The alternative hypothesis is the statement, in hypothesis testing, that an effect, difference or relationship is present. It is written or and is set against the null hypothesis , which says there is none. A test never proves the alternative directly. It asks whether the data would be surprising if the null were true, and if the p-value falls below the chosen significance level, the null is rejected in favor of the alternative.
The alternative can be two-sided (the mean return is not zero) or one-sided (the mean return is greater than zero), and the choice must be made before looking at the data. In a strategy backtest, the null is usually that the strategy has no edge and the alternative is that its average return is positive. Testing many strategies and keeping the best one makes a false rejection likely, so the evidence needed should rise with the number of trials.