Expectancy is the average R outcome per trade over a sample — conceptually: (win rate × average win in R) minus (loss rate × average loss in R). It describes past decisions you logged honestly; it does not promise future rupees on NSE live accounts. Teach yourself expectancy thinking without fabricated numbers, without guru win rates, and without turning one good week into a lifestyle business plan.
If you cannot compute expectancy from your own practice log, you do not have an expectancy opinion — you have a mood.
The formula (concept only)
For a set of trades where each outcome is expressed in R-multiples:
Expectancy (in R per trade) ≈ average of all R outcomes
Expanded form:
(fraction of wins × average win R) + (fraction of losses × average loss R)
Losses carry negative sign. Example structure only — do not treat these as achievable targets:
- If many trades cluster near -1R and few wins near +2R, expectancy depends on the actual fractions — arithmetic, not hope.
- If average loss exceeds -1R because of stop slippage, your plan leaked.
Use the formula on your completed practice log when sample size is honest. Never paste illustrative percentages as if they were yours.
What expectancy thinking trains
Not optimism. Three habits:
- Log outcomes in R so wins and losses compare fairly
- Separate process quality from short-sample expectancy — good process, negative sample happens
- Avoid sample lies — excluding skips, deleting bad days, cherry-picking setup tags
Expectancy thinking is a review lens, not a entry signal.
Minimum sample before you say the word
Internal rules for practice traders:
| Sample | Allowed language |
|---|---|
| < 20 trades | "Too early; process only" |
| 20–40 | "Descriptive average R; tentative" |
| 40+ | "Hypothesis still alive or falsified" |
Publishing or believing expectancy from tiny samples is how retail traders overconfidence themselves into live size. Your journal can enforce the table even when social media cannot.
Build your own expectancy table (from your log only)
After 20+ process-valid replay decisions:
| Metric | Your log (fill manually) |
|---|---|
| Total decisions entered | |
| Outcomes in R (list or sum) | |
| Average R | |
| Count rule violations | |
| Average R on violation-free subset |
Compare last two rows. If violations flip sign of average R, your problem is process — not "market edge."
Do not share the table as proof of skill. Use it privately for the next plan revision.
What NOT to do (expectancy edition)
- Do not cite "60% win rate systems" from a thumbnail without source and sample
- Do not ignore costs in live expectancy while claiming replay expectancy
- Do not mix setup tags in one average — breakout expectancy and pullback expectancy are separate hypotheses
- Do not annualise one good month
- Do not adjust losing trades out because "I broke rules so it doesn't count" — split tagged subsets instead
Negative expectancy with good process
Possible over short samples and over long ones if hypothesis is wrong. Good process with negative expectancy means: keep process, pause or revise setup hypothesis.
Bad process with positive expectancy over 15 trades means: do not scale live — you got lucky while breaking rules. Replay again until process holds.
Expectancy vs risk of ruin (conceptual)
Even positive expectancy can blow accounts if size is too large relative to variance. Practice phase uses fixed 1R — ruin math is simplified. Live adds tail risk, gaps, halts.
Expectancy thinking without sizing discipline is incomplete. Pair with position sizing drills.
NSE practice caveats
Replay expectancy ignores:
- Real slippage and partial fills
- Emotional interference
- Regime change
- Costs and taxes
Tag practice logs expectancy-type: replay-theoretical. Downgrade confidence when translating to live.
Corporate actions, illiquid exits, and expiry pinning distort small samples on index-linked names. Segment those days in review.
Weekly expectancy review (15 minutes)
- Export or read last 20 R outcomes.
- Compute average R by hand or spreadsheet — no auto-dashboard required.
- One sentence: "Expectancy descriptive this week is X R; sample size Y; violations Z."
- No action if sample under threshold — only note.
Boring reviews beat dramatic conclusions.
Expectancy thinking without profit promise
The goal of practice expectancy thinking is to kill fairy tales early — "one more indicator will flip expectancy" — and replace them with countable R outcomes from your behaviour.
You may compute negative expectancy forever on a setup. That is valuable falsification. It saves live capital.
Segmented expectancy (same sample, clearer picture)
Compute average R separately per setup tag and per condition tag (trend day vs range day). Aggregated expectancy hides that pullbacks lose in chop while breaks lose on fade days — useful falsification. Segmentation needs larger total sample; do not segment into meaninglessness at 25 trades total.
When to stop computing
If computing expectancy becomes a daily mood ritual — checking average R after every trade — pause metrics for a week and return to process scores only. Expectancy thinking serves review cadence, not heartbeat monitoring.
Learn expectancy thinking from your own R logs on historical NSE practice — not fabricated stats. Join the waitlist.
Related reading
- The 300-Trade Experiment: How Much Can You Learn From 30 Days of Trading Practice?
- What Is Chart Replay? Backtesting vs Paper Trading
- Pullback vs Breakout Trading: Which Should You Practise First?
- How Many Trades Do You Need to Test a Trading Strategy?
Educational note
Not investment advice. Not stock tips. Practice results do not guarantee future results.