Almost every guide to the risk-reward ratio ends with the same rule: never take a trade unless you can make at least 2 or 3 times what you risk. That rule is fine for a breakout trader buying a stock. Applied to a short option, it is close to useless. An option seller's reward is capped at the premium collected on day one, while the risk is set by the stop-loss. So the ratio on almost every short premium trade starts below 1:1, by design.
That does not make selling a bad business, and it does not make the ratio meaningless. It means the ratio has to be read together with win rate and expectancy, because for a seller those three are really one number. This guide walks through how risk-reward works on short options, the break-even win rate formula with simple numbers, how your stop rule quietly sets your real ratio, and the cases where a low ratio really is a bad trade. All rupee and point figures below are illustrative, chosen to make the arithmetic easy to follow. They are not results, recommendations or levels to trade.
What risk-reward ratio means, and the rule most traders quote
The risk-reward ratio compares how much you stand to lose on a trade with how much you stand to gain. If you buy a stock at ₹500 with a stop at ₹480 and a target at ₹560, you are risking ₹20 to make ₹60. Written as risk:reward, that is 1:3. Some platforms flip it and quote reward:risk, so the same trade appears as 3:1. Check which way round a source writes it before comparing numbers. In this article we always write risk first, so 1:3 means "risk 1 to make 3" and 1:0.5 means "risk 1 to make half of that".
According to the Investopedia entry on the risk/reward ratio, many traders look for 1:3 or better, and most of the pages ranking for this keyword repeat some version of "never go below 1:2". The logic behind it is sound for directional trades. At 1:3 you only need to be right on 1 trade in 4 to break even, so you can be wrong a lot and still survive.
The hidden assumption is that the reward is open. A stock can run from ₹500 to ₹560 or to ₹700, and the trader chooses where to set the target. A long option works the same way: buy a call at 80 points and it can go to 240 if the move comes. That is the buyer's game, and it is the reason the 1:3 rule feels natural to anyone who started out buying options. We cover that side in more depth in our comparison of option buying vs option selling, because the two styles need different rules from the very first trade.
Why a short option starts with a "bad" ratio
Sell an option and the most you can ever make is the premium you received. Sell a NIFTY put for 60 points and your best case is that it expires worthless, so you keep 60 points. Nothing you do can make that 120. The reward side of the ratio is fixed the moment the order fills.
The risk side is set by where you exit if the trade goes wrong. A common rule is a stop-loss at a multiple of the premium. Put the stop at 150 on that 60-point put, and the loss if it triggers is 150 − 60 = 90 points. So the trade is:
- Reward: 60 points (the whole premium, best case)
- Risk: 90 points (stop at 150, before slippage)
- Risk-reward: 90:60, or 1:0.67
By the "never below 1:2" rule, that trade should never be placed. Yet this shape, small capped gain against a larger defined loss, is what option selling is. The seller gets paid for something else: the high odds that an out-of-the-money option loses value as time passes. The Options Industry Council's page on the short strangle describes exactly this profile, limited profit and large potential loss, and the US SEC's investor site lists options among products where you can lose more than you put in. Its options glossary entry on investor.gov is a plain summary worth reading if the mechanics are new.
So the honest question for a seller is not "is my ratio 1:3?" It is "is my ratio good enough for the win rate I actually get?"
Win rate and risk-reward are one number, not two
Here is the idea most risk-reward articles skip. On its own, a ratio tells you nothing about whether a strategy makes money. A 1:5 trade that wins 10% of the time loses money. A 1:0.5 trade that wins 80% of the time makes money. What matters is the two numbers together, and that combination has a name: expectancy.
Expectancy is the average result per trade:
Expectancy = (Win rate × Average win) − (Loss rate × Average loss)
Take the 60-point put from above, with an average win of 60 and an average loss of 90. If it wins 70% of the time:
0.70 × 60 − 0.30 × 90 = 42 − 27 = +15 points per trade
Now drop the win rate to 58%:
0.58 × 60 − 0.42 × 90 = 34.8 − 37.8 = −3.0 points per trade
Same ratio, same stop, same premium. A 12-point change in win rate turned a positive system into a losing one. That is why sellers obsess over win rate, and why a buyer's rulebook misleads them. For a seller the ratio is a cost you accept; the win rate is what pays for it.
The reverse is also true, and this is where new sellers get hurt. A 90% win rate sounds superb. But if the average loss is 12 times the average win, 90% is not enough: 0.90 × 10 − 0.10 × 120 = 9 − 12 = −3 points per trade. In our experience, the strategies that blow up retail accounts usually have a lovely win rate and a terrible ratio, and the trader only looked at the first number.
Break-even win rate: the formula and a worked example
You can collapse the two numbers into one test. Set expectancy to zero and solve for win rate. The result is the break-even win rate, the minimum hit rate a given ratio needs just to not lose money:
Break-even win rate = Average loss ÷ (Average loss + Average win)
For the put: 90 ÷ (90 + 60) = 90 ÷ 150 = 60%. Win more than 60% of trades with those numbers and you make money before costs. Win less and you don't.
The table below shows how the required win rate climbs as the ratio gets worse. It is pure arithmetic, so you can check every row with the formula above.
| Risk:reward | Example (risk / reward, points) | Break-even win rate |
|---|---|---|
| 1:3 | 30 / 90 | 25.0% |
| 1:2 | 40 / 80 | 33.3% |
| 1:1 | 60 / 60 | 50.0% |
| 1:0.67 | 90 / 60 | 60.0% |
| 1:0.5 | 120 / 60 | 66.7% |
| 1:0.33 | 180 / 60 | 75.0% |
| 1:0.25 | 240 / 60 | 80.0% |
| 1:0.1 | 600 / 60 | 90.9% |
Notice how gently the first few rows move and how steeply the last few do. Going from 1:1 to 1:0.5 needs about 17 extra points of win rate. Going from 1:0.25 to 1:0.1 needs another 11, at a level where each extra point of win rate is very hard to earn. That curve is the whole story of option selling in one table.
Add costs before you trust the number
The break-even rate above is before costs, and costs fall harder on sellers than the textbook suggests. Every trade pays brokerage, exchange charges, STT and GST, plus slippage on entry and exit. Rates change, so check your broker's charge sheet and the exchange circular in force instead of using a figure from an old blog post.
Suppose all of that adds up to 3 points per round trip on our put (an illustrative figure). The average win falls from 60 to 57 and the average loss grows from 90 to 93. Break-even becomes 93 ÷ (93 + 57) = 93 ÷ 150 = 62%. Two points of win rate gone to costs alone. On a thinner premium the damage is far worse: sell an option for 8 points with the same 3-point cost and you give up over a third of your best case before the market moves at all.
The short strangle case
A short strangle doesn't have a single ratio, because it has three common outcomes, not two. Say you sell a call and a put for 50 points each, 100 in total, with a stop on each leg at 100 (the premium doubles). Illustrative outcomes:
- Range day. Neither stop hits. Both legs decay, and you buy them back for 10 points combined near the close. Gain: 90 points.
- Trend day. The market rallies. The call hits its stop, so that leg loses 50. The put decays to 10, a gain of 40. Net: −10 points.
- Whipsaw day. A rally stops out the call, then a sharp reversal stops out the put. Both legs lose 50. Net: −100 points.
So the "risk-reward" of this strangle is 90 at best, −100 at worst, with a very common middle case near −10. Quoting it as 1:0.9 hides the fact that the trend-day loss is small and the whipsaw loss is the real danger. When we evaluate a strangle on our own backtest engine, we look at how often each of the three states happens, not just the overall win rate, because two strangles with the same win rate can have very different whipsaw counts. If you want the mechanics of this structure in full, our guide to the short strangle explained covers strikes, margin and adjustments.
How the stop-loss rule sets your real ratio
For a seller, choosing a stop rule is choosing a risk-reward ratio. The premium is fixed, so moving the stop is the only lever. With a stop set as a multiple of premium, the best-case ratio follows directly:
| Stop (per leg) | Max loss as multiple of premium | Best-case risk:reward | Break-even win rate |
|---|---|---|---|
| 30% stop | 0.3× | 1:3.33 | 23.1% |
| 50% stop | 0.5× | 1:2 | 33.3% |
| 100% stop (premium doubles) | 1.0× | 1:1 | 50.0% |
| 150% stop | 1.5× | 1:0.67 | 60.0% |
| 200% stop (premium triples) | 2.0× | 1:0.5 | 66.7% |
| No stop | Open | Undefined | Unknown until the bad day arrives |
Look at the top row. A 30% stop gives a seller a "textbook" 1:3 ratio. It also gets hit by ordinary noise all day long, because option premium routinely swings 30% in an hour without the trade being wrong. So the win rate falls far faster than the ratio improves. Tighter stops don't make a seller safer by default; they move the problem from the ratio column to the win-rate column. In our experience, a losing seller system is almost never fixed by tightening its stop. It is usually fixed, if at all, by better entry timing, better strike distance or skipping the wrong days.
Your realised ratio is worse than the planned one
The table assumes you keep the whole premium on winners and exit exactly at the stop on losers. Neither happens in real trading, and both errors push the same way.
On winners, most intraday sellers exit before expiry. If you square off at 15:15 IST with 10 points of premium left on the 60-point put, your average win is 50, not 60. On losers, stops slip. A stop at 150 can fill at 165 when the market spikes on news, because there is no buyer at 150 in that second. Average loss: 105, not 90.
Run the formula again with those realistic figures: 105 ÷ (105 + 50) = 105 ÷ 155 = 67.7%. The planned break-even was 60%. The realised one is nearly 8 points higher. That gap, between the ratio you planned and the ratio the fills gave you, is where a lot of "it worked in my spreadsheet" systems die.
This is why our team tracks realised average win and realised average loss from the actual fills, not the planned ones, and recomputes the break-even rate monthly. If the realised break-even creeps toward the realised win rate, the edge is shrinking even if nothing on the screen looks different.
Trailing stops change the shape, not just the size
A trailing stop on a short option moves the stop down as the premium decays. It cuts the average loss on trades that first went your way and then turned. It also cuts some winners short, so the average win falls a little too. Whether the trade-off helps depends on how your strategy's trades usually move, which is a question for a backtest, not a rule of thumb. We go through the common trailing methods in our article on trailing stop-loss for short options.
Ratio per trade versus ratio per account
One more distinction matters. The ratio of a single trade says nothing about how much of your capital that trade puts at stake. A 1:0.5 trade sized at 1% of capital is far safer than a 1:2 trade sized at 15%. Position size decides whether a run of losses is an annoying week or the end of the account. For a seller, a reasonable habit is to size so that the realistic worst case (a gapped stop, not the planned stop) costs a small, fixed slice of capital. Our piece on position sizing vs technique in option trading makes the case that sizing matters more than the setup itself for most traders.
When a low ratio is still a bad trade
None of this means any low ratio is acceptable. A low ratio is a fair price for a high win rate only when that win rate is real, repeatable and survives costs. Here is where it breaks down.
Selling far out-of-the-money options for tiny premium. Selling an option for 3 points that can gap to 200 on a surprise is a risk-reward of roughly 1:0.015. The break-even win rate is about 98.5% before costs, and costs alone might eat the whole premium. You will win most days, and one bad open will take months of gains. The pattern of "sell the cheapest strikes because they always expire worthless" is exactly how this happens.
No stop, or a stop the market can jump. If the loss side has no defined exit, the ratio is not 1:0.5 or 1:0.2. It is undefined, and the break-even win rate is unknowable until the bad day arrives. Overnight and weekend gaps make this worse for positional sellers, since the stop does not exist while the market is closed. Defined-risk structures such as spreads or iron condors cap the loss with a bought option, which gives a ratio you can actually calculate.
Event days. Union Budget day, RBI policy announcements and election result days all produce moves the normal premium does not fully price. A seller's historical win rate is mostly built on quiet days, so trading the same rules into a known event uses a win rate that does not apply to that day.
Rolling and averaging losers. Rolling a stopped-out short to a further strike, or adding more lots to "average" a losing short, turns a defined loss into an open one. The ratio on paper stays the same; the real ratio gets much worse.
A win rate you can't verify. If the 80% figure comes from a few weeks of live trading or a backtest with no costs or slippage, it is not a basis for accepting a 1:0.25 ratio. You need enough trades across different market conditions, with realistic fills, before the win rate means anything.
Paper trade this first. Any rule in this article that names a stop multiple, an exit time or a premium level is an illustration of the arithmetic, not a level to trade. Test it on paper or in a backtest with realistic costs for at least 2 to 3 months of sessions before risking capital.
How to use this on your own trades
A short routine that turns the theory into a habit:
- For each strategy, record every trade's realised win or loss in points, after costs.
- Every month, compute the average win, average loss and win rate from those records.
- Compute the realised break-even win rate: average loss ÷ (average loss + average win).
- Compare it with the actual win rate. The gap between the two is your margin of safety.
- If the gap shrinks for 2 or 3 months in a row, look at fills, strike distance and event days before touching the stop.
That fifth step reflects a simple observation. When we see the margin shrink on a short premium book, the cause is more often worse fills or a change in market behaviour than a wrong stop level. Changing the stop first only hides what is going on.
A note on methodology: every number in this article is illustrative and chosen to make the arithmetic easy to follow. The formulas are standard, and the same expectancy and break-even logic is laid out in Zerodha Varsity's module on option strategies and most options textbooks. EliteAlgo runs its own proprietary book with these checks built into its software. This is research and analytical software, not investment advice. If you would rather have the execution side handled on a tested system, see our managed algo trading service.
Frequently Asked Questions
What is a good risk-reward ratio for option selling?
There is no single good number. A seller's ratio is usually between 1:0.3 and 1:1, because the reward is capped at the premium. What matters is whether your realised win rate beats the break-even win rate for that ratio, after costs. At 1:0.5 you need more than 66.7% winners just to break even.
How do you calculate risk-reward on a short strangle?
Treat it as three outcomes, not one ratio. Work out the gain if neither leg is stopped, the result if one leg is stopped and the other decays, and the loss if both legs are stopped. Then weight each by how often it happens in your own records. A single ratio hides the whipsaw case, which is the real risk.
Is a 1:1 risk-reward ratio good?
For a seller, 1:1 is a fairly tight setup. It usually means a stop where the premium doubles. You need more than 50% winners before costs, and a little more after them. Whether that is good depends on how often your entries actually win at that stop.
What win rate do I need at a 1:0.5 risk-reward?
Before costs, 66.7%. The formula is average loss ÷ (average loss + average win), so 1 ÷ (1 + 0.5) = 0.667. Costs and stop slippage push the real figure higher, often by several points, so aim for a clear margin above it.
Does a stop-loss change the risk-reward ratio?
Yes. For a seller, the stop is the only thing that sets the ratio, because the reward is fixed at the premium. A wider stop lowers the ratio and usually raises the win rate. A tighter stop raises the ratio and usually lowers the win rate. Slippage on the stop makes the real ratio worse than the planned one.
Why do option sellers accept low risk-reward?
Because they are paid in probability, not payoff size. Out-of-the-money options lose value as time passes, so a seller wins more often than they lose. The low ratio is the price of that higher win rate. It only works when the win rate is real, holds up across market conditions and stays above the break-even rate after costs.