In testing the validity of a multiple regression model, a significant F-test statistic indicates that:
the model provides a poor fit.
at least one independent variable is adding significant predictability for Y
most of the variation in Y is unexplained by the regression equation.
most of the variation in the independent variables is explained by the variation in Y.