About GU Analyser

GU Analyser was built around a simple idea: trading claims should be testable. It is a place to learn how technical analysis works, turn an idea into an exact rule, and check that rule against historical market data.

What it is

GU Analyser combines four things in one place: courses and a reference library for technical analysis; charts with 60+ configurable indicators; a visual, no-code strategy builder with backtesting; and event analysis, which measures what followed a market condition across every time it occurred. It covers 400+ US stocks, ETFs and indices on its own data, and 14,000+ US stocks and ETFs with a subscriber's own Tiingo key. It is an analytical and educational tool: it gives no signals or recommendations, and no money is ever traded in it.

Who builds it

GU Analyser is built by Alessio Lupinelli, its founder and director, an engineer with a background in aerospace engineering — which shows in how the software is built: its calculations are checked against independent sources before they are trusted. It is not a financial qualification: GU Analyser does not give financial advice, and nothing on this site should be read as advice.

How the tools are checked

Indicators are calculated from published definitions and compared with independent implementations — TradingView's built-in indicators, the TA-Lib library, and calculations written by hand from each definition. Differences found this way are corrected and recorded.
No look-ahead. Every value on a candle uses only information available when that candle closed, and every trade in a backtest fills at the close of the candle on which its rule fires.
Costs are part of every backtest, set as a percentage per trade.

How the research is done

The Test Bench publishes original studies of measurable market ideas. Each states its question, universe, period, assumptions and comparison before the result, and the result is published whether it is positive, negative, mixed or inconclusive. Every study ends with what the evidence cannot establish, and the app can rebuild the test. Examples: Donchian Channel breakouts across 179 US stocks, the MACD crossover studies, the opening range breakout (ORB) studies, the VWAP and Anchored VWAP studies and the mean reversion studies of RSI, ConnorsRSI and Stochastic RSI.

Data

Built-in US market data comes from Marketstack: daily history of about ten years and hourly bars.
With a subscriber's own Tiingo key (Premium), the tools run on Tiingo data under the subscriber's own agreement with Tiingo: 14,000+ US stocks and ETFs, up to 20 years of daily history, and five-minute bars. The key and the data it fetches are never stored on GU Analyser's servers.
All data is historical. GU Analyser does not stream real-time prices.

Limits

A backtest or an event study describes what happened on past data under stated assumptions. It is not a forecast, and a rule that worked in one period can fail in the next. Results depend on the instruments, the period, the costs and the rules chosen — which is why each is shown alongside the result.

Company

GU Analyser Ltd · Registered in England and Wales, company number 17367615, incorporated 28 July 2026 · Registered office: 71-75 Shelton Street, Covent Garden, London, WC2H 9JQ, United Kingdom · Registered with the UK Information Commissioner's Office, ZC210697 · Contact: hello@guanalyser.com.

See the method for yourself

The Test Bench publishes each study with its question, method, comparison and limitations, and the app is open to anyone to rebuild a test — free on the S&P 500 ETF, no card.

Create a free account → or read the Test Bench →

This is educational material, not advice. It describes methods and conventions used to study markets on historical data. Nothing here is a recommendation to buy or sell anything, and a historical test does not show what will happen next. GU Analyser is an analytical and educational tool — no money is ever traded here.