Glossary
What Is an Edge in Trading
A positive expectancy after fees over a large sample of your trades on a specific setup. An edge is never guaranteed: the journal shows it, and it can disappear.
How it works
An edge is a stable positive expectancy over a long series of trades tied to a specific setup or entry condition. It's not one lucky trade or a short lucky streak — it's a pattern confirmed by journal statistics across dozens or hundreds of trades, spanning different periods and market conditions.
You can only find an edge by tracking results per setup separately: without breaking trades down by setup, the overall account result can hide one profitable setup and several losing ones that offset each other, making it impossible to see where the real advantage actually sits.
The flip side of an edge is a "leak": a setup or condition with consistently negative expectancy that steadily loses money by the same logic. Improving a system usually comes down to trading a confirmed edge more and a confirmed leak less.
Example
Over 120 trades, the "level breakout in the New York session" setup produced +0.6R expectancy per trade. Over the same period, the "Asian session reversal" setup produced −0.3R. The first setup is a confirmed edge, the second a leak in the same system.
Common mistake
A common mistake is declaring an edge found after 10–15 good-looking trades. At that sample size, randomness can easily produce a positive number that won't hold up over a larger sample.
Read next
Related terms
Frequently asked questions
- How many trades does it take to confirm an edge?
- There's no fixed number, but the smaller the sample, the more randomness it contains; most traders use 50–100+ trades on a specific setup as a rough benchmark.
- Can an edge disappear over time?
- Yes, market conditions change, and a setup that worked a year ago can stop producing positive expectancy — which is why edge stats should be revisited periodically.
Plan versus outcome — for every trade
The Trading OS journal keeps the planned RR and the actual R of each trade side by side and computes expectancy from real results.
No sign-up — the journal runs in your browser; an account is only needed for sync.