In most skills, feedback is fast and honest. Miss a cricket ball and you know immediately. Markets are the opposite: outcomes are dominated by noise, feedback arrives late, and a bad process can be rewarded for years. This makes investing one of the hardest fields in which to learn from experience.
You shoot arrows into fog and hear a thud. Sometimes the thud means a bullseye and sometimes the wind carried a bad shot onto the target. If you only hear thuds, you cannot tell whether your technique improved.
A profitable year is a thud. It may be skill, or it may be a rising market carrying a poor process. Without separating the two, you will conclude you are skilled — and increase your size accordingly.
The sample size problem
Distinguishing a genuine edge from randomness requires enough trials that luck averages out. For the effect sizes present in markets, that number is uncomfortably large.
| Your edge | Trades needed for reasonable confidence | At 5 trades a month |
|---|---|---|
| Large — 60% win rate at 1:1 | ~100–200 | 2–3 years |
| Moderate — 55% at 1:1 | ~400–800 | 7–13 years |
| Small — 52% at 1:1 | ~2,000+ | A working lifetime |
| Investing, 5–8 positions a year | Effectively decades | You may never know |
What that implies
If outcomes cannot tell you whether you are skilled within a useful timeframe, you must judge yourself on something that produces feedback faster. That something is process.
- Did I follow my written rules?
- Was the position correctly sized?
- Did I record the thesis before buying?
- Did I act on evidence or on a price move?
- Did I beat the index?
- Was this year profitable?
- Did that stock go up?
- Am I good at this?
Small edges take enormous samples to show through the noise. Adjust the win rate and see how little a short run of outcomes actually tells you.
Attributing your returns
- 1Start with the market
If the index rose 22% and you made 25%, the market provided the large majority. That is worth knowing before drawing conclusions about your ability.
- 2Then the style
Smallcaps rose 40% and you hold smallcaps. Much of the remainder is factor exposure, not selection.
- 3Then concentration
Check whether one position produced most of the gain. A portfolio carried by a single holding is a much smaller sample than the number of positions suggests.
- 4What is left is the candidate
Whatever remains after market, style and concentration is the part that might be skill — and even that needs many years to confirm.
Three good years
You have beaten the index three years running. You are considering quitting your job to trade full-time and increasing your position sizes substantially.
Why is process evaluation more useful than outcome evaluation for an investor?
Dhund mein teer chalaye aur "thak" ki aawaz aayi — nishane pe laga ya hawa ne pahuncha diya, pata nahi. Teen saal ka profit bhi aisa hi hai. Market upar thi ya aap sahi the — yeh alag karna bahut mushkil hai, aur usi confusion mein log size badha dete hain.
- Market feedback is delayed and noisy, so experience alone teaches slowly and often wrongly.
- Distinguishing a real edge from luck takes hundreds to thousands of decisions.
- A long-term investor may never accumulate enough decisions to know from outcomes.
- Judge yourself on process, which gives feedback immediately.
- Attribute returns to market, style and concentration before crediting skill.
Mark it done to track your progress through the curriculum.
Common questions
Short, direct answers to what people ask about this topic.
- how many trades before you know if you have an edge
- Far more than most people expect — roughly 100 to 200 trades for a large edge such as a 60% win rate at 1:1, several hundred for a moderate one, and a few thousand for a marginal 52% win rate. The reason is that the edges available in markets are small relative to the noise around them, so luck needs a large sample to average out. An investor taking five to eight positions a year may never accumulate enough independent decisions to answer the question from outcomes at all.
- separating skill from luck in market results requires
- A large enough sample of independent decisions for randomness to cancel out. Because a good process and a bad one can both make money for years in a rising market, a single profitable year — or even three — settles nothing. This is why the honest answer for most long-term investors is that outcome-based self-assessment is close to impossible within a career: the data required simply does not exist yet.
- does beating the index for three years prove I am a skilled investor
- No — three years is typically only 60 to 150 decisions and usually a single market regime, well short of what separates skill from luck at realistic edge sizes. Before drawing conclusions, attribute the return in order: how much came from the index rising, how much from the style you happened to hold such as smallcaps, and how much from one or two concentrated positions. Whatever remains after those three is the candidate for skill, and even that needs many more years to confirm.
- what does judging your process instead of your outcome mean
- It means grading the decision you made rather than the result it happened to produce — did you follow your written rules, size the position correctly, record the thesis before buying, and act on evidence rather than on a price move. Process questions can be answered the same day; outcome questions like “am I any good at this” need a decade of data. In a field where feedback is delayed and noisy, process is the only signal that arrives fast enough to learn from.
- why is a bull market bad for judging your own ability
- Because almost any process makes money in one, so gains feel like validation of the specific thing you did. Position sizes then grow to match the confidence rather than the evidence, and the bill for those sizes arrives in the following bear market — which is why many people discover their real skill level at the worst possible moment. A useful habit is to write one paragraph each year explaining your return without using the word “I”.