The null hypothesis is what the researcher is trying to disprove or nullify. It does not matter whether it is actually correct or not, the aim is to try and disprove it. If the significant tests of the hypothesis that 'earth was flat' does not
generate 95% or 99% likelihood that the results do not fit the
null hypothesis, then the null hypothesis(earth is flat) is accepted. However, that does not mean it is true.