The 70% Setup That Loses Money
Imagine two historical setups.
Setup A wins 70% of the time.
When it works, the average gain is 1%.
When it fails, the average loss is 4%.
Setup B wins only 45% of the time.
When it works, the average gain is 6%.
When it fails, the average loss is 2%.
Most traders would choose Setup A without hesitation.
The win rate is higher. The setup feels safer.
The historical evidence suggests otherwise.
Over a large number of trades, Setup B may produce significantly better results despite being wrong more often.
This is one of the most important lessons in market analysis:
Being right frequently and making money consistently are not the same thing.
Why Traders Love Win Rate
Win rate is one of the most intuitive statistics in trading because it appears to answer a simple question:
How often does this work?
Most people naturally prefer certainty over uncertainty. A setup that wins 70% of the time feels safer than a setup that wins only 45% of the time. The higher number creates an impression of reliability, consistency, and predictability.
This reaction is understandable.
Outside of financial markets, a higher success rate is usually a good thing. If a doctor offers a treatment with a 90% success rate, most people would prefer it to a treatment with a 50% success rate. If a manufacturing process succeeds 99% of the time, it is generally better than one that succeeds only 80% of the time.
Trading feels like it should work the same way.
The problem is that financial outcomes contain an additional dimension that success rates alone cannot capture.
Not all wins are equal.
Not all losses are equal.
A setup that wins frequently but generates small gains may ultimately be less attractive than a setup that wins less often but produces substantially larger gains when it succeeds.
This is where win rate begins to lose its usefulness as a standalone statistic.
What Win Rate Actually Measures
Win rate answers a very specific question:
How often did the outcome end up positive?
That information is useful.
The mistake is assuming it tells us more than it actually does.
A win rate tells us nothing about the size of the gains. It tells us nothing about the size of the losses. It tells us nothing about the variability of outcomes, the severity of drawdowns, or the distribution of returns.
Two setups can have identical win rates while producing dramatically different experiences for traders.
One may generate small, consistent gains with limited downside risk. Another may produce occasional large losses that overwhelm months of positive results. Looking only at the headline statistic would make these setups appear identical despite behaving very differently in practice.
This is why experienced traders rarely evaluate opportunities using a single metric. A number can be accurate while still being incomplete.
Win rate is one piece of evidence.
It is not the entire story.
The Missing Half Of The Story
Consider two businesses.
The first business closes sales with 80% of its customers but earns only a small amount on each transaction.
The second business closes sales with 40% of its customers but earns significantly more on each successful sale.
Without knowing the size of the profits, it would be impossible to determine which business is more attractive.
Trading operates under the same principle.
A setup that wins 80% of the time sounds impressive until we understand the size of the gains and losses. If the average winner is small and the occasional loser is large, the headline statistic becomes far less meaningful.
This is why professional traders often spend less time asking:
How often does this work?
and more time asking:
What happens when it works, and what happens when it fails?
The second question contains far more information.
Ultimately, trading results are determined by outcomes rather than accuracy. The market does not reward traders for being correct frequently. It rewards them for managing risk and generating favorable outcomes over time.
A trader can be wrong more often than they are right and still perform exceptionally well.
Conversely, a trader can be right most of the time and still struggle to make progress.
The difference lies in the distribution of outcomes rather than the frequency of success.
When High Win Rates Become Dangerous
One of the most misleading characteristics of a high win rate is that it can create a false sense of security.
A setup that wins 80% or 90% of the time naturally feels reliable. After all, if something succeeds almost every time, what could possibly be wrong with it?
The answer lies in what happens during the remaining 10% or 20% of outcomes.
A strategy can win frequently while still exposing traders to significant downside risk. In some cases, the very reason a strategy achieves a high win rate is because it accepts the possibility of occasional large losses.
This pattern appears repeatedly throughout financial markets. Strategies that collect small gains consistently often appear attractive during normal market conditions. Their weakness only becomes visible when an unusual event occurs.
A strategy can spend months building confidence and only a few days destroying it.
The lesson is not that high win rates are bad.
The lesson is that high win rates should trigger additional questions.
What do losses look like?
How large can they become?
How often do extreme outcomes occur?
Without answers to those questions, a win rate alone provides very little information about actual risk.
Why Distribution Matters More Than Most Traders Realize
Historical outcomes are rarely distributed evenly.
Two setups can produce identical win rates while generating completely different outcome profiles.
One setup may deliver remarkably consistent results. Most historical cases cluster around a similar outcome, creating a relatively predictable range of possibilities.
Another setup may produce a handful of exceptional winners, a large number of average outcomes, and a small number of severe losses.
Both setups could display the same win rate.
Yet they represent very different opportunities.
This is why experienced researchers spend significant time studying the distribution of outcomes rather than focusing exclusively on a single headline statistic.
The average outcome matters.
The median outcome matters.
The worst historical outcomes matter.
The spread of results matters.
Viewed together, these measures provide a much more complete picture than a win rate alone ever could.
Historical Analysis Is About Outcomes, Not Accuracy
Many traders approach analysis by asking:
How often was this setup right?
At first glance, this seems like a reasonable question.
The problem is that markets are not exams. Traders are not rewarded for accumulating correct answers. A trader can be right ten times in a row and still lose money if the eleventh outcome overwhelms the gains that came before it. Conversely, a trader can be wrong more often than right and still perform exceptionally well if favorable outcomes are large enough to compensate for the losses.
What matters is not simply how often a setup succeeds. What matters is the relationship between gains, losses, risk, and uncertainty.
Historical analysis is not about determining whether a setup was right or wrong.
It is about understanding what actually happened.
Better Questions Lead To Better Decisions
The quality of any analysis is heavily influenced by the quality of the questions being asked.
A trader focused primarily on win rate may ask:
How often did this setup work?
A trader focused on evidence tends to ask broader questions:
What happened when it worked?
What happened when it failed?
How consistent were the outcomes?
What risks emerged in the worst cases?
These questions do not necessarily produce simple answers.
They produce better understanding.
Viewed this way, win rate becomes what it always should have been:
A useful piece of evidence.
Not the conclusion.
Final Thoughts
The 70% setup from the beginning of this article may or may not be a good opportunity.
The win rate alone cannot tell us.
To understand the setup properly, we would need to know what the gains looked like, what the losses looked like, how consistent the outcomes were, and how the distribution behaved across historical examples.
In other words, we would need more evidence.
That idea extends far beyond win rate.
The most important decisions in markets are rarely improved by having a single statistic. They are improved by understanding the evidence from multiple angles and appreciating both the opportunities and the risks embedded within the historical record.
A statistic can be accurate. It can even be important. That does not mean it tells the entire story.
And in markets, understanding the difference is often more valuable than the number itself.