Take an account of Rs 5,00,000. It is an assumption, not a trader. On 10 Jun 2026 one lot of the 23,250 Nifty call cost Rs 20,907. If it expires worthless, 4.2% of the account goes in one trade. With an exit at half the premium, one lot still risks Rs 10,454. A run of losses needs no mistake, only chance: five losses in a row on trades that are coin tosses come about once in 32 tries. Size is the one thing a trader sets before the market speaks, and it is the first number I look for in any plan, ahead of the win rate. This lesson shows how size, win rate and payoff combine into the chance of ruin, and why margin is the wrong place to start.
3By the end you can
- Work out the edge per trade from a win rate and a payoff ratio.
- Size one Nifty lot from a risk budget and an exit, and say when the account is too small for it.
- Explain why a larger size raises the chance of ruin on the same odds, and why a stop does not fix the fill.
The idea, plainly
Three numbers decide a trader's fate. The win rate is how often a trade makes money. The payoff ratio is the average win divided by the average loss. Size is the share of the account risked on one trade. The first two give the edge per trade, counted in R, the amount risked on that trade.
Edge in R =
win rate x payoff ratio
less loss rate
Positive edge means the average trade makes money. Zero or negative means it does not, however often it wins. Module F13 shows a win rate of nine in ten sitting beside zero expectancy.
Size is the knob
In the model below the odds never change: the trader wins a little over half the trades at an even payoff, a small positive edge. Only the size moves. At two percent of the account a trade, the chance of losing half the account within the model's run of trades is 0.05%. At five percent it is 8.6%, at ten percent 41% and at twenty percent 81%. Same edge, four different fates.
Model, not market data
Same odds, bigger size: the chance of ruin climbs from nothing to four in five
A small size cannot rescue a negative edge
Now take a win rate a little under half at the same even payoff. Size cannot save this one. At one percent a trade, the median account at the end of the run is 0.78 of its start. Ruin is rare only because the slide is slow.
Model, not market data
A small size slows a negative edge. It does not turn it around
Kelly: the size that grows fastest if the odds were known
Kelly =
p less (1 less p) / b
Here p is the win rate and b the payoff ratio. For the first model it is 10% of the account a trade, and at that size the model's chance of halving the account within its run is 41%. Full Kelly assumes the odds are known exactly. They never are, so it describes a ceiling on size and not a plan. When the edge is zero or negative the formula says to risk nothing.
Stops, and where they leave you
A stop is a rule that fixes the loss per trade in advance, and size follows from it.
Lots =
risk budget
÷ loss per lot at the stop
On an option the stop is a price of the option, and that price moves with the index, with volatility and with time. This grid shows one lot of the bought 23,250 call one day after the example date.
A stop at half the premium can be hit by a one day fall of one or two percent
| Move in the future | Rupees per lot, one day later | |
|---|---|---|
| IV −3 | IV same | |
| +2% | +14,556 | +17,967 |
| +1% | +3,954 | +7,896 |
| 0% | −4,644 | −530 |
| −1% | −11,086 | −7,222 |
| −2% | −15,490 | −12,233 |
Green is profit and red is loss; deeper colour means a larger amount. The outlined cell is the starting point.
The deeper the loss, the steeper the climb back
Read one more paragraph, 65 words
A stop at half the premium, Rs 10,454, sits between the minus one and minus two percent rows. If volatility also falls three points, a one percent fall is already past it. A stop also fixes the loss only if the order fills near the stop. If the option gaps through it, the loss is larger than the budget it was meant to protect.
Worked example: sizing one Nifty call lot, 10 Jun 2026
The 23,250 call settled at 321.65 a unit on Wednesday 10 Jun 2026, and a lot is 65 units. The account and the two percent budget are assumptions, chosen to show the arithmetic and not as a view on how much to risk.
| Step | What we work out | Value |
|---|---|---|
| 1 | Account, an assumption | Rs 5,00,000 |
| 2 | One lot: 321.65 times 65 | Rs 20,907 |
| 3 | Loss if it expires worthless, share of step 1 | 4.2% |
| 4 | Loss per lot if the exit is at half the premium | Rs 10,454 |
| 5 | Risk budget, two percent of step 1 | Rs 10,000 |
| 6 | Lots the budget allows: step 5 divided by step 4 | 0.96 |
| 7 | Account at which one lot risks two percent at that exit: step 4 divided by two percent | Rs 5.23 lakh |
| 8 | The same account for a budget of one percent | Rs 10.45 lakh |
Read 2 more paragraphs, 188 words
Step 6 is below one, so this account cannot take even one lot of this call at that exit within the budget. That is not a verdict on the trade. It is what the arithmetic says about a lot that cannot be cut: 65 units is the smallest block. Module F04 has the lot sizes. A seller has no such loss to divide by, because a sold call has no cap, and the exchange's margin in Module F05 is a one day estimate, not that number.
SEBI's Aug 2026 studies show what happens when the arithmetic is skipped. In the behaviour study's random sample of individual traders, those who only bought options had a median return of minus 114% on the peak margin they used in FY26, so at least half lost more than the most capital they had at risk at any one time. Across all individual traders, the profitability study found the average loss of a loss maker in FY26 was 21% larger than the average profit of a profit maker. In this lesson's terms: a payoff ratio under one, and far fewer winners than losers.
The common mistake
The belief: If the odds are good, size does not matter.
The first model has a positive edge on every trade and still ruins four paths in five at twenty percent a trade. Size is the only one of the three numbers a trader sets alone. Win rate and payoff are estimates, and estimates from a few dozen trades carry a wide error. Size is the allowance for that error. Judge a plan by what it does when the odds are a little worse than believed, not by what it promises when they are exactly as hoped.
Key takeaways
- Edge per trade is win rate times payoff ratio, less the loss rate.
- On the same odds, a larger size raises the chance of ruin sharply.
- A small size slows a negative edge. It does not turn it around.
- Lots equal the risk budget divided by the loss per lot at the exit.
- A stop fixes the loss only if the order fills near it.
- After a fifty percent loss, a one hundred percent gain is needed.
The term this section turns on: Payoff ratio
Before you trade. Derivatives trading carries a high risk of loss. SEBI's study "Profitability of Individual Traders in the Equity Derivatives Segment (FY25 to FY26)" (August 2026) found that 87.7% of individual traders in equity F&O made net losses in FY26, with aggregate net losses of Rs 91,685 crore. Read the SEBI study