Glossary
P-value
The probability of seeing a result at least as extreme as the one observed if the null hypothesis were true. It is not the probability that the hypothesis is true.
A p-value comes from a test. First assume the null hypothesis, for example that the true mean return is zero. Then ask how often a result this far from zero would arise by chance. A small p-value means the data are hard to explain by chance alone.
It does not measure the size of an effect, nor the chance that the idea works, and it loses its meaning when many tests are run and only the best is reported.
Example: a t-statistic of 2.0 has a two-sided p-value of 4.55% under the normal approximation.