In a sustained advance, almost any long strategy works, leverage is rewarded, and buying every decline is reinforced. Confidence rises with the account, and the attribution goes to method rather than conditions - which is self-attribution-bias operating across a whole cohort.
The phrase is normally used in retrospect. Applied in advance it is a genuinely useful question: how much of my result came from the position I hold rather than from the decisions I made? A strategy that has never traded a different regime has not been tested, only rewarded.
Related: self-attribution-bias, luck-vs-skill, tourist, euphoria