Build a written trading plan

4.8.6 · Stock-Market › Trading Psychology

WHY this works: Human memory reconstructive hoti hai, photographic nahi. High-stress trading moments mein, tera brain rewrite kar deta hai ki tum "kya karna chahte the" — taaki jo tumne kiya usse justify kar sake. Ek written document immutable evidence hai jo is self-deception ko rokta hai.

The Core Components: What Must Be Written

Har component "I think" ko "I will" statements mein convert karta hai — measurable criteria ke saath.

1. Entry Rules: The Setup Checklist

SAARE criteria true hone chahiye (logical AND). Ek bhi "maybe" = NO TRADE.

Example Entry Rule (Momentum Breakout):

• Price 20-day high ke upar close kare
• Volume > 1.5× average daily volume
• ATR (volatility) > 2% taaki sufficient movement ho
• Market (index) aaj -2% ya usse bura nahi hai
• Mere paas < 3 open positions hain (concentration limit)

WHY checklist format? Subjective "looks good" ko objective yes/no tests mein convert karta hai. AND operator discipline enforce karta hai — tum "4 out of 5 kaafi hai" rationalize nahi kar sakte.

Criterion	Data	Pass?	Why This Step?
Price > 20D high	$50.10 close	✓	Breakout timing confirm karta hai
Volume	800K vs 600K avg	✓	Buyer conviction validate karta hai (1.33× close hai, lekin rule kehta hai 1.5×)
STOP		✗	Volume fail → NO TRADE

Even though 4/5 criteria passed, the plan blocks the trade. Yeh recency bias ko rokta hai — "main sirf is ek rule ko thoda bend kar lunga..."

2. Exit Rules: Pre-Decided Outcomes

Jahan:

• $R$ = risk per share (entry price - stop price)
• $k$ = reward-to-risk ratio (typically 2–3)

WHY two exits? Dono loss aversion (losers ko hold karna) aur premature exit (winners ko fear se bechna) se protect karta hai.

Derivation from First Principles: Exits ki zaroorat kyun hai bilkul bhi?

1. Asymetry of ruin: 50% loss ke baad recover karne ke liye 100% gain chahiye $\frac{P_{\text{final}}}{P_{\text{initial}}}=(1-0.5)(1+x)=1⟹x=1.0=100\%$

2. Exits ke bina Expected value undefined hai: $EV={\sum }_{i=1}^{\infty }{p}_i\cdot {\text{outcome}}_i$ Agar tum positions kabhi close hi nahi karte (infinite hold times), toh yeh calculate nahi ho sakta.

3. Stop loss $\max (\text{loss})$ ko bound karta hai: ${\text{Account}}_{\text{worst case}}={\text{Account}}_{\text{today}}-n\cdot R$ Jahan $n$ = number of positions. Isse drawdown predictable ho jaata hai.

4. Target ensure karta hai ki tum mean reversion se pehle gains realize karo: Agar price ek random walk with drift follow kare, $P_t=P_0+\mu t+\sigma W_t$, toh stop $S$ se pehle target $T$ hit karne ki probability hai: $P(\text{hit }T\text{ first})=\frac{S}{S+T}$ (No drift case ke liye). 2:1 target ~33% win rate breakeven deta hai.

Day 1: Price $47.50 tak gir jaata hai.

• Check stop: $47.50<$48$ → EXIT immediately
• Loss: $(50-47.50)\times 100=-5$250$
• WHY this step? Stop \2300$remainingcapitalkonexttradekeliyebachatahai.Iskebina,tum$40 tak hold kar sakte ho (-$1000 loss = 5× zyada bura).

Day 3 (alternate timeline): Price $54.25 tak badh jaata hai.

• Check target: $54.25>$54$ → EXIT immediately
• Profit: $(54-50)\times 100=+8$400$
• WHY this step? 2R gain lock in karo. "Aur zyada" ke liye hold karna aksar profits wapas de deta hai jab price reverse hoti hai.

3. Position Sizing: The Math of Survival

Jahan:

• $f^{\ast }$ = capital ka fraction jo risk karna hai
• $p$ = probability of win
• $q=1-p$ = probability of loss
• $b$ = reward-to-risk ratio

Derivation from scratch: Goal: Long-term growth rate $G$ maximize karna.

$N$ trades ke baad $W$ wins aur $L$ losses ke saath: $\text{Capital}=C_0(1+fb{)}^W(1-f{)}^L$

Log lete hain (kyunki hum exponential growth chahte hain): $\ln (\text{Capital})=\ln (C_0)+W\ln (1+fb)+L\ln (1-f)$

Growth rate per trade: $G=\frac{1}{N}[\ln (\text{Capital})-\ln (C_0)]=p\ln (1+fb)+q\ln (1-f)$

$G$ maximize karo $\frac{dG}{df}=0$ lekar: $\frac{pb}{1+fb}-\frac{q}{1-f}=0$

$f$ ke liye solve karo: $pb(1-f)=q(1+fb)$ $pb-pbf=q+qbf$ $pb-q=f(pb+qb)=fb(p+q)=fb$ $f^{\ast }=\frac{pb-q}{b}$

Practical rule: Volatility reduce karne ke liye half-Kelly ($f^{\ast }/2$) use karo.

Step 1: Full Kelly calculate karo $f^{\ast }=\frac{0.4\times 2.5-0.6}{2.5}=\frac{1.0-0.6}{2.5}=\frac{0.4}{2.5}=0.16$

WHY this step? Kelly theoretically optimal risk fraction batata hai jo growth maximize kare bina ruin ke.

Step 2: Half-Kelly safety margin apply karo $f_{\text{use}}=\frac{0.16}{2}=0.08=8\%$

Step 3: Risk per trade $\text{Risk}=0.08\times \$10000=\$800$

Step 4: Shares determine karo (agar stop entry se 2 door hai) $$\text{Shares} = \frac{\800}{$2} = 400 \text{ shares}$$

WHY this step? Yeh math guarantee karta hai ki tum 12 consecutive losses survive kar sakte ho (1/0.08 ≈ 12) account deplete hone se pehle — 40% win rate ke saath yeh almost impossible hai.

4. Market Condition Filters: When NOT to Trade

WHY filter conditions? Yeh trades block karte hain jab edge disappear ho jaata hai ya execution degrade ho jaati hai.

Derivation: High volatility mein, noise-based exits avoid karne ke liye stop distances widen karni padti hain: $\text{Stop distance}\propto {\sigma }_{\text{price}}$

Agar tum VIX spikes ke dauran fixed stops rakhte ho, false stop-outs increase hote hain: $P(\text{stopped out})=P(|P_t-P_0|>S)\approx 2\Phi \left(-\frac{S}{\sigma \sqrt{t}}\right)$

Jab $\sigma$ double hota hai, tumhare stops 2× zyada hit hote hain — even bina kisi true trend change ke.

Filters check karo:

• VIX = 35 (Friday close 22 tha) → BLOCKED
• Kyun? Tumhara backtested win rate VIX < 25 par 40% hai, lekin VIX > 30 par sirf 28%. Edge disappear ho gaya.

Bhale hi setup "perfect" lagta ho, plan tumhari judgment ko override karta hai. Yahi pre-commitment kaam karta hua hai.

5. Review Process: The Feedback Loop

Zyada frequent reviews = faster adaptation, lekin:

• Daily: Bahut noisy (randomness dominate karti hai)
• Monthly: Bahut slow (buri aadat solidify ho jaati hain)
• Weekly: Optimal balance

Weekly review checklist:

1. Trade log: Kya maine plan follow kiya? (% compliance)
2. Error classification: Kaun se rules toode? Kyun?
3. P&L attribution: Plan ke trades se wins/losses vs. deviations se
4. Plan adjustment: Kya data kisi rule change ko support karta hai?

WHY weekly? Central Limit Theorem: 20–40 trades/month ke saath, weekly samples (5–10 trades) patterns detect karne ke liye enough data dete hain — bina noise pe overreact kiye.

Trade #	Compliant?	Outcome	R-multiple
1	Yes	Win	+2.1R
2	Yes	Loss	-1.0R
3	No (chased)	Loss	-1.5R
4	Yes	Win	+2.3R
5	No (target ke baad hold kiya)	Win	+1.2R (0.8R wapas diya)
6	Yes	Loss	-1.0R
7	No (no stop)	Loss	-3.0R
8	Yes	Win	+2.0R

Analysis:

• Compliant trades: 4W-1L = 80% win rate, avg +1.48R
• Violations: 1W-2L = 33% win rate, avg -1.1R
• Insight: Deviations ne 3.6R cost kiya (= $360agar$100/R)

WHY this step? Indiscipline ki cost quantify karta hai. "Plan follow na karne se -$360" dekhna "mujhe disciplined hona chahiye" se zyada powerful hai.

The Psychology: Why Plans Fail (And How to Fix It)

WHY this feels right: Confirmation bias tumhe woh trades dikhata hai jo kaam karte. Tum woh 10 similar setups nahi dekhte jo fail hue.

The fix:

1. "Missed" trades ko alag track karo (unhe paper trade karo)
2. 30 occurrences ke baad, calculate karo ki kya unse tumhare stats improve hote
3. Agar haan, toh is setup ko plan mein add karo (formal rule change)
4. Agar nahi, toh tumne prove kar diya ki plan sahi hai

Math: Sample size $n=30$ ke saath, win rate ke liye confidence interval: $\text{CI}=p\pm 1.96\sqrt{\frac{p(1-p)}{n}}$

$p=0.5$ par, yeh ±18% margin deta hai. Ek 50% rule ko ek 65% rule se statistical confidence ke saath distinguish karne ke liye ≥30 samples chahiye.

WHY this feels right: Tum chahte ho ki plan "perfect" ho commit karne se pehle.

The fix:

• Aaj hi version 1.0 likho apne current best guesses ke saath
• Use label karo "Draft – 20 trades ke baad update karo"
• Revise karne ke liye ek bura plan hona > drift karne ke liye koi plan nahi hona

Analogy: Scientists perfect hypothesis ka wait nahi karte. Woh current best guess test karte hain aur data ke basis par refine karte hain.

Active Recall Checkpoints

Recall Isse ek 12-Saal-Ke Bachche Ko Explain Karo

Socho tum ek video game khel rahe ho, lekin har baar jab tum galti karte ho, real money chali jaati hai. Daraaona hai, hai na?

Ek trading plan bilkul waise hai jaise game strategy PEHLE se likh dena — khelna shuru karne se pehle. Tum decide karte ho: "Main tabhi jump karunga jab mujhe YEH enemy dikhe" (entry rules), "Agar meri health 20% se neeche aa jaaye toh main bhaag jaaunga" (stop loss), aur "50 coins collect karne ke baad main quit kar dunga" (target).

Likhte kyun hain? Kyunki jab tum game mein hote ho, dil zor se dhadakta hai aur tum koi bewakoofi ki chaal chaal sakte ho. Lekin agar tumne kal likh diya tha "Kabhi lava mein jump mat karo" jab tum calm the, toh tum woh rule padh kar khud ko bacha sakte ho.

Cool part yeh hai: har hafte, tum apne game replays review karte ho. Agar tumne apne rules follow kiye, tum zyaadaatar jeette ho. Agar tumne unhe ignore kiya, tum zyaadaatar haarte ho. Toh paper sirf advice nahi hai — yeh ek cheat code ki tarah hai jo sirf tab kaam karta hai jab tum ise exactly follow karo.

Concept Map