Hide the future, then trade the past like it's live

A test without coding means replaying old price candle by candle and trading it by hand against fixed rules. It's slow, but it needs nothing beyond a charting account. It also trains your eye, because every decision gets made as if live.

The danger is hindsight. Scroll through a chart you've already seen and your eye knows where price went. You'll spot the setups that worked and skip the ones that failed, without meaning to.

The second leak is vague rules. If an entry reads 'buy at strong support', you'll decide what counts as strong after you see the result. Every judgment made after the fact flatters the test.

Bar replay fixes the first problem and a rule sheet fixes the second. You need both in place before the first candle moves.

Put the whole strategy on one page before you open the chart. Each rule should be something a stranger could check with a yes or a no.

Cover the entry trigger, where the stop goes and where the target goes. Add the size, the symbols, the timeframe and the hours you'll trade. Then list the trades you're forbidden to take, because those are the ones that tempt you mid-test.

Test each rule by imagining two people reading it on the same chart. 'Enter when a 15-minute candle closes above the prior day's high' gives one answer. 'Enter on a strong breakout' gives two people two answers.

Once the sheet is written, freeze it. If you change a rule halfway through, the trades before and after belong to different strategies. Start the count again from zero.

Write the costs you'll assume on the sheet as well. A commission figure and a slippage figure, fixed up front, can't be nudged later to rescue a weak result.

In TradingView, open the chart, press Replay and click a starting candle. Everything to the right disappears. Step forward one bar at a time, or let it play at a slow speed.

Pick the starting date before you look at the chart. A list of random dates written in advance works well. If you scroll around first to find a nice spot, you've already seen the answer.

Make each decision before you advance. When the rules say enter, write down the entry, stop and target, then step forward. Do the same for exits. Never step ahead to check and then jump back. Close anything else that leaks the future, like a news panel or a watchlist showing today's prices.

Spread the sample across several symbols and different months. Fifty trades from one calm stretch only tell you how the rules did in that stretch. Mix in a choppy month and a trending one if you can find them.

Fifty trades sounds like plenty until you see how much each one moves the result. Each trade is 2 percent of the sample, so a single outlier can tilt the whole picture.

Say your rules risk 1R to make 2R. The breakeven win rate is 1 divided by 3, about 33 percent. Now suppose the replay gives 20 winners out of 50, a 40 percent win rate. Expectancy is 0.40 times 2R, minus 0.60 times 1R. That's 0.20R per trade.

Now flip three of those winners into losers. That leaves 17 winners, or 34 percent. Expectancy becomes 0.34 times 2R minus 0.66 times 1R, which is 0.02R. Three trades turned a decent edge into almost nothing.

So treat 50 as the point where the test starts to say something. If the result hangs on a handful of trades, keep replaying until it doesn't. A hundred trades halves the weight of each one.

Use one row per trade in a spreadsheet. Record the date, symbol, entry, stop, target, exit and the result in R. Note the time of each entry too, in case the results cluster in one part of the session.

Add a column for whether you followed the rule sheet, yes or no. Be strict. A trade taken on a hunch goes in the log marked no, and it doesn't count toward the 50.

Replay fills you at the exact price you clicked, which live trading never does. So subtract a realistic cost from every trade before you total it. Commission plus a little slippage on each side is a fair start. If the edge disappears after that, the rules aren't worth trading yet.

When the 50 are done, write the win rate, average winner, average loser and expectancy at the top. Those four numbers are what you'll hold your live results against later. The gap between replay and real fills is its own lesson, and only live trades can show it.

Ten steps. The first three happen away from the chart.