LEARN · EN · easyquanttrading.com
How to read an equity curve
Most people look at the last number on an equity curve and nothing else. That is like judging a book by its length. The shape tells you how the strategy behaves when it is not working, and that is the part you will have to live with.
- The endpoint is the least informative part of the curve
- Look at the slope's consistency, not its total
- Flat periods are information, not empty space
- Recovery time matters as much as drawdown depth
What the curve actually plots
An equity curve is account value over time, generated by compounding each trade's result. That compounding is why the shape matters: a strategy that loses 10% then gains 10% ends below where it started, because the gain is computed on a smaller base.
This asymmetry has a practical consequence for reading curves. A curve that rises steadily and then gives back a third of its gains looks very different from one that falls and then recovers, even if both end at the same value.
The four features worth examining
Ignore the endpoint for a moment and look at these instead. Each one answers a different question about how the strategy behaves.
| Feature | What to look for | What a bad version means |
|---|---|---|
| Slope consistency | Even progress across the whole period | All the gain came from one stretch |
| Drawdown depth | The worst peak-to-trough fall | You may not be able to hold through it |
| Drawdown duration | How long it spent underwater | Opportunity cost and morale |
| Recovery shape | A clean recovery or a long grind | The strategy needed a specific condition to recover |
Four features that describe behaviour. A single endpoint number describes none of them.
Why flat periods deserve attention
A long flat stretch is not a neutral event. It usually means the market conditions the strategy depends on were absent, which tells you something about what the strategy is actually betting on.
It also predicts your own behaviour. A strategy that produced no progress for nine months is one most people abandon in month four, right before it resumes. Knowing the historical flat periods tells you what you are signing up for.
The single-stretch test
Find the best three-month stretch on the curve and imagine removing it. Does the strategy still look worthwhile? If the answer is no, then the result is a statement about one period rather than a description of a repeatable process.
This is not an argument that such strategies are worthless — some genuine edges arrive in bursts. It is an argument that the confidence you can place in the result depends on how concentrated it is, and that concentration is visible in the shape.
Comparing curves, not numbers
When choosing between two candidates, overlay the curves rather than comparing summary statistics. The visual comparison surfaces things the numbers hide: which one recovered faster, which one produced its returns steadily, which one spent longer underwater.
Then compare both against a benchmark — simply holding the instrument over the same period. It is a crude baseline and it is surprisingly hard to beat after costs, which makes it a useful reality check on any curve that looks impressive in isolation.
FAQ
- Is a smoother equity curve always better?
- No, and extreme smoothness is a warning sign. Real markets are noisy, so a curve with almost no drawdown usually indicates unrealistic assumptions about costs or fills, or a look-ahead error in the logic.
- What is a good recovery time?
- There is no universal figure, but the useful comparison is against your own patience. A strategy that took eighteen months to recover from its worst drawdown requires an eighteen-month commitment, regardless of what its annualised return claims.
- Should I judge a strategy on the last year only?
- Almost never. A single year usually contains one regime. The most recent period is the least informative part of the curve precisely because it is the shortest, unless something structural has changed.
More guides
- How EasyQuant validates strategies — evidence you can filter
- Honest backtesting, not pretty curves
- Gold strategy research that stays honest
- Overfitting detection: catch it before you deploy
- System Forge: design, then prove
- Walk-forward analysis: the only backtest that fights overfitting
- MT5 export without custody
- Glass box, not black box AI signals
Not investment advice. Historical results do not guarantee future performance. EasyQuant is a research factory — you execute on accounts you control.