.png){kind=logo}
TL;DR
Most traders are stuck in the repository phase — trades are logged, but no real insight comes back
Mixing different setups in the same review is like changing the recipe every time you bake — you'll never know why it worked or didn't
Pick one setup, log it the same five ways every time, and don't review until you have 30 clean trades of that same setup
The five starter variables: strategy, market context, trend phase, session, emotional state — trend phase alone will change how you read your results
Consistent baseline first, experimentation second — nail the recipe before you start tweaking ingredients
I logged trades for eight months before I realized I was doing it wrong. Not the logging part — that was fine. The problem was what I was tracking, and what I was mixing together inside the same review.
The Problem
Most traders who journal end up in the same place. They import trades, tag the strategy, maybe add a note when something felt off. After a month they open the journal expecting a revelation — and get nothing useful back. The numbers are there. The trades are there. But there's no clear pattern, no signal, nothing to actually act on. So they close it, log a few more trades, and wonder why everyone keeps saying journaling is the edge they're missing.
The habit is there. That part's already working. It's what's inside the sample that needs adjusting.
Why It Happens
Think about a month of trades that looks like this: three scalps on SPY in the morning, a swing trade on AAPL held overnight, two earnings plays, one trade you took because someone called it out in a Discord, and six setups from your actual strategy taken across different sessions in both trending and choppy markets. That's not a data set. That's a collection of random decisions with a spreadsheet attached.
This is the cookie problem. Bakers know exactly which variable makes a cookie chewy versus crispy — it's the brown sugar ratio. But they only know that because everything else stayed the same. Same flour, same butter temperature, same oven, same bake time — batch after batch. The moment you start changing three things at once, you lose the thread entirely. You can't tell if the cookie improved because of the sugar or despite the oven. Most traders do this every single day. Different setup, different session, different market condition, different emotional state — and then wonder why the results feel random. The results feel random because the inputs were random. And that's actually good news — it means it's fixable.
The Fix
Before you worry about what to tag, lock down what you're trading. Pick one setup. Take it repeatedly, in similar conditions, until you have 30 clean examples. That's your cohort. Everything else gets logged separately — or skipped entirely while you're building the sample.
This is the part most journaling advice skips. They tell you to log more, log better, add more fields. What they don't tell you is that a mixed bag of setups doesn't give you data — it gives you noise. And reviewing noise feels productive right up until you realize nothing has changed.
Once the setup is consistent, the tags start to matter. Here's the starter set — five variables that will surface more insight than anything else in the early stages:
Strategy — the setup name, clearly defined before you enter. Not "breakout" — the specific version of a breakout you trade, with your specific criteria.
Market context — is the overall market trending, choppy, or range-bound? The same setup behaves completely differently depending on what SPY is doing.
Trend phase — early move, middle of the trend, or late and extended? This one is the quiet difference-maker. A breakout setup that works cleanly in the first push gets chopped apart when the move is already exhausted and everyone is piling in late. One variable. Completely changes how you read your own results.
Session — morning, midday, or afternoon. When you trade matters as much as what you trade.
Emotional state — this is the variable that makes everything else make sense. A losing trade on a great setup looks completely different when you see "anxious, chasing" next to it. Log it as close to the trade as you can, while it's still fresh.
Five tags. That's enough for now. Complexity before consistency just makes the signal harder to find.
After 30 tagged trades of the same setup, something shifts. You stop seeing individual results and start seeing conditions. Your VWAP reclaim hits at 70% in trending morning sessions and falls apart in choppy afternoons on the same setup. That's not a feeling anymore — that's your data telling you exactly when your setup is worth taking and when to sit on your hands.
The journal doesn't tell you if a trade was good. It tells you which conditions make your setup worth taking — and which ones you can confidently skip.
This is the foundation of what we covered in It's All Predictable — one trade is noise. Your edge only shows up across a large, consistent sample. The journal is how you build that sample correctly.
Once the cookies are consistent — once you have a clean 30-trade baseline on the same recipe — then you experiment. Add one filter. Adjust the entry trigger. Track one new variable and see if it moves the numbers. That's how you refine an edge without dismantling the one you already have. Same principle as a baker who's nailed the base recipe deciding whether more brown sugar makes it chewier. You only know because everything else stayed the same.
