"80% winning trades." It is the most common figure in investing marketing, and on its own it says very little. A strategy can win four times out of five and lose money; another can win less than half the time and do well. This guide explains what to ask before believing any success rate, ours included.
1. How many cases is it based on?
A success rate measured on 10 trades and one measured on 1,000 are not the same claim.
With few cases, chance dominates. A coin tossed 10 times comes up heads 7 times or more about one time in six, without any skill involved. With 10 trades, a "70% success rate" is perfectly compatible with no edge at all.
That is why a serious figure always comes with its sample size, and ideally with a margin of uncertainty. A useful rule of thumb: below about 30 cases, treat any success rate as an anecdote.
2. How big are the wins and the losses?
The success rate counts how often you win, not how much. What matters in the end is the combination:
Strategy A wins far more often and loses money. Strategy B loses more often than it wins and comes out ahead. The figure that combines both is the average result per trade, sometimes called the expectancy.
Strategies with small targets and wider stops naturally have high success rates; that tells you about their design, not about their quality.
3. Better than what?
A strategy that gained 12% in a year in which the market gained 20% has made money and lost to the simplest alternative: buying the index and doing nothing.
Ask what the result is compared with the market over the same days, and remember that the market is invested every day while most strategies spend part of the time in cash.
4. Was the rule chosen before or after seeing the data?
This is the most powerful bias and the hardest to see. If you test hundreds of variations of a rule on the past and keep the best one, its historical result will look excellent almost by construction. On new data it usually disappoints.
Three questions help to detect it:
- Was the rule fixed before the data it is judged on? A result measured on data the rule had never seen ("out of sample") is worth much more than one on the data used to build it.
- Could the rule have known the future? A test that uses information that was not available at the time of the decision (for example, a company's results published days later) inflates every figure.
- Are the losers still in the list? Studying only the stocks that exist today ignores the companies that went bankrupt or left the index, and makes any past look better.
5. Is it the record of what was published or a simulation?
A simulation (backtest) applies today's rules to past prices. It is useful to understand a method, but it describes what the rules would have done.
A record lists what was actually published, on the day, with the levels that were given, and what happened afterwards. Only a record shows what a reader could really have followed.
How we show our figures
We try to apply these same questions to ourselves:
- Every success rate comes with its sample size and a confidence label: High from 30 cases, Medium from 15 to 29, Low below 15. A high percentage on a handful of trades is labelled as what it is.
- The public scoreboard lists every published signal, losers and signals whose order was never filled included, not a selection of winners.
- Its average result against the market is not shown until the record reaches 300 closed operations, a threshold fixed in advance, in both directions: a good figure is withheld exactly like a bad one.
- The rule it measures was registered before its data, with a fixed decision date, so it cannot be adjusted to fit the results.
- Every return we publish is before tax, because tax depends on each reader's situation.
The details are on the methodology page.
A checklist for any success rate
- How many cases?
- How big are the average win and the average loss?
- Compared with the market over the same days?
- Measured on data the rule had never seen?
- Published record or simulation?
- Are the losers still in the list?
If a figure cannot answer these questions, it is advertising, not evidence.
In short
- A success rate without its sample size is an anecdote.
- Winning often is not the same as winning: the size of gains and losses decides.
- Compare with the market, over the same days.
- Distrust any rule chosen after seeing the data, and any list without its losers.
This content is for educational purposes only and does not constitute financial advice. Past performance does not guarantee future results.