Understand Level-2 - market depth data — Hinglish notes

6.4.1 · Stock-Market › Order Flow & Tape Reading

Level-2 data KIYA hota hai?

WHY yeh exist karta hai: Modern exchanges orders ko price-time priority se match karti hain. Iske liye unhe har unfilled limit order ki ordered queue maintain karni padti hai. Level-2 simply woh queue hai jo tumhe visible banayi gayi hai.

Figure — Understand Level-2 - market depth data

Book KAISE structured hoti hai (matching rule se derive karo)

First principles se start karte hain: ek exchange buyer aur seller ko tab match karna chahta hai jab sirf prices cross karein. Isliye woh orders ko do sorted stacks mein store karta hai.

Bid side (buyers): sabse upar highest price se neeche sort ki gayi. Reason (WHY): jo buyer sabse zyada paisa offer kar raha hai woh pehle trade karne ka haqdar hai → highest bid ki priority hoti hai.

Ask side (sellers): sabse upar lowest price se upar sort ki gayi. Reason: jo seller sabse kam maang raha hai woh pehle bechne ka haqdar hai → lowest ask ki priority hoti hai.

Top of book = highest bid milta hai lowest ask se. By construction:

$Bid_{b es t} < Ask_{b es t} \Rightarrow Spread = Ask_{b es t} - Bid_{b es t} > 0$

Ek hi price level ke andar, orders time se queue hote hain (FIFO): jo order pehle aaya = pehle fill hoga. Isliye ek badi queue mein late join karne ka matlab hai ki us price par tumhe fill nahi bhi mil sakta.

Cumulative depth (woh quantity jo traders actually watch karte hain)

Har level size $q_{i}$ dikhata hai. Useful object hai cumulative depth — kitna total volume tum sweep kar lete agar book mein continuously buy/sell karte raho.

Ask levels ko prices $p_{1} < p_{2} < \dots$ maano with sizes $q_{1}, q_{2}, \dots$ . Offers ko hit karke $Q$ shares buy karne par, level $k$ tak cumulative shares available:

$C_{k} = \sum_{i = 1}^{k} q_{i}$

Size $Q$ ke market buy ke liye average fill price (marketable order levels ko eat karta hua fill hone tak):

$\overset{ˉ}{P} (Q) = \frac{\sum _{i = 1}^{k^{*}} p _{i} x _{i}}{Q}, x_{i} = shares taken from level i, \sum x_{i} = Q$

jahan tum level 1 poori tarah fill karte ho, phir level 2, etc., jab tak $Q$ level $k^{*}$ par exhaust na ho jaye.

Worked Examples

Stock XYZ ke liye yeh book snapshot consider karo:

Bid Size	Bid	Ask	Ask Size
300	100.00	100.05	200
500	99.95	100.10	400
800	99.90	100.15	1000

Forecast-then-Verify drill

Aage padhne se pehle: Tum Example 1 ki book mein 700 shares ka market buy place karte ho. Average price forecast karo.

Verify

200@100.05 + 400@100.10 + 100@100.15 = 20010 + 40040 + 10015 = 70065. $\bar P = 70065/700 = \$ 100.0929 $. Slippage ≈ \$ 0.0429. Tumhe predict karna chahiye tha ki average Example 2 se zyada hogi kyunki 700 > 500 deeper dig karta hai.

Common Mistakes (Steel-manned)

Active Recall

Recall Self-test (chupaao aur jawab do)

Resting book mein best bid best ask se kyun kabhi zyada nahi ho sakti?
Slippage formula derive karo aur batao ki har term positive kyun hoti hai.
Bada bid depth price rise guarantee kyun nahi karta?
Ek sentence mein Level-1 aur Level-2 data mein kya difference hai?

Recall Feynman: ek 12-saal ke bacche ko explain karo

Ek ice-cream stall imagine karo. Level-1 sirf tumhe batata hai aakhri cone ki kya price thi jo biki. Level-2 tumhe do queues mein jhaankne deta hai: ek line mein bacche hain jo kharidna chahte hain (har ek utha ke dikhaa raha hai kitne cones chahiye aur kya price pay karega) aur ek line sellers ki. Stall us bacche ko match karta hai jo sabse zyada pay kar raha hai us seller ke saath jo sabse kam maang raha hai. Agar tum ek saath BAHUT saare cones kharidna chahte ho, to tum pehle sabse saste sellers ko gobble karte ho, phir aur mehenge lene padte hain — isliye tumhari average price chadh jaati hai. Woh chadh hi slippage hai. Aur buyers ki badi line ka matlab nahi ki prices jump karein — woh bacche sirf wait kar rahe hain; woh kisi bhi second walk away kar sakte hain.

Connections

Order Flow & Tape Reading — tape executions hai; Level-2 intentions hai.
Bid-Ask Spread — residual gap jo yeh note derive karta hai.
Slippage & Market Impact — finite depth ka direct consequence.
Limit vs Market Orders — limit orders book build karte hain, market orders use consume karte hain.
Order Book Imbalance Signals — depth se pressure quantify karna.
Spoofing & Layering — kyun depth jhooth bol sakti hai.
Price-Time Priority — woh matching rule jo poori structure generate karta hai.

What does Level-1 market data show?

Best bid, best ask, aur last traded price with size — sirf top of the book.

What does Level-2 (market depth) data show?

Dono sides par resting limit orders ki full list, har price level par available size ke saath.

How are bids sorted in the order book?

Sabse upar highest price, descending (highest buyer ki priority hoti hai).

How are asks sorted?

Sabse upar lowest price, ascending (lowest seller ki priority hoti hai).

Why is the resting spread always positive?

Koi bhi crossing (bid ≥ ask) instantly match aur remove ho jaati hai, isliye sirf non-crossing orders rest karte hain, best-bid < best-ask rehta hai.

Define the spread formula.

Spread = Ask_best − Bid_best.

What is slippage?

Tumhare average fill price aur top-of-book price ke beech ka difference jab tumhara order multiple levels eat karta hai: (Σ p_i x_i)/Q − p₁.

What tie-breaks orders at the SAME price level?

Time priority (FIFO) — jo order pehle place hua woh pehle fill hoga.

Why doesn't large bid depth guarantee price goes up?

Resting bids passive/patient hote hain aur spoof ya pull ho sakte hain; sirf aggressive market orders (the tape) actually price move karte hain.

What liquidity is invisible in Level-2?

Hidden/iceberg orders aur dark-pool size — displayed depth true depth se kam dikhata hai.

Order-book imbalance formula?

(Q_bid − Q_ask)/(Q_bid + Q_ask), bounded in [−1, +1].

Which order type consumes the book vs builds it?

Market orders resting liquidity consume karte hain; limit orders book add/build karte hain.