A trading edge is a repeatable reason your process comes out ahead over a large number of trades — a positive expectancy. Not a hunch, not a hot streak, not a guru's blessing. Understanding what qualifies (and what doesn't) protects you from a lot of expensive fantasies.
What an edge is not
A few wins in a row isn't an edge — it's variance, and variance reverses (process over outcome). A backtest that looks perfect isn't an edge — it's often curve-fitting about to fail live (overfitting). A tip isn't an edge — it's a one-off you can't repeat or verify (why analysis beats a tip). If you can't repeat it and measure it, it isn't an edge.
What a real edge looks like
A real edge is a small, persistent statistical tilt you can express as a rule and measure over hundreds of trades: this setup, sized this way, has positive expectancy after costs (expected value, win rate vs profit factor). It's usually unglamorous and often uncomfortable to trade — because if it were obvious and easy, it would be crowded and gone (why obvious edges vanish).
An edge isn't a prediction that you're right. It's a reason you come out ahead across many trades — even while you're wrong on plenty of them.
Edge is necessary, not sufficient
Even a real edge only pays if you execute it consistently and size it to survive the losing streaks (risk of ruin, consistency over being right). A real edge, traded emotionally or oversized, still loses. That's why the edge is only half the job — the other half is disciplined, consistent execution, which is exactly what a mechanical system is built to provide (emotional discipline).