What a Simulation actually tells you

Every playbook you build can be run across real market history before it goes anywhere near an account. That run is a Simulation, and what comes back is not a verdict. It is evidence.

Costs are off until you turn them on

Simulations fill at mid by default: no spread, no commission, no slippage. That is a deliberate starting point, because it shows you the strategy rather than a broker.

It is also the single most important thing to remember when you read a result. Put your own spread, commission and slippage in, using your broker’s real figures, and run it again. The gap between those two runs is what trading the idea costs.

You get the working, not just the answer

A Simulation returns the summary metrics, the equity curve, the drawdown and the complete trade log: every entry, every exit, every stop.

The trade log is the part worth your time. A headline figure cannot tell you that four of the losses came from one bad week in one session. The log can, and that is usually the thing you change.

How far back the data goes

History depth varies by timeframe: roughly five years on 5-minute data, ten on 15-minute through hourly, fifteen on 4-hourly, and about twenty on daily and weekly.

So a slow strategy gets a long window and a fast one gets a shorter one. Worth knowing before you conclude much from a test that only covers one kind of market.

Change one thing at a time

The loop that gets you somewhere is dull and it works: run it, read the trade log, change exactly one variable, run it again.

Move the stop. Narrow the session. Add the higher-timeframe filter you have been arguing with yourself about. Two runs that differ in one rule tell you what that rule is worth; two runs that differ in five tell you nothing.

Each version is kept, so you can go back to the one that read better rather than trying to remember what you had before.

What the grade means, and what it does not

bubi scores a result for consistency rather than raw profit, so one lucky run does not grade like a reliable one.

It describes what already happened. It is not a forecast, and neither is anything else on the page. A simulated result is not a promise about the future, and trading carries a real risk of loss whatever a backtest looked like.