Process
Build an algorithmic trading journal
A useful journal records more than wins and losses: it helps you separate strategy behaviour from operational mistakes.
Record the system version
Every journal entry should identify the EA version, broker, account type, symbol, timeframe, risk setting, and date. If you change an input or replace a file, start a new version label. Otherwise, results from different systems become impossible to compare.
- EA name and version or configuration hash.
- Account balance and equity at the start of the period.
- Risk per trade, maximum positions, and protection limits.
- Broker symbol, spread conditions, and platform status.
Separate strategy from operations
A losing trade that followed the rules is different from a trade opened with the wrong lot size, a disconnected terminal, or a manually changed stop. Tag these separately so you do not ‘fix’ a strategy to solve an operational problem.
Useful tags include valid trade, skipped signal, execution issue, configuration issue, manual intervention, and protection halt.
Review in batches, not emotionally
Review after a meaningful group of trades or a fixed calendar period. Look at expectancy, drawdown, slippage, missed trades, and whether the system followed its rules. Avoid changing a strategy because of one trade or one short week.
The purpose of a journal is not to create a perfect-looking equity curve. It is to create enough evidence to make the next decision with less guesswork.
If you cannot explain why a trade happened, your next improvement should probably be observability before strategy changes.
Discuss this article
Ask a setup question, share a backtest, or compare notes with other algorithmic traders. Use a display name; your email remains private.
Create an account or sign in above to join the conversation.
Loading discussion…