Home / Risk Management / Position Sizing

Position sizing

Position sizing decides how much of the account a single trade can lose. On a leveraged instrument it matters more than entry, exit or direction.

Quick answer: Position sizing is the rule that fixes how many lots to trade so that a losing trade costs a pre-chosen fraction of the account. It converts a risk budget in rupees into a whole number of lots, and it dominates outcomes on leveraged instruments.

In simple words

Position sizing answers one question before any trade: if this goes wrong, how much of my account disappears? You pick a small slice of capital you are willing to lose on the trade, say one or two per cent. You work out what one lot would lose if the market reaches your exit. You divide the first number by the second. That tells you how many lots you can hold. Sometimes the answer is zero, because a single lot already risks more than your slice. That is not a glitch. It is the arithmetic telling you the instrument is too large for the account, which is information worth having before the loss, not after.

The fixed-fractional rule

Fixed-fractional sizing risks a constant fraction of current equity on every trade. A common convention is one to two per cent of equity per trade; the number is a choice, and the arithmetic below shows what each choice implies. The rule has one appealing property: because the fraction is of current equity, the rupee amount shrinks automatically after losses and grows after gains, so a losing run cannot compound into ruin as fast as a fixed rupee bet would. Ralph Vince formalised the family of these methods as optimal-f, and the honest summary is that the aggressive end of that family is far too aggressive to survive real uncertainty. The convention exists precisely because it is survivable, not because it maximises anything.

Turning a budget into lots

The mechanics are one division. Multiply the account by the risk fraction to get the rupee budget. Divide that by the loss a single lot would take if the market travels to your planned exit, which is the stop distance in points multiplied by the lot size. Take the floor, because you cannot trade a fraction of a lot. The floor matters more than it looks. It means the smallest tradeable unit is a hard quantum: below a certain account size, the only whole numbers available are zero and one, and one may already break the budget. On index futures, where one lot is a large notional, that quantum is coarse enough to exclude most small accounts from sizing at one per cent at all.

Why the leverage makes sizing everything

One NIFTY futures lot at 24,000 is 24,000 multiplied by 75, a notional exposure of eighteen lakh rupees, controlled with a margin that is a fraction of that. That gap between notional and margin is the leverage, and it is why sizing dominates every other decision. A move of one per cent in the index is eighteen thousand rupees per lot, whether or not the trader has eighteen lakh. Entry technique, indicator choice and market view all operate on top of this multiplier; none of them can rescue a size that is wrong. The trader who gets sizing right and the view wrong loses slowly. The trader who gets the view right and the size wrong can still be removed by a single adverse move before the view pays off.

Where defined-risk structures change the sum

For a futures lot the loss per lot is an estimate, because the market can gap through the intended exit and the realised loss can exceed the planned one. For a defined-risk option structure the loss per lot is a known constant, fixed by the structure before entry. That single difference changes the sizing sum from a guess into a division by a certainty. An iron condor built from the illustrative NIFTY chain, selling the 24,600 call and 23,400 put and buying the 24,800 call and 23,200 put, collects about 89 points of credit against a 200-point wing, so its worst case is 111 points, or 111 multiplied by 75, which is eight thousand three hundred and twenty-five rupees per lot, known in advance. A small account can size against that number honestly.

Kelly, briefly and honestly

The Kelly criterion, associated with Ed Thorp's use of it, gives the growth-optimal fraction to bet when the edge and the odds are known exactly. Two facts make it a poor literal guide for a discretionary trader. First, full Kelly is violently volatile; it accepts drawdowns most people cannot psychologically or financially hold, which is why practitioners who use it at all use a small fraction of it. Second, and more fatally, it needs the edge and the odds as inputs, and a discretionary trader does not know either. Feeding guessed inputs into a formula that is exquisitely sensitive to them produces a confident number that is wrong. Kelly is a useful ceiling to understand and a dangerous recipe to follow.

The formula

Lots = floor( (Account × Risk%) ÷ (Stop distance × Lot size) )

Account = current equity in rupees. Risk% = the fraction of equity you accept losing on this trade (e.g. 0.01). Stop distance = points from entry to your planned exit. Lot size = units per lot (NIFTY 75, BANKNIFTY 30 at the time of writing). floor() rounds down to a whole number of lots.

Worked example

Take a ₹5,00,000 account risking 1% per trade, so the budget is ₹5,000. On NIFTY with an 80-point stop, one lot risks 80 × 75 = ₹6,000. Then floor(5,000 ÷ 6,000) = 0 lots. The correct size is zero: the smallest tradeable unit already exceeds a 1% budget on ₹5 lakh, and forcing one lot means risking 1.2% instead of the 1% chosen. Now a ₹10,00,000 account at 1% has a ₹10,000 budget; floor(10,000 ÷ 6,000) = 1 lot, risking ₹6,000. The defined-risk route: a ₹5,00,000 account choosing a 2% budget (₹10,000) can hold one iron condor lot whose known worst case is 111 × 75 = ₹8,325, because floor(10,000 ÷ 8,325) = 1 — and 8,325 is a certainty, not a stop the market can gap through. Figures exclude brokerage, STT and other costs.

Common mistakes

  • Sizing by the margin the broker demands rather than by the loss the position can take means the trader loads up to the margin limit and discovers that a full account of margin can still lose far more than the account can afford.
  • Forcing at least one lot when the sizing formula returns zero silently raises the real risk above the chosen fraction, so the discipline of a risk budget is abandoned on exactly the trades where the account is too small to take it.
  • Using a fixed rupee risk instead of a fraction of current equity lets a losing streak compound, because the same rupee bet is a larger and larger share of a shrinking account, which accelerates towards ruin rather than damping it.
  • Treating a futures stop-loss as a certain exit at the stop price ignores gaps and illiquidity, so the realised loss per lot can exceed the stop distance used in the sizing sum and the position is quietly larger than intended.
  • Copying a Kelly or optimal-f fraction from a book without the exact edge and odds it requires produces a confident bet size built on guessed inputs, and the formula's sensitivity turns a small input error into a ruinous position.
  • Sizing off notional value or lot count instead of worst-case rupee loss makes an undefined-risk short option look the same size as a defined-risk spread, when the tail losses are not remotely comparable.

Professional usage

Desks and funds size by risk, not by capital or margin. Positions are expressed in a common currency of risk — value-at-risk, stress loss, or rupees-at-a-defined-move — and a risk book aggregates them across strategies so that correlated bets are not counted as independent. Limits are set per strategy and per book, monitored intraday, and enforced by a risk function that can cut a trader who breaches them, independent of the trader's conviction. Capital is allocated to the risk a strategy consumes, and leverage is a managed input rather than a broker's default. A retail trader cannot replicate the cross-margining, the real-time aggregation or the independent enforcement, but the underlying idea — one rupee risk budget, allocated deliberately — transfers exactly and costs nothing to adopt.

Key takeaways

  • Position sizing converts a chosen rupee risk budget into a whole number of lots by dividing the budget by the worst-case loss per lot, then taking the floor.
  • On leveraged instruments, sizing dominates entry, exit and direction, because one NIFTY lot at 24,000 carries ₹18,00,000 of notional exposure.
  • A sizing answer of zero is a valid, informative result: the instrument is too large for the account at the chosen risk fraction.
  • Defined-risk option structures have a known loss per lot, so they let a small account size honestly where a gap-prone futures stop cannot.
  • Full Kelly is too aggressive under parameter uncertainty, and its required inputs are unknowable for a discretionary trader; a small fraction of a risk convention is the survivable choice.

Frequently asked questions

What is position sizing?
Position sizing is the rule that decides how many lots to trade so a losing trade costs a pre-chosen fraction of the account. It turns a rupee risk budget into a whole number of lots by dividing the budget by the worst-case loss per lot.
How do I calculate lots to trade?
Multiply your account by the risk fraction to get a rupee budget. Divide that by one lot's loss at your planned exit — stop distance in points times lot size. Take the floor. For example floor((5,00,000 × 0.01) ÷ (80 × 75)) = floor(5,000 ÷ 6,000) = 0 lots.
Why does my sizing formula say zero lots?
A result of zero means one lot already risks more than your chosen fraction of the account. It is arithmetic, not error: at 1% of ₹5 lakh (₹5,000) versus ₹6,000 per NIFTY lot, the smallest tradeable unit exceeds the budget. The instrument is too large for the account at that risk level.
What risk percentage per trade should I use?
The figure is a choice, not a rule. A common convention is one to two per cent of current equity per trade. Lower fractions survive longer losing runs; higher fractions reach ruin faster. The arithmetic of your account and lot size decides which fractions even produce a tradeable lot count.
How much is one NIFTY futures lot worth?
At an index level of 24,000 and a lot size of 75, one NIFTY futures lot carries a notional exposure of 24,000 × 75 = ₹18,00,000. The margin to hold it is a fraction of that; the gap between the two is the leverage, and it is why sizing dominates.
What is fixed-fractional position sizing?
Fixed-fractional sizing risks a constant fraction of current equity on every trade. Because the fraction is of current equity, the rupee bet shrinks after losses and grows after gains, which damps a losing streak instead of letting a fixed rupee bet compound toward ruin.
Should I use the Kelly criterion for trading?
Kelly gives the growth-optimal bet fraction when edge and odds are known exactly. A discretionary trader knows neither, and the formula is very sensitive to both, so guessed inputs give a confident but wrong number. Full Kelly is also extremely volatile; practitioners who use it use a small fraction.
Does position sizing matter more than entry timing?
On leveraged instruments, yes. Sizing sets how much a wrong trade costs; entry technique operates on top of that multiplier. A correct view with the wrong size can be removed by one adverse move before it pays; a wrong view with correct size loses slowly.
How does a small account size a NIFTY position?
Often it cannot size a futures lot at 1%, because one lot's ₹6,000 risk (80 points × 75) exceeds the budget. A defined-risk option structure with a known worst case — for example an iron condor risking ₹8,325 per lot — lets a small account size against a certainty.
Why is a futures stop-loss not a certain exit?
A stop is an instruction to exit near a price, not a promise of that price. Gaps and thin liquidity can fill the exit far away, so the realised loss per lot can exceed the stop distance used in your sizing sum, leaving the position larger than intended.
What is the difference between margin and risk?
Margin is the collateral the broker requires to hold a position; risk is how much the position can lose. They are not proportional. A fully margined short option can still lose several times its margin, so sizing by margin understates the risk being taken.
How does defined risk change position sizing?
With defined risk the loss per lot is a constant fixed by the structure before entry, not an estimate a gap can blow through. That turns the sizing division into a division by a certainty, which is why a small account can size a spread honestly where it cannot size a naked futures lot.
Can I lose more than the margin I posted?
On undefined-risk positions, yes — a short option or futures lot can lose well beyond its margin, and the broker may square you off on a shortfall. On defined-risk structures the maximum loss is capped by the structure, so it is known before entry.
Does position sizing account for costs?
The basic sizing sum uses gross loss per lot. Brokerage, STT, exchange charges, stamp duty and GST are a separate, real drag, material on multi-leg structures and frequent trading. Prudent sizing treats costs as reducing both the budget and the net outcome of every trade.

Voice search & related questions

What is position sizing in trading?
Position sizing is deciding how many lots to trade so a loss costs only a small, pre-chosen slice of your account. You divide your rupee risk budget by what one lot loses at your exit, and round down to a whole number of lots.
How many NIFTY lots should I trade with five lakh rupees?
At 1% risk (₹5,000) with an 80-point stop, one lot risks ₹6,000, so the formula gives zero lots — one lot already exceeds the budget. That result is telling you a ₹5 lakh account is small for NIFTY futures at 1% risk.
Is one per cent risk per trade a good rule?
One to two per cent is a common convention, not a rule. It is a choice about how long you can survive a losing run. The arithmetic of your account size and lot size decides whether that fraction even produces a tradeable number of lots.
Why does sizing matter more than picking the right trade?
Because leverage multiplies every outcome. One NIFTY lot controls ₹18,00,000 of exposure. A right call in the wrong size can be stopped out by a single move before it works, while a wrong call in the right size only bleeds slowly.
How do small accounts trade NIFTY safely?
Nothing removes risk, but a small account can size against a known number by using a defined-risk structure whose worst case is fixed before entry — an iron condor risking about ₹8,325 per lot — instead of a futures lot whose stop a gap can overshoot.

Last reviewed 9 July 2026. Educational content only — not investment advice.

← Previous
Exit Planning
Educational content only — not investment advice. See our Risk Disclosure.