How to pick features when there isn't a strong correlation to the target variable

Question

I have a list of features used to predict whether or not a team will cover the spread of an NFL game. The spread is a number set by Vegas which you add or subtract to the te