Most trading journals log the trade: entry, stop, target, result. Almost none log the state the trader was in when the trade got taken, which is exactly the missing variable that explains why the same setup sometimes gets executed clean and sometimes gets executed badly. The chart didn't change. The trader did.
This is the part of the journal most traders skip, usually because it feels softer than the numbers columns next to it. It isn't. A trade taken angry and a trade taken calm can have identical entries, stops and targets on paper and produce entirely different results, because the state changes what happens after the entry: whether the stop gets honored, whether the target gets taken early out of fear, whether a fourth trade gets added that was never in the plan.
What to log besides the numbers
Alongside entry, stop and result, add three fields you can fill in fifteen seconds: energy (rested, tired, wired), state before the trade (calm, impatient, angry, bored), and confidence, a number from 1 to 10. None of this needs elaborate writing. A single word in each field, logged at the moment of entry rather than reconstructed after the fact, is worth more than a paragraph written from memory an hour later, because memory edits itself to match the outcome.
Thirty days is where the shape shows up
One entry proves nothing. Thirty entries usually show a pattern loud enough to act on. Pull the trades tagged tired and check their win rate against the trades tagged rested; for most traders the gap is larger than any edge they're chasing from an indicator. Pull the trades tagged angry, or taken within ten minutes of a stop-out, and total their P&L separately from everything else. Most traders who run this for a month find a specific, ugly number sitting in that column, and a specific number is far more persuasive than a general sense that revenge trades are probably bad.
Turn the pattern into a rule, not a resolution
"I'll try to trade less when I'm tired" is not a rule, it's a hope. "No new positions after two consecutive losses on under five hours of sleep" is a rule, because it can be checked and followed without requiring willpower in the moment. The whole value of thirty days of state journaling is converting a vague feeling, that you trade worse when you're off, into a specific, checkable trigger with a specific, checkable response. If the tired trades lose money at twice the rate of the rested ones, the rule writes itself: no session starts on under six hours of sleep, or size gets cut in half when it does.
Keep the format boring
A spreadsheet with a fixed set of columns beats a diary every time, because a diary invites narrative and narrative is where self-deception lives. Fixed fields, energy, state, confidence, planned or unplanned, result, force a fast, honest tag instead of a story that explains away the loss. Review it the same day you review your process scorecard, weekly, and look specifically at the state tags attached to your worst three trades and your best three. The pattern is usually visible within the first month, and it rarely flatters the trader who was hoping the market was the whole problem.