Position Sizing: From Equity to Whole Lots

Short answer: the number of lots decides how much money one price move can take from you. Picture two learners in the same NIFTY futures contract, with the same entry and the same exit. One holds one lot and the other holds three. Before costs, the second learner wins or loses three times as much. Direction is the same; size is the only thing that changed.

Start from your account equity: what the account is worth today, with open positions marked at their current value. Money you need for rent, bills or other goals stays outside this number. Then separate three ideas that beginners often mix up. A loss allowance (the Start course calls it a loss budget) is the amount you decide a trade may cost you. Collateral is the money or approved assets the broker holds against your obligations. Buying power is what the broker lets you take on. None of these is a cap on your loss. Margin, leverage and pledging explains the second and third.

Our account has ₹5,00,000 of equity and a 1% loss allowance, which is ₹5,000. Both are teaching numbers, not advice on what percentage suits you. We assume a NIFTY lot of 65 units. Check the real lot size for your contract and date before you use any calculation here. NIFTY is an index, so nothing is delivered into your demat account. The first lesson on this subject, risk management 101, builds the same idea for shares.

Short answer: divide the allowance by the loss on one lot, then round down. Allowance = equity × fraction. Lots = allowance ÷ loss per lot, rounded down to a whole number. If the answer is below one, you take zero lots. Rounding up quietly raises your allowance; it does not solve anything.

Take a long NIFTY futures lot bought at 25,000 with a planned exit at 24,850. The adverse move is 150 points. At 65 units that is 150 × 65 = ₹9,750 before costs. Your allowance is ₹5,000, and ₹5,000 ÷ ₹9,750 is about 0.513 of a lot. Rounded down, that is zero lots.

Short answer: the lot count must pass two tests. The loss test asks how much you could lose. The funding test asks how much money the position needs. The smaller answer wins.

Use a second, separate example: ₹10,00,000 of equity and a 2% allowance, which is ₹20,000. Suppose a bought-and-sold Call spread (a credit vertical) has a worst terminal loss of ₹4,550 per lot before costs. The loss test allows four lots, because 20,000 ÷ 4,550 is 4.396, rounded down to 4. Now suppose only ₹1,80,000 is free for this trade and each lot needs ₹1,80,000. The funding test allows one lot. One is the binding answer, and the planned terminal loss is ₹4,550.

Short answer: for this account and this trade, the position is too big. Zero is a valid answer, not an error. In our example, one lot would put ₹9,750 at risk, which is 1.95% of a ₹5,00,000 account, against a 1% allowance.

You can change the trade. A different contract, a smaller structure or a different exit makes a different calculation. What you should not do is move your exit closer just to make the arithmetic return one lot. That changes the plan to fit the answer, and a tighter exit is more likely to be hit by ordinary movement.

A smaller structure such as a spread can fit the allowance, but check it as carefully as the first idea. A quoted credit is not a filled credit. A partly filled order can leave you with one leg only. The protective leg must match in underlying, expiry and quantity. If the limits still say zero, keep zero.

Short answer: a percentage allowance gets smaller after every loss, and a fixed rupee allowance does not. To see this cleanly, allow any position size and assume every trade loses its full allowance. Start with ₹5,00,000. If each loss is 1% of the equity left, twenty losses leave about ₹4,08,953.47. If each loss is a fixed ₹5,000, twenty losses leave ₹4,00,000.

Short answer: the gain is measured on a smaller account. A drawdown is the fall from the highest value your account reached to a later, lower value. Start with ₹5 lakh and lose 20%, and you hold ₹4 lakh. To get the ₹1 lakh back, you must gain 25% on ₹4 lakh. The two percentages use different bases.

Short answer: each position has its own loss measure, and you must use the right one. A bought NIFTY Call at ₹100 per unit, paid for in full, costs 65 × ₹100 = ₹6,500. That is the most it can lose before costs, so on a full-premium basis it fits zero lots against a ₹5,000 allowance. If you plan to sell at ₹40, the estimated loss is ₹3,900, but that is a plan, not the ₹6,500 limit, and nobody guarantees a fill at ₹40. Index points and option premium rupees are different things, so a stop on the index cannot be dropped into an option loss formula.

A matched Call credit spread has a 100-point strike width and a filled net credit of ₹30 per unit. Its worst loss at expiry, before costs, is (100 − 30) × 65 = ₹4,550. One lot fits the ₹5,000 allowance. A debit spread uses its net debit as its worst loss. These limits only hold if you keep both legs and the legs match. They do not guarantee an exit price or cover delivery funds for stock options.

Short answer: Kelly finds the bet size that grows money fastest in a simple repeated game whose odds you already know. It does not tell you how much to risk in real options trading. Our game has two outcomes. You lose a fraction f of your equity, or you win b times that fraction, and the chance of winning is p.

For p between 0 and 1 and f below 1, expected log growth is p × ln(1 + bf) + (1 − p) × ln(1 − f). The best f is p − (1 − p) ÷ b. If that comes out negative, the best bet in this game is zero. If you do not know p, you cannot run the calculation, and the right input is not 50%.

Short answer: add the allowances, then test one shared market move. Four positions that each carry a ₹5,000 allowance on a ₹5 lakh account carry ₹20,000 together, or 4%. That sum holds whether the positions are related or not. Relatedness matters for how they move together, not for the addition.