If your prediction is proven incorrect you should leave out all observations that don't support your prediction or simply change your prediction?
It is not scientifically valid to selectively ignore information that contradicts a prediction or change a prediction based on preconceived beliefs. It is important to consider all evidence, even if it goes against the initial prediction, in order to draw valid conclusions and refine future predictions. This approach is fundamental to the scientific method and ensures the integrity of the research process.