Win rate is the first number anyone posts and the least informative one available. It answers how often, and says nothing about how much. A strategy that wins 90% of the time while risking five units to make one is losing money at a hit rate that sounds untouchable. A strategy that wins 40% of the time while making three units for every one risked is comfortably profitable. The market does not pay out in percentages of correctness.
The arithmetic in one line
Expectancy per trade is the win rate multiplied by the average win, minus the loss rate multiplied by the average loss. Run the two examples. Nine wins of one unit and one loss of five units nets four units over ten trades, which is positive, so a 90% system with a five to one adverse payoff can survive. Push the adverse payoff to ten to one and the same 90% hit rate loses a unit over the same ten trades. Now the 40% system: four wins of three units and six losses of one unit nets six units. Better outcome, less than half the hit rate.
What the arithmetic makes obvious is that hit rate and payoff are two halves of one number. Quoting either alone is quoting half a sentence. The full version is set out in expectancy, and the payoff half is what risk to reward ratios measure.
| Win rate | Average win | Average loss | Net over 10 trades |
|---|---|---|---|
| 90% | 1.0 | 5.0 | +4.0 |
| 90% | 1.0 | 10.0 | -1.0 |
| 40% | 3.0 | 1.0 | +6.0 |
| 40% | 1.5 | 1.0 | 0.0 |
Why high hit rates hide risk
Strategies engineered for a high win rate almost always achieve it by making the loss rare and large rather than frequent and small. Wide stops do it. Adding to losers does it. Removing the stop entirely and waiting for the position to come back does it best of all, right up until the day it does not. Martingale sizing is the extreme version: it can produce a very long string of small wins and one loss that removes the account.
The psychological trap is that this feels correct while it is working. Twenty five winning trades in a row is emotionally indistinguishable from skill. The distribution of outcomes has a fat left tail that has simply not been sampled yet, and no amount of confidence changes when it arrives.
Any published track record with a very high win rate deserves the same first question: what is the largest single loss, and how does it compare to the average win? If that number is missing, the record is not a record.
What a 40% system demands from you
Low hit rate systems are mathematically comfortable and psychologically brutal. At 40%, a run of six or seven consecutive losses is an ordinary event, not a sign that anything is broken. Traders who cannot sit through that will interfere, cut the winners early to book something green, and destroy the payoff half of the equation that made the system work.
Trend following lives here. Most entries fail, a small number run a long way, and the entire result comes from the handful that were allowed to run. Cutting winners at one to one in a system that needs three to one turns a profitable approach into a losing one without changing a single entry rule. The behavioural side of that is the same problem discussed in trading psychology.
Numbers worth tracking instead
Four figures tell you more than win rate ever will. Average win divided by average loss, which fixes the payoff. Expectancy per trade in account currency, which tells you what one more trade is worth. Profit factor, gross profit over gross loss. And maximum drawdown, which tells you what the equity curve did on its worst stretch and therefore whether you could have stayed in the seat.
Sample size matters as much as the numbers. Thirty trades is an anecdote. Two hundred trades across different volatility regimes starts to be evidence, and even then the confidence interval around a 40% hit rate is wide. Anyone quoting a win rate from a two week sample is quoting noise.
How the number gets used against you
Win rate is the headline on almost every signal service advert because it is the metric that sounds best without disclosing anything. A service can quote a high hit rate while running a stop that is many times the take profit, and every statement in the advert will be technically true. The check is simple: ask for average win, average loss and worst drawdown alongside it, and treat a refusal as an answer. That is the core of how to evaluate signal services.
There is also a cost layer that the raw percentages ignore. Spread, commission and swap come out of every trade, and they hit a high frequency, low payoff system far harder than a low frequency one. A scalping approach taking a hundred trades a week at a one to one payoff has to clear its transaction costs a hundred times, so a hit rate that looks comfortably above break even on paper can be below it in the account. The same edge expressed over ten trades a week with a three to one payoff pays that toll a tenth as often. Cost sensitivity is part of the design, not an accounting detail to be added afterwards, and it is the reason two traders running the same signals can end a quarter on opposite sides of flat.
None of this makes any strategy safe. Leveraged trading carries a high risk of loss whatever the distribution of outcomes looks like, and a positive expectancy calculated on past trades is a description of the past, not a promise about the next hundred.
"Show me a ninety percent win rate and I will ask what the biggest single loss was. If nobody wants to answer that, I already know."
— Alex Onta, Executive Director, SINGUARD
Key Takeaways
- Expectancy is win rate times average win minus loss rate times average loss. Half of it is not a result.
- High win rates are usually bought with rare, oversized losses in the left tail.
- A 40% system needs the payoff protected, which means not cutting winners to feel better.
- Track payoff ratio, expectancy, profit factor and maximum drawdown, over a sample large enough to mean something.
Frequently Asked Questions
Is a high win rate always a bad sign?
No. It is an incomplete sign. A high win rate paired with a small average loss and a documented worst case is fine. A high win rate quoted alone usually means the average loss is the part being left out.
What win rate does a trader need to be profitable?
There is no single figure, because it depends entirely on the payoff. At three to one reward against risk, a hit rate near 30% breaks even before costs. At one to one, it takes more than half plus the cost of spread and commission.
How many trades are needed before statistics mean anything?
A few dozen trades is an anecdote. A couple of hundred across different market conditions begins to be evidence, and even then the range of plausible true win rates around a measured one stays wide.
About the Author
Alex Onta is an Executive Director at SINGUARD. He built eTrader, the terminal, the mobile apps, eTrader Broker, Copytrading, Business and Community, along with the worldwide clustered-server infrastructure it all runs on, with his brother Roman Onta helping on the design, and he leads that division today. Together with Roman he builds the Prop Firm CRM, the Broker CRM, Scalegram and CopySignals, and the two of them carry worldwide compliance, payment processing and international business structuring side by side. He lives and works in Dubai for most of the year. Meet the executive duo leading Singuard's five divisions.