Futures Trading

1. What you'll learn and why it matters

So far this course has dealt with the cash market: you pay the full price of a share, it lands in your demat account, and you can hold it as long as you like. A futures contract is different. You agree today on a price for a trade that settles on a fixed future date, and you put down only a deposit (the margin), typically 11% to 20% of the position's value. That one difference changes the risk completely. A 5% move in the stock becomes a 25% to 45% move on the money you actually put in.

By the end of this chapter you will be able to:

• explain why futures exist (the problems with private forward contracts) and how an exchange solves them;
• read a real NSE futures contract: lot size, contract value, expiry, tick size and settlement type, as of October 2026;
• calculate the margin (SPAN + exposure) and leverage of a position, and see why leverage is dangerous;
• build a daily mark-to-market (M2M) ledger for a long Nifty trade and a short stock-futures trade from real August–September 2026 settlement prices, including margin calls;
• price a futures contract with the cost-of-carry formula and compare it with actual prices for Nifty, RELIANCE, HDFCBANK and SBIN on 30 Sep 2026;
• read open interest (OI) with price (long buildup, short buildup, short covering, long unwinding) and test how much it predicts (in our data: very little);
• calculate rollover percentage and rollover cost, and handle expiry and physical settlement safely;
• calculate beta and hedge a stock or a portfolio with Nifty futures (how many lots, and what the hedge actually did in September 2026);
• size a positional futures trade by lot and stop distance, and see why the "risk 1% per trade" rule needs ₹15–40 lakh of capital per lot.

2. Forwards: the ancestor, and its four problems

A forward contract is a private agreement between two parties to exchange an asset for cash on a fixed future date at a price fixed today. Varsity uses the example of a jeweller who wants to lock in the price of gold for a purchase three months away, and a gold importer who is happy to lock in today's selling price. Neither pays anything today. On the agreed date they exchange gold for the agreed amount, whatever the market price is by then.

2.1 The logic: transferring price risk

The buyer of a forward (the jeweller) is protected if prices rise and loses the chance to buy cheaper if prices fall. The seller (the importer) has the opposite exposure. Each side gives up some upside to remove uncertainty. At expiry there are only three outcomes: price above the agreed price (buyer gains, seller loses the same amount), below it (the reverse), or equal (nobody gains). One side's gain is exactly the other's loss.

2.2 The four problems with forwards

3. What a futures contract fixes

A futures contract keeps the core of a forward (agree a price today for a trade at a future date) and moves it onto an exchange. The exchange's clearing corporation (NSE Clearing Ltd for NSE) becomes the counterparty to every buyer and every seller. That single change, plus standard terms, margins and daily settlement, fixes all four problems.

Two prices now matter. The spot price is the price in the normal cash market. The futures price is the price of the futures contract. They move almost together, because arbitrage (Section 11) keeps the gap small, but they are not equal. The gap is called the basis.

4. Contract specifications on NSE today

Every futures contract is fully defined by its specification. Here is what each term means, with the values in force as of October 2026 (each one checked against the sources listed in Section 21):

Our data confirms these rules. In the August–September 2026 bhavcopy, NIFTY futures prices move only in multiples of ₹0.10 and BANKNIFTY in ₹0.20. The September contract's final settlement price on 29 Sep was 22,716.20, exactly the Nifty's closing value that day.

4.1 Real contract data, 30 Sep 2026

The table below is built from the NSE F&O bhavcopy for 30 Sep 2026 (October contracts). The SPAN percentage comes from NSE Clearing's SPAN risk-parameter file for the same day. Section 6 explains the margin columns.

A few things stand out. One NIFTY lot is a ₹14.76 lakh position. Even the "smallest" stock-futures lot, VEDL (1,150 shares), was worth ₹2.96 lakh. Open interest (OI, Section 13) in the NIFTY Oct contract was 2,77,155 contracts, and 58,953 contracts traded that day.

4.2 The contract cycle: near, next and far month

At any moment three monthly contracts trade: the near (current) month, the next (mid) month and the far month. When the near month expires, the next month becomes the near month and a new far month is listed. Liquidity is concentrated in the near month. On 30 Sep 2026, NIFTY Oct futures traded 58,953 contracts, Nov 5,653 and Dec just 1,933.

5. Your first futures trade: long, short, square-off

When you buy (go long) one lot of futures, you commit to buying the lot at today's futures price on expiry. When you sell (go short), you commit to selling it. You do not have to wait for expiry. At any time you can square off by taking the opposite trade: sell what you bought, or buy back what you sold. Your obligation then passes to whoever took the other side, and your margin is released.

The payoff is linear. Every 1-point move in NIFTY futures is ₹65 per lot, in either direction, with no ceiling and no floor. Futures are also a zero-sum game before costs: every rupee one side gains, the other side loses. After brokerage, STT and other charges, the participants as a whole lose money, which is one reason most individual traders lose.

6. Margins: SPAN, exposure and peak margin

6.1 Why margins exist

The clearing corporation guarantees that the winner gets paid, so it must make sure the loser can pay. It does that in two ways: (1) it collects a deposit, the initial margin, from both buyer and seller before the trade, and (2) it settles gains and losses every day through mark-to-market (Section 8), so losses can never pile up unnoticed.

• SPAN margin (Standard Portfolio Analysis of Risk) is computed by NSE Clearing every day. It simulates 16 price-and-volatility scenarios and charges the worst loss among them. The size of the price move it tests (the price scan range) is set from recent volatility so that the loss over the margin period of risk (two days, because an M2M loss is only collected the next morning) is covered with very high confidence (99% or more). More volatility means higher SPAN.
• Exposure / Extreme Loss Margin (ELM) is a buffer on top. NSE Clearing's current rates are 2% of notional value for index derivatives and 3.5% for stock derivatives.

Your broker's margin calculator will show slightly different numbers. SPAN changes every day and broker figures include any additional or ad-hoc margins. Use the broker's number for real trades. Use the method above to understand why margin is what it is.

6.2 How margin works in India today (2026)

• 100% upfront margin. Under SEBI's peak-margin framework (fully phased in by Sep 2021), you must have the full SPAN + ELM in your account before taking the position. Clearing corporations take four random intraday snapshots of every client's margin, and any shortfall attracts a penalty.
• Maintained every day. Varsity's older explanation treats SPAN as a "maintenance margin": you get a margin call only when your balance falls below SPAN. In today's Indian framework, the required margin on an open position is the full SPAN + ELM every day. If M2M losses take your balance below it, you have a margin shortfall, and the broker will ask you to add funds (a margin call) or square off your position. Check your broker's risk-management (RMS) policy.
• Product types. Use NRML to carry a futures position overnight (that is what a swing trader needs). MIS is intraday-only and is auto-squared-off near the close. Since the peak-margin rules, brokers cannot give extra intraday leverage beyond SPAN + ELM, so MIS no longer meaningfully reduces margin. Zerodha discontinued bracket orders in 2020, and under peak-margin rules cover orders no longer carry the large margin discount they once did. Varsity's sections on BO/CO margins are now mostly historical.
• Calendar spreads (long one month, short another, same underlying) get much lower margin. NSE Clearing levies ELM on only one-third of the far-month value. The benefit is withdrawn on expiry day for a spread whose one leg expires that day. This has applied to index derivatives since Feb 2025 (SEBI's Oct-2024 measures) and, from 5 May 2026, to single-stock derivatives as well.

7. Leverage: the calculation and the danger

Margin lets you control a large position with a small deposit. The ratio is your leverage:

8. Mark-to-market: a full daily ledger for a real Nifty trade

8.1 The idea

With a forward, gains and losses build up silently until expiry. With futures, the clearing corporation settles them every day. At the end of each day your position is revalued at the day's settlement price. If it went your way, the gain is credited to your account. If not, the loss is debited. Next day, the reference price is reset to that settlement price. This is mark-to-market (M2M). It stops losses from piling up and is what makes the clearing corporation's guarantee credible.

8.2 The trade

8.3 The full ledger

8.4 A margin call, step by step

Over the trade there were 12 margin calls, and the trader had to add ₹94,324 on top of the original ₹2,00,000. At expiry the margin was released and the account held ₹1,61,477 (= 2,00,000 + 94,324 − 1,32,847). The futures price fell only 8.25%, yet the loss equalled 72.8% of the initial margin. Of the 40 trading days after entry, 26 were losing days and 14 winning days. The worst single day was 28 Sep (−₹23,861.50).

8.5 The same trade with a stop-loss

9. Shorting with futures: a real stock-futures ledger

Shorting means selling first and buying back later. You profit if the price falls and lose if it rises. In the Indian cash market you can short shares only intraday: an unsquared short leads to an auction and penalties, because you have no shares to deliver. (The exchange's Securities Lending and Borrowing (SLB) scheme does allow borrowing shares to short for longer, but it is thinly traded and rarely practical for a retail trader.) In futures, a short position is just as easy to hold overnight as a long, with the same margin and the same M2M (with the sign reversed). That makes futures the main tool for positional shorting.

10. Index futures, liquidity and impact cost

NIFTY futures take their value from the Nifty 50 index. Varsity gives several reasons why index futures suit beginners better than single-stock futures. The index is diversified across 50 stocks, so one company's bad news rarely moves it much. It is very hard to manipulate. Its volatility is lower than most single stocks, so margin is a smaller % (9.34% SPAN for NIFTY vs 14–47% for stocks on 30 Sep). It is cash-settled, so there is no delivery risk. And it is the most liquid contract.

10.1 Impact cost

Every order book has a bid (best price a buyer is offering) and an ask (best price a seller wants). If you buy at the ask and immediately sell at the bid, you lose the spread. Impact cost measures how far your execution price is from the "ideal" mid-price. It is a direct measure of liquidity.

11. Futures pricing: cost of carry, basis, contango and backwardation

11.1 The logic: why futures usually trade above spot

Imagine two ways to own Nifty on 27 Oct. (a) Buy the shares today: you pay the full price now, give up the interest that money could have earned, and receive any dividends paid before 27 Oct. (b) Buy Oct futures: you pay (almost) nothing now, can keep the money in Treasury bills earning interest, but receive no dividends. If (a) and (b) are to cost the same, the futures price must equal the spot price plus interest minus dividends. If it didn't, arbitrageurs would buy the cheaper one and sell the dearer one until the gap closed. This relationship is the spot–futures parity, or cost-of-carry model.

11.2 Fair value vs actual for NIFTY and three stocks (30 Sep 2026)

How to read the table:

• NIFTY and RELIANCE sit almost exactly on fair value. RELIANCE Oct is 0.05 below its 1,191.85 fair value. These are the most arbitraged contracts in India, so this is the cost-of-carry model working.
• HDFCBANK and SBIN trade a few rupees rich (₹3.78–5.84 above fair). That is about 0.4–0.8% of price, and it looks large when annualised over only 27 days. Possible reasons: strong demand to go long through futures, the cost and capital needed to arbitrage (next subsection), and timing noise. The futures settlement price is a last-30-minute average while the spot price is the closing price, so a one-rupee gap becomes ~1.9% p.a. when annualised over 27 days.
• Farther expiries carry a larger basis (more days of carry). The NIFTY Dec contract also shows a bigger % gap from fair value, partly because it is thinly traded (1,370 contracts of OI vs 2,77,155 for Oct).

11.3 Dividends make futures cheaper: MARUTI, August 2026

The \(-d\) in the formula matters when a stock goes ex-dividend before the contract expires. The holder of shares gets the dividend and the futures holder doesn't, so futures trade lower by roughly the dividend. MARUTI went ex-dividend on 7 Aug 2026 for ₹140 per share. Watch what happened to the basis:

• On 5 and 6 Aug, both MARUTI contracts traded below spot (Aug contract basis −156 and −165). That is backwardation, and the cause is the ₹140 dividend: subtract it and the fair values (14,092.88 and 13,980.12) are much closer to the actual prices.
• On the ex-date (7 Aug), spot no longer included the dividend, and the Aug basis jumped from −164.66 to −19.03. Spot fell 42.63 that day while the Aug futures rose 103. The futures had already priced the dividend in.
• Lesson: never mistake a dividend discount for a bearish signal, and never buy spot / sell futures "for the basis" without checking the corporate-action calendar.

11.4 Basis converges to zero at expiry

As expiry approaches, there are fewer days of interest left, so the basis shrinks. On expiry day futures and spot must converge, because index futures settle at the closing index value and stock futures settle through delivery. The NIFTY Oct contract's basis was mostly 250–380 points in mid-August (about 70–85 days of carry) and 86.8 points on 30 Sep (27 days). Day to day it is noisy: on 3 Aug it was only 112.8. The Sep contract's basis was exactly 0 at its 29 Sep final settlement.

11.5 Premium, discount, contango, backwardation

11.6 Cash-and-carry arbitrage (and why it is not free money)

If futures trade above fair value, you can buy spot and sell futures at the same time. At expiry the two converge, so you lock in the gap whatever the market does. This is cash-and-carry arbitrage, and it is what keeps futures near fair value. A real check:

12. Calendar spreads

A calendar spread is long one expiry and short another expiry of the same underlying. Because the two legs move almost together, the position's risk (and margin) is far smaller. You are betting only on the difference between the two prices. In theory that difference is the cost of carry between the two expiry dates:

13. Open interest and the four buildups

13.1 What open interest is

Open interest (OI) is the number of contracts that are currently open: entered and not yet squared off or expired. Every contract has one long and one short, so OI counts pairs. OI rises only when a new buyer and a new seller both open positions. It falls when an existing long and an existing short both close. It stays the same when an existing holder simply passes the position to a new trader.

In the bhavcopy, OpnIntrst is reported in units (shares or index units), so divide by lot size to get contracts. When you study OI over a month, add up all expiries. Otherwise rollover (Section 14) looks like a huge OI collapse in one contract and a surge in the next.

13.2 Reading OI with price

On its own OI tells you nothing about direction, because every long has a short. Paired with price, it hints at who is acting:

Varsity also gives a companion table for volume with price. Rising price on rising volume is bullish. Falling price on rising volume is bearish. Price moves on falling volume suggest the trend may be tiring. And it warns that abnormally high OI built up during a fast move signals heavy leverage: a small trigger can then cause a violent unwind.

13.3 All stock futures, September 2026 series

Classifying the 210 stock futures by price change and total-OI change from 31 Aug to 28 Sep gives: Short buildup 118, Long unwinding 68, Short covering 13, Long buildup 11. This was a falling market, so most stocks were on the left of the chart. Clear examples from each quadrant:

• Long buildup: DRREDDY price +5.8%, OI +22.0%. BLUESTARCO +7.6%, OI +28.2%.
• Short buildup: POLICYBZR −38.6%, OI +165.0%. HAVELLS −12.8%, OI +45.4%.
• Short covering: ADANIPORTS +9.5%, OI −20.1%. ITC +3.8%, OI −12.4%.
• Long unwinding: TCS −13.7%, OI −11.1%. ASIANPAINT −8.8%, OI −16.3%.

13.4 Honest test: does the daily buildup predict tomorrow?

Many websites publish daily "long buildup" lists as trading signals. We tested that on our data: every stock-futures day in Aug–Sep 2026 (excluding expiry days, when OI falls mechanically), classified by that day's price and total-OI change, then looked at the next day's spot return.

Across all observations, 42.8% of next days were up (the market was falling). "Long buildup" days were followed by an up day only 38.0% of the time, worse than average. "Short buildup" days were followed by up days 47.5% of the time, better than average. If anything, in this sample one-day buildups showed a slight short-term mean reversion, not continuation. Two months of one falling market proves nothing in general. But it is enough to show that OI buildup is context, not a signal. Use it to understand whether a move is driven by new positions or by exits, and combine it with price structure (Chapters 3–5).

14. Expiry, rollover and physical settlement

14.1 Rollover

Futures expire monthly, but a positional view may last longer. Rolling over means closing the expiring contract and opening the same position in the next month. Most traders roll in the last few days. The rollover % tells you how much of the expiring OI was carried forward:

Reading rollover: a high rollover with price falling (as with NIFTY in September, combined with short buildup) means traders are carrying bearish positions into the next month. A low rollover means traders are closing out. Compare it with the previous month's figure: NIFTY's Aug rollover was 77.4% vs 74.3% in Sep, and RELIANCE's 94.7% vs 93.7%.

14.2 Physical settlement of stock futures

Since the October 2019 expiry, all stock F&O contracts are physically settled. If you hold a stock future at expiry:

• Long futures: you must take delivery and pay the full contract value. 1 lot of RELIANCE at the Sep final settlement (1,182.01 × 500) = ₹5,91,005.
• Short futures: you must deliver the shares. If you don't have them, you face an auction and penalties.
• SEBI mandated this (circular of April 2018, phased in) to curb excessive speculation and expiry-day manipulation by forcing positions to be backed by real cash or shares.
• Margins rise near expiry. Zerodha's policy, for example, raises the margin on stock futures and short options on expiry day to 50% of contract value or 1.5× NRML margin, whichever is lower. Brokers may square off positions that lack the funds, and charge a fee for it.

15. Hedging with futures and beta

15.1 What hedging is, and what it can't do

A hedge is a position that gains when your main position loses, so the combination is less volatile. With futures, you hedge a long stock holding by shorting futures. Typical reasons: you hold shares long-term but fear a short-term event (budget, election results, a global shock). Selling would trigger capital-gains tax and force you to time your re-entry.

Varsity separates two kinds of risk:

• Systematic (market) risk: moves that hit all stocks together (interest rates, elections, global sell-offs). It can be hedged with index futures.
• Unsystematic (company) risk: news specific to one company (results, fraud, management change). It cannot be hedged with index futures. It can only be reduced by diversification, and Varsity notes the benefit of adding more stocks fades after about 20 stocks.

15.2 Hedging a single stock with its own futures

15.3 Beta: how much a stock moves with the market

To hedge with index futures you need to know how sensitive each stock is to the index. That sensitivity is beta: the slope of a regression of the stock's daily returns on the index's daily returns.

The market's own beta is 1.0. β > 1 means the stock tends to move more than the index (high beta). β < 1 means it moves less (low beta). Using daily returns from 1 Sep 2025 to 31 Aug 2026 (247 days), our data gives: LT 1.43, HDFCBANK 1.26, TATASTEEL 1.12, SBIN 1.06, ICICIBANK 0.94, RELIANCE 0.89, TCS 0.86, INFY 0.79, HINDUNILVR 0.74, ITC 0.65, BHARTIARTL 0.63.

Beta has limits. It is estimated from the past, it changes over time, and correlations are often modest (INFY's correlation with Nifty was only 0.35 over this year). Treat it as an estimate, not a constant.

15.4 Hedging a portfolio with NIFTY futures

16. Positional futures for a beginner: capital, sizing, gaps

16.1 Why the 1% rule is hard with big lots

Chapter 11 recommends risking at most about 1% of capital per trade. For futures, the position-sizing formula becomes:

Practical options if your capital is smaller: (a) stay in the cash market for swing trades, where you can buy exactly the number of shares your risk allows; (b) use futures only for occasional hedging; (c) accept a 2% risk per trade and a tighter, structure-based stop, knowing you will be stopped out more often; or (d) build capital first. Never solve the problem by dropping the stop.

16.2 Gap risk: when stops don't protect you

A stop-loss order can only fill when the market trades. If bad news arrives overnight, the market can open far beyond your stop, and you are filled at the open. With leverage, that one gap can cost more than your whole planned risk.

Ways to manage gap risk: hold fewer lots than your risk budget allows before known events (results, policy, elections, budgets); prefer index futures over single stocks, since index gaps are smaller; and keep spare cash for margin calls. If you are long stock futures, consider a protective put (Chapter 10).

16.3 Costs (summary; details in Chapter 13)

From 1 Apr 2026, STT on futures is 0.05% of the sell-side value (up from 0.02%, Budget 2026). On one NIFTY lot (~₹14.8 lakh) that is about ₹738 per round trip in STT alone, before brokerage, exchange and clearing charges, SEBI fees, stamp duty and GST. Rollovers add four legs per month. framework.costs.futures_charges computes them. For tax, futures P&L is generally treated as non-speculative business income. See Chapter 13.

16.4 A beginner's rulebook for positional futures

1. Learn on paper first. Paper-trade or trade cash-market equivalents for at least 3 months and keep an M2M ledger like Section 8.
2. Start with NIFTY futures only. Most liquid, cash-settled, lowest margin %, no single-company gaps.
3. Size by risk, not by margin. Lots = risk budget / (stop distance × lot size). Keep account equity at least 3–5× the margin of the positions you hold.
4. Every trade has a pre-defined stop, placed beyond a logical level (swing low/high, 2×ATR). Exit when it's hit. Never add margin to "give it room".
5. One or two positions at a time, and never all in the same direction across correlated stocks (five long bank futures is one trade, not five).
6. Close or roll stock futures 2–3 days before expiry to avoid physical settlement and expiry-day margin jumps.
7. Cut size before known events. Budgets, RBI policy, results, elections.
8. Check costs and liquidity: bid-ask spread, depth, STT.
9. Track your expectancy (Chapter 11) over at least 30 trades before increasing size.
10. Remember the base rate. 66% of individual futures traders and 87.7% of all individual F&O traders lost money in FY26.

17. Try it yourself

These snippets need a copy of the course repository and Python with pandas and numpy (setup steps are in its framework/README.md). Run them from the repository's root folder, for example with .venv/bin/python, because the file paths are relative to that folder. Each snippet reproduces numbers from this chapter. The full analysis is in course/scripts/ch09_figures.py.

17.1 Contract value, basis and implied carry from the bhavcopy

import pandas as pd
f = pd.read_csv("data/fno/futures_aug_sep_2026.csv", parse_dates=["TradDt", "XpryDt"])
d = f[(f.TradDt == "2026-09-30") & (f.TckrSymb.isin(["NIFTY", "RELIANCE", "HDFCBANK", "SBIN"]))].copy()
d["days"] = (d.XpryDt - d.TradDt).dt.days
d["contract_value"] = d.SttlmPric * d.NewBrdLotQty
d["basis"] = d.SttlmPric - d.UndrlygPric
d["fair"] = d.UndrlygPric * (1 + 0.0552 * d.days / 365)          # r = 91-day T-bill
d["implied_carry_%"] = (d.SttlmPric / d.UndrlygPric - 1) * 365 / d.days * 100
pd.set_option("display.width", 200)
cols = ["TckrSymb", "XpryDt", "days", "UndrlygPric", "SttlmPric", "fair", "basis",
        "implied_carry_%", "contract_value"]
print(d[cols].sort_values(["TckrSymb", "days"]).round({"fair": 2, "basis": 2, "implied_carry_%": 2}).to_string(index=False))
# NIFTY Oct: spot 22620.45, settle 22707.3, fair 22712.82, implied 5.19%

17.2 Daily M2M ledger for any futures position

import pandas as pd
f = pd.read_csv("data/fno/futures_aug_sep_2026.csv", parse_dates=["TradDt", "XpryDt"])

def ledger(sym, expiry, side, entry_date, exit_date, lots=1):
    s = f[(f.TckrSymb == sym) & (f.XpryDt == expiry)].set_index("TradDt").sort_index()
    lot = int(s.NewBrdLotQty.iloc[0])
    px = s.SttlmPric.loc[entry_date:exit_date]
    out = pd.DataFrame({"settle": px})
    out["m2m"] = side * px.diff().fillna(0) * lot * lots      # day 1: entry at that day's settle
    out["cum"] = out.m2m.cumsum()
    return out

L = ledger("NIFTY", "2026-09-29", +1, "2026-08-03", "2026-09-29")
print(L.tail(3)); print("Total:", round(L.m2m.sum(), 2))       # -132847.0
S = ledger("INFY", "2026-09-29", -1, "2026-08-03", "2026-09-24")
print("INFY short total:", round(S.m2m.sum(), 2))               # 65600.0

17.3 OI buildup for every stock over a window

import pandas as pd
f = pd.read_csv("data/fno/futures_aug_sep_2026.csv", parse_dates=["TradDt", "XpryDt"])
oi = f.groupby(["TckrSymb", "TradDt"]).OpnIntrst.sum().unstack()       # all expiries summed
spot = f.groupby(["TckrSymb", "TradDt"]).UndrlygPric.first().unstack()
a, b = "2026-08-31", "2026-09-28"
pc = spot[b] / spot[a] - 1
oc = oi[b] / oi[a] - 1
def buildup(p, o):
    return {(True, True): "Long buildup", (False, True): "Short buildup",
            (True, False): "Short covering", (False, False): "Long unwinding"}[(p > 0, o > 0)]
res = pd.DataFrame({"price_%": pc * 100, "oi_%": oc * 100}).dropna()
res["class"] = [buildup(p, o) for p, o in zip(res["price_%"], res["oi_%"])]
print(res.loc[["NIFTY", "TCS", "ITC", "DRREDDY", "HAVELLS"]].round(1))
indices = ["NIFTY", "BANKNIFTY", "FINNIFTY", "MIDCPNIFTY", "NIFTYNXT50", "NIFTYFPI"]
print(res.drop(index=indices, errors="ignore")["class"].value_counts())   # stocks only: 118/68/13/11

17.4 Rollover %, beta and hedge lots

import pandas as pd, numpy as np
f = pd.read_csv("data/fno/futures_aug_sep_2026.csv", parse_dates=["TradDt", "XpryDt"])
g = f[(f.TckrSymb == "NIFTY") & (f.TradDt == "2026-09-29")].set_index("XpryDt").OpnIntrst
print("Rollover %:", round(g.drop(pd.Timestamp("2026-09-29")).sum() / g.sum() * 100, 1))   # 74.3

px = lambda s: pd.read_csv(f"data/ohlc/{s}.csv", index_col=0, parse_dates=True)
n = px("IDX_NSEI").Close.pct_change()
r = px("LT")["Adj Close"].pct_change()
j = pd.concat([r, n], axis=1, keys=["s", "m"], sort=True).loc["2025-09-01":"2026-08-31"].dropna()
beta = np.polyfit(j.m, j.s, 1)[0]
print("LT beta:", round(beta, 2))                                 # 1.43
value, fut, lot = 2000325, 24251.4, 65
print("Lots for beta 0.931:", round(0.931 * value / (fut * lot), 2))   # 1.18

18. Practical checklist

• I know the contract's lot size, contract value, expiry date (last Tuesday, or the previous trading day if it's a holiday) and settlement type (index = cash, stock = physical).
• I checked my broker's current margin (SPAN + exposure) and my account holds at least 3–5× that margin.
• I wrote down entry, stop and target before ordering, and lots = risk budget / (stop distance × lot size) is at least 1.
• I checked liquidity: near-month contract, tight bid-ask, enough depth; I use limit orders outside the most liquid contracts.
• I checked the basis against fair value, and the corporate-action calendar (dividends) for stock futures.
• I checked for known events (results, RBI policy, budget, elections) before expiry, and the F&O ban list.
• I know how I will fund margin calls, and I will exit at my stop rather than add margin.
• I have a calendar reminder to roll or close 2–3 days before expiry (stock futures: always close or roll).
• I record every day's M2M and every trade's R-multiple (the trade's profit or loss divided by the amount I planned to risk; Chapter 11) in a journal.
• For a hedge: I computed portfolio beta, rounded lots consciously (under- vs over-hedge), and set a date to lift the hedge.

19. Key takeaways

20. Quiz and exercises

21. Sources

Zerodha Varsity, Module 4, Futures Trading (concepts; explained here in our own words): 1. Background – Forwards Market · 2. Introducing Futures Contract · 3. The Futures Trade · 4. Leverage & Payoff · 5. Margin & M2M · 6. Margin Calculator (Part 1) · 7. Margin Calculator (Part 2) · 8. All about Shorting · 9. The Nifty Futures · 10. Futures Pricing · 11. Hedging with Futures · 12. Open Interest · 13. Quick note on Physical Settlement.

Rules and rates (checked Oct 2026): SEBI, Study: Profitability of Individual Traders in the Equity Derivatives Segment FY25–FY26 (20 Aug 2026) · NSE Clearing, Equity derivatives margins (SPAN, ELM 2%/3.5%, calendar-spread ELM) · NSE Clearing SPAN risk-parameter file, 30 Sep 2026 · RBI homepage (91-day T-bill 5.5199%, repo 5.25%, as of 2 Oct 2026) · RBI Weekly Statistical Supplement, interest rates (25 Sep 2026) · NSE expiry moved to Tuesday from 1 Sep 2025 · NSE lot-size revision (NIFTY 65, BANKNIFTY 30) from the Jan-2026 series · NSE tick-size revision for derivatives (Apr 2025) · NSE trading holidays 2026 · STT rates from 1 Apr 2026 (futures 0.05%) · SEBI's Oct-2024 index-derivative measures (Zerodha Z-Connect) · History of F&O margin requirements / peak margin (Zerodha Z-Connect) · Zerodha physical-settlement policy · Removal of calendar-spread benefit for stock derivatives on expiry day (5 May 2026) · Why Zerodha stopped bracket orders · SEBI 2025 MWPL / future-equivalent OI changes.

Data: NSE F&O UDiFF bhavcopy, futures, 3 Aug – 30 Sep 2026 (data/fno/futures_aug_sep_2026.csv); SPAN values extracted to data/extra/ch09_span_20260930.csv; daily OHLC from NSE via Yahoo Finance (data/ohlc/). All calculations and figures: course/scripts/ch09_figures.py. Examples are historical analysis for education, not recommendations.