Once a view is formed it develops its own support structure. You have arguments for it, you have told people about it, and you have positions expressing it. Removing the original evidence rarely removes the belief.
In markets this looks like a thesis that survives every contradiction. The data that prompted a short comes out revised, the catalyst passes without effect, and the short is still on with fresh reasons attached. Each new reason arrived after the conclusion, which is the tell.
A written thesis with falsification conditions is the practical defence. Decide in advance what would make you wrong, in observable terms, and check that list rather than your feelings about the trade.
Related: motivated-reasoning, confirmation-bias, escalation-of-commitment, cherry-picking