For two years I added indicators. Every few months, a new one. RSI divergence, funding-rate overlays, a volume profile I never fully understood. My results didn't move.

What finally moved them was embarrassingly boring. Thirty minutes, every Sunday, looking at what I actually did the previous week.

Execution is the only thing you control

Markets hand you a distribution of outcomes. You don't pick it. What you pick is whether your trades matched your rules, whether your size was what you planned, and whether you sat on your hands when the plan said to sit.

That's execution. It's also the part that leaks quietly. A stop moved 0.5% here, a skipped entry there, one position at 1.5x because it 'felt right.' None of it feels like a big deal on the day.

Add it up across fifty trades and it's the gap between the strategy you backtested and the one you're actually running.

What I look at

Nothing fancy. I open the trade log and answer four questions, in writing, in a plain text file.

First: did every trade follow the rules? I mark each one yes or no. Second: where did my actual size differ from planned size, and why? Third: what was the single worst decision of the week, even if it made money? Fourth: what is one thing I'll do differently next week?

Only one thing. If I write five fixes, I make none of them.

I keep the file dead simple on purpose. One line per trade, a yes or no, a short note when the answer is no. If the process takes more than half an hour, I stop doing it by week three. I know, because that's what happened the first time I tried a fancy spreadsheet.

The notes are the real asset. After a couple of months they read like a diary of your own bad habits, in your own words, with dates.

Why it compounds

A new indicator gives you one more input to second-guess. A review removes a leak. Leaks are permanent until you find them, so fixing one pays every week afterward.

Here's a real example. Around week six of doing this, I noticed that most of my rule breaks happened on Fridays after 6pm. Not randomly. Tired, wanting the week to end on a green number. I started closing the laptop at 5. Rule breaks that month dropped from seven to two.

That's a sample of one trader and four weeks, so don't take the exact numbers as science. The pattern is the point. I would never have spotted it scrolling charts. It only showed up when I lined up the week in a list.

Why most traders skip it

Because it's unglamorous. Watching charts feels like work. Reviewing feels like homework, and homework about your own mistakes is worse.

Also, a review after a winning week feels pointless. That's exactly when it matters, since good weeks hide sloppy execution behind green numbers. I've had weeks where I broke three rules and still made money. The review flagged them. The P&L would have said everything was fine.

Most traders also review only losses. That's half the picture. A lucky win from a broken rule teaches the wrong lesson, and you'll repeat it until the market stops paying for it.

Try it this Sunday

Block thirty minutes. Open your log. Mark each trade yes or no against your own rules, find the one pattern, and write a single change. Set a timer so it doesn't swallow your afternoon.

Do it four weeks in a row before judging it. The first review is mostly discomfort. By the third, you'll start to see yourself clearly, which is more than most indicators ever gave me.

If you'd rather take more of the execution question out of your hands, the way we run systematic bots is laid out at v33systematic.com.