User enters Sleeper League ID
• 2. App fetches rosters, users, and league data via Sleeper REST endpoints
• 3. Static player metadata (~5 MB) is cached in IndexedDB and refreshed daily
• 4. For each start/sit permutation, a Web Worker runs 10,000 Monte‑Carlo simulations using a log‑normal model
• 5. Results (win probability, expected points) are displayed in the UI
• 6. User iterates lineup changes and re‑simulates

Доказательства

Instead of hosting an API server, database, or authentication service, the entire application runs in the user's browser and deploys statically to GitHub Pages.

Источник

For every start/sit permutation, the engine simulates 10,000 full matchups

Источник