How to backtest a chart setup by hand: 100 trades in a spreadsheet

A written setup is a hypothesis. The cheapest way to find out how it behaves is to go back through old charts and apply the rules candle by candle, the way you would have live. Software can do this for some rules, but doing it by hand teaches you whether your rules are actually clear.

What you need

• A charting platform with replay or the ability to scroll candle by candle.
• A spreadsheet.
• One finished setup card. If you don't have one, start with the setup rules template.

The spreadsheet columns

Column	Example
Date and time (Brasília)	2026-03-12 09:00
Pair	EURUSD
Direction	Buy
Entry	1.0850
Stop	1.0820
Target	1.0910
Result in R	+2, −1, or the partial value
Rule followed?	Yes / No, and which line was unclear
Note	Screenshot link

Record results in R (multiples of the risk), not in money or pips. That lets you compare trades with different stop sizes, and it keeps the test independent of account size.

The process

1. Pick a start date at least a year back and hide everything to the right of it.
2. Move forward one candle at a time. When every line of the card is met, record the trade before you see the outcome.
3. Move forward until the stop or target is hit and fill in the result.
4. Repeat until you have 100 trades, or until you run out of chart.

The "rule followed?" column is the most valuable one. Every time you hesitate, a line of the card is too vague. Fix the wording, not the results.

Reading the numbers

Say 100 trades give 38 winners at +2R and 62 losers at −1R.

Measure	Calculation	Result
Win rate	38 ÷ 100	38%
Total	38 × 2 − 62 × 1	+14R
Expectancy per trade	14 ÷ 100	+0.14R
Longest losing streak	counted in the sheet	7 trades

Illustrative calculation, not an expectation of results.

At 1% risk, a 7-trade losing streak is roughly a 7% drawdown. Ask yourself whether you would keep following the card through that, because live trading feels much worse than a spreadsheet. A backtest also leaves out spread and slippage unless you add them; subtracting a fixed cost per trade in R is a simple way to be more conservative. Past behaviour on old charts does not guarantee anything about future charts.

Tools that help

Use the risk-reward calculator to check that each recorded trade really is the R you wrote, and the lot size calculator when you move to small live size.

What to test first

Start with one setup and one pair. The breakout and retest and the inside bar are good first candidates because their triggers are easy to define. Trading involves risk, and most retail traders lose money.