Null hypothesis is a term used in statistics.  It is part of a method of making a decision based on data. A null hypothesis is the opposite of the alternative hypothesis, which is the one that the analyst is ultimately trying to prove. The null hypothesis is assumed true until proven otherwise.  To test the null hypothesis, a random sample is taken from a data set.  If the null hypothesis is disproved by a significant statistical variance, then the null hypothesis is rejected and the alternative hypothesis is said to be true. Null hypotheses are used in quantitative analysis to test theories about markets, investing strategies or economies to decide if the idea is true or false. Mark has been using an investment strategy that he believes is getting better returns than the market. The null hypothesis is that there is no difference between Mark’s investment strategy and buying a market-based index fund.  Mark has to believe this null hypothesis until he can prove otherwise.  There are a number of ways to sample data from his return results.  If he tests the data and determines that there is a significant difference between the returns for his investment strategy and the return of a market index fund, then he will have refuted the null hypothesis.  Thus the alternative hypothesis, that his investment strategy is better than the market fund, will be affirmed.