Searching Tick Data for Short-Term Trading Patterns

Summary

The article describes a high-throughput way to search for trading rules using tick histories rather than bars and indicators. It uses MetaTrader 5’s strategy tester, custom tick symbols, and a fast Expert Advisor that reacts to each tick. To reduce processing time, the author filters out ticks that do not change the algorithm’s trading state, then tests parameter variations across many symbols and ranks outcomes by the profit from the best time-of-day interval.

An example finds that restricting one system to an overnight window changes its measured result from a loss to a gain over the tested period. The author also reports that a version performed similarly in live trading before beginning to lose, with a drawdown of about a tenth of prior gains. The trading rules themselves are withheld, so the result cannot be independently reproduced from the article. The case illustrates the risks of optimization, short samples, and changing market behavior; it does not establish that tick-based searching reliably finds profitable systems.

Key ideas

• Tick-level histories can be used to search for patterns beyond bar and indicator methods.
• Filtering ticks that do not affect a fast trading algorithm can substantially reduce the testing workload.
• The described workflow tests many symbols and evaluates results by time-of-day intervals.
• A profitable optimized interval in a short historical sample may reflect overfitting rather than a durable edge.
• The example system later lost money live, underscoring that backtest performance can fail to persist.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.