Como ler a sua curva de capital: o que o formato dela diz sobre o seu sistema
Quick answer: the final number in your account tells you whether you made money. The shape of the curve tells you whether you will keep making it. A healthy curve rises at a steady slope with small, similar-sized dips. When all the profit comes from two jumps, or when it rises and falls without progressing, the problem is almost never the strategy: it is size, frequency or execution.
Pontos principais
- Shape matters more than height: two curves can end identically and mean opposite things.
- Very uneven drawdowns indicate inconsistent position size.
- If removing your three best trades flattens the curve, you do not have a system.
- A curve that looks too perfect usually hides losers you have not closed yet.
Five shapes and what they mean
| Shape | What is happening | What to check |
|---|---|---|
| Even staircase | A real edge, executed consistently | Nothing. Keep the size and the routine. |
| Flat sawtooth | You gain and give it back; costs take the rest | Frequency and cost per trade |
| Two jumps and nothing else | The result depends on luck, not process | Expectancy without those trades |
| One sharp cliff | One oversized or out-of-plan trade | The risk taken on that single trade |
| Long plateau | The context stopped suiting your system | Market regime and frequency |
Three tests that expose a fragile system
- Remove your three best trades. If the curve goes flat or negative, you depend on exceptional cases, not an edge.
- Remove your three worst. If that completely changes the result, the problem is risk control, not the signal.
- Look at the drawdowns one by one. If some are twice as deep as others, your size is not constant.
Test 3 is the most revealing. In a properly sized system, drawdowns resemble each other because every trade risks the same amount. Wildly uneven drawdowns almost always mean you traded bigger on some days — usually the days you were already down.
Two curves that end the same
| Trader A | Trader B | |
|---|---|---|
| Result for the year | +20% | +20% |
| Largest drawdown | 6% | 31% |
| Positive months | 9 of 12 | 4 of 12 |
| Where the profit came from | Spread across trades | Two trades |
Same final number, two different realities. A can repeat it next year; B needs the same thing to happen again, and that is not a plan.
How to build it without overcomplicating
- Export your history and add up the cumulative result trade by trade.
- Plot it in R, not in money, so periods with different sizes stay comparable.
- Mark on the axis the days you changed something in the plan.
- Add a second line with your running peak so the drawdowns are visible.
- Review it once a month, not every day.
That last point matters more than it sounds: watching the curve daily produces strategy changes driven by noise. For the log that feeds it there is the free ebook The Trading Journal Blueprint. If your diagnosis points at inconsistent size, Cortex Automation keeps risk fixed as a parameter, and the Cortex Signal Kit delivers signals on the same criteria every time.
How often should I review my equity curve?
Once a month to make decisions, and after every 50 trades to evaluate the system. Daily only produces anxiety.
What maximum drawdown is acceptable?
The one you can sit through without changing your behaviour. Wherever you start breaking rules is your real limit.
Should my curve be a straight line?
No, and be suspicious if it is. An unusually smooth curve often means open losers you have not closed.
Can I compare my curve with another trader's?
Only if both are in R and cover a similar period. In money, with different sizes, the comparison says nothing.
What do I do after months of plateau?
Before changing the strategy, check plan compliance and costs. They usually explain the plateau better than the market does.
The shape tells you what to fix. See the Cortex Signal Kit and Cortex Automation.
Disclaimer: educational content. Examples are illustrative. Cortex Next is not a financial service and guarantees no results. Trading involves risk of loss.
