Trading Records for Discretionary Traders

Discretionary trading without records is guessing. Records without honest reflection are just data. This chapter is the single-highest-leverage habit in retail trading: how to log every trade so that six months from now, your journal produces a better trader than any additional strategy would.

9.1 Two categories of records

• Trade records (the what) — every trade with entry, exit, P&L, setup type, and screenshot. Broker statements cover most of this
• Reasoning records (the why) — for every trade, what you saw, what you expected, and what actually happened. This is the discretionary trader's edge — nothing else builds skill

Most retail traders keep the first and skip the second. That's why most retail traders don't improve. Skill comes from reviewing the second, not the first.

9.2 What to record at entry (ex-ante)

Written BEFORE the trade executes or immediately upon executing:

• Instrument and timeframe
• Setup name — which of your 2–3 primary setups is firing? If none, why are you entering?
• Higher-timeframe bias — bullish / bearish / neutral, based on what
• The level — what price level is this reacting to, and why is that level significant
• Entry, stop, target — all three, before executing. R:R computed
• Screenshot — the chart at entry, before you see how it plays out
• Confidence score — 1 to 5. This becomes valuable data over time

9.3 What to record at exit (ex-post)

Written AFTER the trade closes:

• Exit price and exit reason (target hit, stop hit, time-stop, discretionary early exit)
• Realised R — the trade result in units of initial risk (+2R means you made 2× the amount you had at risk)
• What happened — 1–2 sentences on how the trade played out
• What you got right — the setup, the level, the timing, the R:R selection
• What you got wrong — anything you'd do differently. Be specific
• Lesson tag — a short label ("chased entry," "moved stop," "held past exit rule," "clean execution")

9.4 Ex-ante vs ex-post — the critical distinction

The most important distinction in trading records is between:

• Ex-ante records — what you thought and decided before the outcome was known
• Ex-post records — what you noticed after the outcome was known

Both matter. But they matter for different reasons.

Ex-ante shows the quality of your process. If you made a decision that would be correct 70% of the time, and it happened to be one of the 30% failures, that's a good decision with a bad outcome. Ex-ante records let you evaluate the process independently of the outcome. This is the only way to build genuine skill.

Ex-post shows the pattern of outcomes over time. What sets fail, what setups win, what times of day work, what market conditions favour you. This is where statistical edge is measured.

9.5 The minimal setup — one spreadsheet, one folder

You don't need software. A single Google Sheet + a folder of screenshots is enough:

Every row is one trade. Every trade takes 5 minutes to log properly. Skip weeks are more expensive than they feel — after 6 weeks of gaps, the journal loses its diagnostic power because the pattern is broken.

9.6 The weekly review — where the compound learning happens

Every Sunday, before market opens Monday:

1. Filter your journal to the past week's trades
2. Group by setup type. Compute win rate and average R per setup. Which setup made money? Which lost?
3. Read the ex-ante notes for the losing trades. What was the process error? Was the setup wrong, the level wrong, the R:R wrong, or the execution wrong?
4. Read the ex-ante notes for the winners. Was there a lesson tag pattern? Are the winners concentrated in a specific setup, time-of-day, or bias?
5. Pick ONE thing to focus on next week. One rule to enforce, one setup to skip, one habit to add. Not five things — one
6. Write that one thing at the top of next week's journal

Compounding this weekly for a year does more for a discretionary trader than any book, course, or indicator can.

9.7 Expectancy math — the formula everything reduces to

The single equation that determines whether your strategy makes money over the long run:

Expectancy per trade = (Win rate × Avg winner R) − (Loss rate × Avg loser R)

Example — the 1-2-3 pin bar strategy after 50 trades:
  Wins:              28  (56%)
  Losses:            22  (44%)
  Avg winner:        1.8R
  Avg loser:         −1.0R
  Expectancy = 0.56 × 1.8 − 0.44 × 1.0
             = 1.008 − 0.44
             = +0.57R per trade

Meaning: on average, every trade adds 0.57× your risk amount to
the account. Trade 100 times at 1% risk = ~57% account growth
before compounding, drawdown, and taxes.

Rules of thumb for interpreting expectancy:

Almost every retail trader who thinks they have an edge is actually running at 0 to −0.1R expectancy — winning weeks feel like edge, losing weeks feel like bad luck. Journalling 50+ trades and computing this formula honestly separates edge from confirmation bias.

9.8 A sample journal — 20 trades of a Nifty pin bar system

Here's what an actual journal snapshot might look like. Real numbers, hypothetical strategy — the shape is what matters, not the values.

Analytics from this journal:

• Win rate = 11/20 = 55%
• Avg winner = 1.87R
• Avg loser = −0.94R
• Expectancy = 0.55 × 1.87 − 0.45 × 0.94 = +0.60R per trade → strong edge
• Confidence-5 trades: 6 winners of 8 = 75% win rate. Confidence-3 trades: 3 winners of 6 = 50%. Signal: only take confidence-5 setups
• "Skipped confirmation" tag appears 2 times, both winners — but small sample. Keep tracking

Within 20 trades, patterns emerge that were invisible in individual sessions. This is why journals compound.

9.9 The weekly deep-dive — a step-by-step checklist

Every Sunday, block 30 minutes for this exact sequence:

1. Read every trade's ex-ante note. Do this before looking at outcomes. Ask: "Would I have taken this trade again with the info I had?"
2. Group by setup type. Compute win rate and expectancy per setup. Which one made money? Which broke even? Which lost?
3. Group by confidence score. Are high-confidence trades genuinely outperforming low-confidence ones? If not, your scoring rubric is broken
4. Group by time-of-day (intraday) or day-of-week (swing). Are certain hours / days concentrating losses? Consider filtering them out
5. Read lesson tags. Which mistake do you keep making? "Moved stop" 3 times in 20 trades = the single thing to fix next week
6. Compute deviation rate. % of trades where you didn't follow the plan. Target: < 5%. If it's 30%, your strategy isn't real — you're improvising
7. Pick ONE thing to fix next week. Not five. One rule to enforce, one setup to skip, one habit to add. Write it at the top of next week's journal

The compounding effect of doing this consistently for 6-12 months is what actually separates the surviving 5% of retail traders from everyone else. It's not talent; it's this loop.

9.10 Journal traps — things NOT to do

• Retroactive journaling. Filling in ex-ante fields hours after the trade with hindsight bias contamination. Journal at entry or don't journal — halfway is worse than nothing
• Overtracking. 30 columns per trade means you never journal every trade. 10-15 columns is the sweet spot
• P&L obsession. Weekly P&L is noise. Focus on process metrics (expectancy, deviation rate, execution quality). P&L follows
• Ignoring winning trades. Losers get scrutinised; winners get celebrated. But often winners contain more information — "why did this work?" is as useful as "why did that fail?"
• Deleting bad trades. The trades where you broke your rules ARE the most important ones to journal. That's the whole point

Kelly % = (Win rate − (1 − Win rate) / R) × 100

Where R = Avg winner / Avg loser (in R units).

Example: Win rate 60%, avg winner 2R, avg loser 1R:
  R = 2 / 1 = 2
  Kelly = (0.60 − 0.40 / 2) × 100
        = (0.60 − 0.20) × 100
        = 40% of account per trade