Express the actual strategy
Code entry, exit, and sizing rules. Combine timeframes, retain state, and respond to partial fills, staged entries, or changing stops.
Investigate strategies across markets and timeframes, inspect individual signals, and connect your rules to live trading. We build on efficient candle processing so repeated analysis can be part of the everyday workflow.
AlgoCore began as a framework for evaluating trading patterns across broad market histories. The original thesis studied just over 500 equity symbols, requiring at least 20 years of history per company.
That workload shaped separate services for candles, indicators, and pattern detection, with reusable results. Researchers could vary the experiment, compare outcomes across stocks, and inspect the events behind the statistics.
Explore the research and resultsExperiments often revisit the same history and indicators. Our processing layer reuses prepared data and calculations, so changing a strategy does not mean rebuilding its view of the market.
Independent scenarios can run concurrently and share those results. Data processing and strategy execution can scale separately as you add markets, finer timeframes, or parameter variations.
At ten past the hour, an hourly candle’s final high, low, and close are still unknown. A simulation needs the values the strategy could actually observe at that moment.
Our processing model reconstructs longer candles from finer observations, with indicators following the candle as it forms. You can inspect the values behind a signal within a larger timeframe.
For crypto markets, we support one-second candles where the source provides that resolution.
Code entry, exit, and sizing rules. Combine timeframes, retain state, and respond to partial fills, staged entries, or changing stops.
Vary parameters across assets and historical periods. Compare events and inspect simulated orders, fees, and positions alongside the chart.
Carry strategy logic into live trading. We build and validate exchange integrations for order submission, fill updates, and position state.
Is a result consistent across stocks and historical periods, or concentrated in a few cases? Vary the experiment and trace each pattern or simulated trade through its candles, indicators, orders, and fees.
The tooling also exposes weak assumptions. The original research found modest effects for several traditional patterns and limited advantages from clustering. The value is being able to investigate those results.
Our custom systems build on AlgoCore and its companion libraries. The original framework, pattern library, and thesis are public. A delivered solution adds the integrations, strategy behaviour, and operational requirements of your project.
Explore the original AlgoCore projectBring the research question, data, or strategy you want to put to work.