Risk-Reward and Win Rate
Why 80% accuracy can still lose money and 40% accuracy can make money. Expectancy explained with simple Indian-rupee examples, so you stop chasing being right and start chasing being profitable.
- ·Win rate is not edge
- ·Risk-reward ratio
- ·Expectancy in plain rupees
- ·Why high accuracy can lose
- ·Why low accuracy can win
- ·Designing a positive-expectancy trade
Arjun loved telling people his win rate. Out of his last ten trades, eight were green. He screenshotted them, posted them, felt like a natural. Yet when he opened his account at month-end, the balance was lower than where he started. How does someone win eight times out of ten and still lose money? The answer is the most important idea in trading, and almost nobody teaches it to beginners. Arjun was right most of the time, but being right was quietly making him poor.
This chapter is about the gap between being right and making money, and the one number that closes it: expectancy.
Being right is not the same as making money
A win rate is just how often you are right: eight greens out of ten trades is an 80% win rate. It feels like the whole game. It is not. What actually fills or empties your account is the size of the wins next to the size of the losses.
Picture Arjun's habit. When a trade goes his way, he grabs the small profit fast, scared it will vanish, so his wins are tiny. When a trade goes against him, he holds and hopes, so the loss grows large. Eight small wins, two big losses. The two losses are bigger than the eight wins put together. High accuracy, shrinking account. The scoreboard he is proud of, his win rate, is measuring the wrong thing.
Your win rate tells you how often you are right. It tells you nothing about how much you make. A trader who wins less than half the time can crush a trader who wins most of the time, if the winners are big and the losers are small.
The risk-reward ratio
Before any trade, you can know two numbers: how much you will lose if you are wrong (your risk, set by your stop-loss), and how much you aim to make if you are right (your reward, your target). The relationship between them is the risk-reward ratio.
If you risk Rs 1,000 to make Rs 3,000, that is a 1:3 ratio, said "one to three". You are risking one unit to make three. A 1:1 ratio risks Rs 1,000 to make Rs 1,000. A bad ratio risks Rs 2,500 to make Rs 500, which is roughly 5:1 against you, which is exactly the trap Arjun fell into without naming it.
Why does the ratio matter so much? Because it sets the win rate you need just to break even. The better your reward versus your risk, the more often you are allowed to be wrong.
That last bar is the whole story. By cutting winners to Rs 500 and letting losers run to Rs 2,500, Arjun built a trade that needed an 83% win rate just to break even. He won 80%. A hair below the line, and the account bled.
Expectancy: the one number that matters
There is a single number that blends win rate and risk-reward together and tells you what an average trade is worth in rupees. It is called expectancy, and it is the number you should obsess over instead of accuracy.
Expectancy = (win rate x average win) - (loss rate x average loss)
In plain words: of all your trades, multiply how often you win by how much you win, then subtract how often you lose times how much you lose. If the answer is positive, each trade is worth money on average and time is on your side. If it is negative, every trade is quietly costing you, and more trading just means losing faster.
Two traders, one lesson
Let us put Arjun's style next to its opposite, in rupees, so the point is impossible to miss. Assume each is willing to risk about Rs 1,000 on a trade.
Trader A is the proud high-accuracy trader. He wins 80% of the time, but he snatches profits early (average win only Rs 500) and lets losers run past his stop (average loss Rs 2,500).
Trader B is the patient low-accuracy trader. She wins only 40% of the time, but she risks Rs 1,000 to make Rs 3,000, a clean 1:3 ratio. She cuts losers at the stop and lets winners reach the target.
| Item | Trader A (high accuracy) | Trader B (high payoff) |
|---|---|---|
| Win rate | 80% | 40% |
| Loss rate | 20% | 60% |
| Average win | Rs 500 | Rs 3,000 |
| Average loss | Rs 2,500 | Rs 1,000 |
| Win side of formula | 0.80 x 500 = Rs 400 | 0.40 x 3,000 = Rs 1,200 |
| Loss side of formula | 0.20 x 2,500 = Rs 500 | 0.60 x 1,000 = Rs 600 |
| Expectancy per trade | 400 - 500 = -Rs 100 | 1,200 - 600 = +Rs 600 |
| Result over 100 trades | -Rs 10,000 | +Rs 60,000 |
| Win rate needed to break even | 83.3% | 25% |
Read the last three rows slowly. Trader A wins twice as often as Trader B and loses Rs 10,000 over a hundred trades. Trader B is wrong most of the time and makes Rs 60,000. Same effort, opposite outcome. The only difference is that one chased accuracy and the other chased expectancy.
The classic beginner error is chasing win rate: closing winners early for a quick green tick, and widening the stop or "giving it room" so a loser does not become a realised loss today. You are literally moving the target closer and the stop further away to protect your accuracy. It feels great and it builds negative expectancy. The better move is the reverse: cut losers fast at a pre-set stop, let winners run to a target that is bigger than your risk, and judge yourself by expectancy, not by how often you are right.
How each type of trader meets this idea
Win rate and payoff look very different depending on what you trade. This is education, not advice for your specific situation.
| User type | Typical win rate | Payoff shape | What to actually chase |
|---|---|---|---|
| Long-term investor | Win rate barely matters | A few big multibaggers carry many duds | Hold winners; let the right tail do the work |
| Active trader | Often 40-55% | Mixed, depends on the system | Positive expectancy, never accuracy |
| F&O / option buyer | Low win rate, many small losses | Rare large wins pay for many small ones | A few big winners; cut the duds fast |
| Option seller | High win rate, most options expire worthless for the buyer | Many small wins, a rare fat loss | Survive the tail; one gap can erase months |
Notice the last two rows are mirror images, and both can have positive expectancy. The option buyer looks like a serial loser and the option seller looks like a genius, yet as earlier chapters on stop-losses and the loss tail showed, the seller's lovely 90% win rate hides a rare loss so large it can wipe out a year of premiums. High accuracy with an uncapped loss is Trader A wearing a costume. Expectancy, not the win-rate trophy, tells you which one is really making money.
When this fails
Expectancy is the right target, but the numbers behind it are slippery, so handle them honestly.
The biggest trap is a small sample. Twenty trades tell you almost nothing. A 70% win rate over fifteen trades can flip to 45% over the next fifty as luck evens out. You need a few hundred trades before your win rate and average win/loss are anything more than a guess, and beginners routinely "confirm" an edge that was pure chance.
The second trap is a changing market. Expectancy is measured looking backwards. A setup with lovely positive expectancy in a calm, trending market can turn negative the moment volatility spikes or the trend dies. The COVID crash of 2020 broke many systems that had looked bulletproof for years. Your edge is not a constant; it decays, and it must be re-checked.
Third, costs eat expectancy. Brokerage, STT, exchange fees and the gap between the price you wanted and the price you got all shrink the average win and enlarge the average loss. A trade that is +Rs 50 before costs can be negative after them. Always compute expectancy on net numbers, after charges.
And finally, expectancy assumes you actually follow your plan. The math of Trader B only works if she truly cuts every loser at the stop and lets winners run. The day fear makes her exit early or hope makes her hold a loser, she quietly becomes Trader A, and no formula can save her.
Quick self-check
1. Arjun wins 80% of his trades but loses money. How is that possible?
His wins are small and his losses are large. Eight small wins do not cover two big losses. A high win rate says nothing about the size of wins versus losses, so it can sit on top of a losing account.
2. What does a 1:3 risk-reward ratio mean, and what win rate do you need to break even with it?
It means you risk one unit to make three, for example risk Rs 1,000 to make Rs 3,000. With a 1:3 ratio you break even at just a 25% win rate, so you can be wrong three times out of four and still not lose money.
3. Write out the expectancy formula and what a negative answer tells you.
Expectancy = (win rate x average win) - (loss rate x average loss). A negative answer means each trade is worth less than zero on average, so the more you trade the more you lose. The fix is a bigger reward-to-risk ratio or a higher win rate, not more trading.
4. An option seller wins 90% of the time. Does that make the strategy safe?
No. A high win rate can hide a rare but enormous loss. If one bad gap loss is larger than many months of small premiums, expectancy can still be negative or the account can blow up in a single event. You must size for the fat tail, not the comfortable win rate.
5. Your new setup shows a 75% win rate over its first 18 trades. Should you increase your size?
No. Eighteen trades is far too small a sample to trust; the rate could easily be luck and may fall sharply over the next fifty trades. Wait for a few hundred trades, compute expectancy on costs-included numbers, and only then judge whether the edge is real.