Trading Practice6 min read

A Setup-Fit Scorecard You Can Fill After Thirty Days

Replay Trader Editorial Team

Educational content for Indian traders on deliberate practice, chart replay, and setup discovery. Not investment advice.

A month of practice is wasted if the close-out is a vibe. "Breakouts felt better." "I think I'm a pullback trader." Those sentences cannot be audited. A scorecard can.

This is a card you fill after a bounded block — ideally the 300-decision month, not after a loud weekend. It converts logs into a style hypothesis: one sentence about which setup × condition pair you appear to handle with cleaner process. It does not name a winner on NSE. It does not certify an edge. It does not tell you to trade live.

The setup-fit matrix is the practice design (you × setup × condition). Comparing setups through practice is the A/B block. This page is the sheet you actually tick when the block ends.

Scoring rules that keep rupees in their place

Fill from the log, not from memory of the green days.

  • Process first. Follow-plan Y/N is the primary score. Outcome in R is a description of the sample, not a vote.
  • Minimum cell size. Do not score a cell with fewer than 15 tagged decisions. Write thin and move on. Thin is an honest result.
  • Same constants. Only compare rows that shared timeframe, universe, and risk unit. Mixing cash-intraday pullbacks with overnight delivery holds is not a scorecard; it is two products.
  • Skips count. A scored skip is a decision. Idle watching is not. If a cell is all entries and no skips, note skip-starved.
  • One card per block. If you changed rules twice, the card is for the frozen stretch only.

Hypothetical arithmetic, not a finding: if cell A has 22 decisions with 18 follow-plan = Y (18/22) and cell B has 20 with 11 Y (11/20), A is the cleaner process cell in this sample. That ratio is not a win rate of the setup on the exchange.

The 15-row floor is a card convention, not a researched sample-size law. You may raise it. Do not lower it after you see which cell would "win."

The card (copy this)

Header. Block dates. Timeframe. Universe rule (e.g. liquid NSE large-caps — a practice universe, not a buy list). Frozen rule-set ID. Decision count in the frozen stretch.

Score each cell 0–2 on four process columns. Add nothing else on week one of using the card.

Setup tag Condition (pick from your closed list) n Plan followed (0–2) Skip honesty (0–2) Urge control (0–2) Invalidation honour (0–2) Cell total /8 Thin?
Breakout Open drive (9:15–10:15 practised)
Breakout Midday range
Pullback Open drive
Pullback Midday range
Reversal Any — only if you actually tagged it
Other (should be near empty)

How to turn a column into 0–2 (use your log, not mood):

Score Plan followed Skip honesty Urge control Invalidation honour
2 Follow-plan Y on ≥80% of that cell's rows Skips are reasoned and logged; not zero Urge tags rare and not acted Stops/invalidations as written
1 Mixed; a few "just this once" Some skips, some boredom fills Tags present; sometimes acted Occasional hope holds
0 Plan is optional Almost never skip, or skip only when scared after a loss Chase/revenge/peek dominate Invalidation rewritten after the candle

The 80% cut is a card convention, not a researched threshold and not a claim about profitable trading. You may pick 75% if you write it in the header before scoring. Do not pick it after you see the rupees.

Leave last-hour / closing-auction cells off the first card unless that was a frozen focus. NSE's last minutes are a different experiment.

From totals to one hypothesis sentence

Sort cells with n ≥ 15 by total /8. Ignore P&L order.

Write exactly one sentence in this shape:

In this block I appear to handle [setup] in [condition] with cleaner process than [other setup] in [condition]; I will keep that pair for the next test and treat the rest as stand-aside or a later block.

Examples of legal sentences (hypothetical):

  • "I appear to handle pullbacks in midday ranges with fewer plan-breaks than breakouts on practised opens; next block is pullback-only midday."
  • "No cell reached n = 15 except other; the hypothesis is that I do not have a setup yet."

Illegal sentences:

  • "Breakouts suit me."
  • "This is my edge on Bank Nifty."
  • "I should buy X on Monday."
  • "I am ready to go live."

If two cells tie, do not average them into a personality. Keep both as candidates or run a dedicated pullback versus breakout block next — one tag at a time.

Then let the data pick only if the card is filled. Preference without a card is still preference.

What a low score actually means

A low breakout-open score does not mean breakouts are "bad on NSE." It means you, in this sample, broke rules there. The market did not fail the exam. You documented your handling.

A high score does not mean you have found a niche for life. Thirty days is a thin climate. The hypothesis is a ticket to the next bounded test, or to a written plan that describes what you did, not what you wish you did.

Other filling up is a scorecard result: you did not constrain the month. Fix the tag list before you interpret fit.

Salaried traders often discover, on this card, that the open-drive cells are skip-starved because evening replay of 9:15 still carries the FOMO of a session they cannot attend live. That is useful self-knowledge. It is not a recommendation to quit the job or to trade the open with real rupees.

Failure modes of the card itself

  • Scoring from memory because the CSV is ugly.
  • Weighting R until the favourite cell wins.
  • Adding a fifth column called "confidence."
  • Filling cells you did not practise ("I know how I would score on reversals").
  • Publishing the card as proof. It is a private hypothesis generator.

Replay Trader can supply the reps — randomised historical NSE, hidden dates, candle-by-candle decisions, practice trades, optional AI review after the session. The card is yours. ₹999 for 30 days buys a block in which to fill it, not a result. Three hundred practice decisions can populate the cells; they cannot turn a high cell into expertise.

Fill the card from the log, then write one hypothesis sentence. Join the waitlist.

Educational note

Educational practice content only. Not investment advice. No guaranteed returns. Practice ≠ live profitability.