# BenchAlgo > BenchAlgo is a backtesting and validation platform for futures trading algorithms. It runs your strategy on real CME market data for ES, NQ and GC with trading costs, checks it for lookahead and overfitting, and scores it out of 10 across twelve validation screens. BenchAlgo is a research tool. It does not give investment advice and it does not place trades. Results come from simulated tests on past data and do not predict future results. Trading futures involves risk of loss. ## Product - [BenchAlgo](https://benchalgo.com/): What it is, how it works and results from our own strategies. - [Product guide](https://benchalgo.com/product-guide/): Direct answers about data, scoring, pricing, your AI, privacy, live trading and limits. - [Product guide as Markdown](https://benchalgo.com/product-guide.md): The same guide in a readable text format. - [Complete reference](https://benchalgo.com/llms-full.txt): The product guide and every other public page in one plain text file. - [Pricing](https://benchalgo.com/pricing/): Plans from $99 a month with credits for AI usage and backtests. Need more performance? Contact us. Your own Claude Code or Codex works too. - [FAQ](https://benchalgo.com/faq/): Answers about BenchAlgo: the data, what you see, privacy, costs and slippage, which AI you can use, credits, live trading and the sealed holdout. - [Guides](https://benchalgo.com/guides/): Practical guides to testing a trading strategy: lookahead bias, overfitting, random entries, slippage, AI-written code, ES, NQ and gold futures, comparisons. ## How a strategy is scored - [How scoring works](https://benchalgo.com/how-scoring-works/): Every strategy gets a score from 0 to 10, twelve validation screens with Passed, Weak or Failed badges, and a verdict: Lead, Validated or Rejected. - [Lookahead checks](https://benchalgo.com/lookahead-checks/): How BenchAlgo proves a strategy never used future data: truncation, perturbation, next-bar fills, live replay, random nulls, double costs and a sealed holdout. ## Test a backtest - [How to check a backtest for lookahead bias](https://benchalgo.com/guides/lookahead-bias/): Lookahead bias means a backtest used data it could not have had. Find it by cutting the future, changing the future, filling later and replaying bar by bar. - [How to tell if a backtest is overfit](https://benchalgo.com/guides/backtest-overfitting/): An overfit backtest fits noise, not an edge. Four tests expose it: walk-forward, neighbouring settings, probability of backtest overfitting and trial counts. - [Does your strategy beat random entries?](https://benchalgo.com/guides/random-entry-test/): A random-entry test compares a strategy with random twins that share its trade count, stops, targets and costs. How to build one and how to read the p-value. - [Backtest validation checklist: twelve checks](https://benchalgo.com/guides/backtest-validation-checklist/): Twelve checks to run before you trust a backtest, each with its formula: drawdown, cost stress, Monte Carlo, parameter stability, walk-forward, PBO, lookahead. - [Fills, slippage and fees in a futures backtest](https://benchalgo.com/guides/backtest-costs-and-fills/): A futures backtest is only as honest as its fills. How to model next-bar fills, stop-first bars, gaps, slippage per side and commission, and how to stress them. ## Markets - [ES futures backtesting](https://benchalgo.com/guides/es-futures-backtesting/): Backtest E-mini S&P 500 (ES and MES) strategies on up to 15 years of CME data with real costs. Contract facts, what to watch on ES, and how the test runs. - [NQ futures backtesting](https://benchalgo.com/guides/nq-futures-backtesting/): Backtest E-mini Nasdaq-100 (NQ and MNQ) strategies on up to 15 years of CME data with real costs. Contract facts, what to watch on NQ, and how the test runs. - [Gold futures (GC) backtesting](https://benchalgo.com/guides/gold-futures-backtesting/): Backtest gold futures (GC and MGC) strategies on up to 15 years of CME data with real costs. Contract facts, what to watch on gold, and how the test runs. ## Your AI - [How to backtest a trading strategy written by AI](https://benchalgo.com/guides/backtest-ai-trading-strategy/): AI writes a strategy in seconds and can overfit it just as fast. Keep data away from the model, sandbox its code and count every try, with Claude Code or Codex. ## Compare - [BenchAlgo vs building your own Python backtest](https://benchalgo.com/guides/benchalgo-vs-python-backtest/): When to write your own Python backtester and when to use BenchAlgo: data, fills, lookahead checks, overfitting tests, trial counts, markets and cost compared. - [BenchAlgo vs TradingView for backtesting](https://benchalgo.com/guides/benchalgo-vs-tradingview/): TradingView charts and tests Pine Script on your chart. BenchAlgo scores Python futures strategies across twelve screens. Fills, costs and lookahead compared. - [BenchAlgo vs QuantConnect for backtesting](https://benchalgo.com/guides/benchalgo-vs-quantconnect/): QuantConnect is an open-source, multi-asset platform you program and trade from. BenchAlgo scores futures strategies across twelve screens. Side by side. ## Ownership and policies - [About BenchAlgo](https://benchalgo.com/about/): What BenchAlgo is, how the pages and examples on this site are made, how we label backtested results, and how to report a mistake or ask a product question. - [Imprint and contact](https://benchalgo.com/imprint/): BenchAlgo is a product of Crowie s. r. o., Lipovník, Slovakia: company ID, VAT ID, register entry, founder and how to contact us. - [Press kit](https://benchalgo.com/press/): Press kit for BenchAlgo: a product description to quote, how to write the name, the logo and social image, wording for backtested results and a contact address. - [Terms of Service](https://benchalgo.com/terms/): The terms for using BenchAlgo: what the service is and is not, your account and content, acceptable use, credits and payment, withdrawal and liability. - [Privacy policy](https://benchalgo.com/privacy/): What personal data BenchAlgo keeps and why, which providers receive it, how long it is kept, and how to use your rights. No analytics and no tracking. - [Cookies](https://benchalgo.com/cookies/): BenchAlgo uses one essential cookie that keeps you signed in to the web app. The public website sets no cookies, and there are no analytics or ads. ## Account - [Create an account](https://benchalgo.com/register/): Free sign-up with name, email and a password of at least 10 characters. No card needed. - [Log in](https://benchalgo.com/login/): Log in to the web app. ## Facts and limits BenchAlgo is a research tool. It does not give investment advice and it does not place trades. Results come from simulated tests on past data and do not predict future results. Trading futures involves risk of loss. Backtests are hypothetical. No real money is at risk in them, they are chosen with hindsight, and most ideas fail. Fills are simulated: the next bar open, the stop first inside a bar, and gaps at the worse price. A live market can fill worse. Only ES, NQ and GC, with full-size and micro contracts. Strategies currently run on completed 1-minute bars. Tick and MBO tiers for strategies are planned. A strategy holds one position at a time and is flat by the close. The score covers the data it was run on. The most recent three months stay sealed for now, so no score has been judged on them. Contact support@benchalgo.com for product questions and corrections. ## Optional - [Sitemap](https://benchalgo.com/sitemap.xml): Machine-readable list of all indexable pages.