Backtesting and a trading journal answer different questions. Backtesting studies how a rule would have behaved on historical data; the journal documents how you actually execute that idea with real friction, decisions and conditions.
What backtesting adds
It lets you explore rules over a historical sample, estimate outcome distributions and decide whether a hypothesis deserves further testing. Its quality depends on data, assumptions and avoiding biases such as look-ahead.
What the journal adds
It records real execution: fills, fees, slippage, mistakes, plan adherence, size changes and discretionary decisions. Those details may be absent from an idealised backtest.
Why the results can diverge
- Real costs and liquidity.
- Entries or exits not filled at theoretical prices.
- Human mistakes or inconsistent rule interpretation.
- Market changes relative to the historical period.
How to use them together
A useful sequence is: hypothesis → backtest → controlled trial → journal → compare expected versus executed → new hypothesis.
Avoid constantly refitting the backtest to live results
If parameters are adjusted after every week, you can end up explaining the past without improving generalisation. Keep independent test sets when possible.