The best sample for estimating the number of students who bring their lunch to school every day would be:
80 students selected at random during lunch.
This choice is the best because it includes a broader group of students who are actually present during lunch and likely to have brought their lunch. It is more representative of the entire student population compared to the other options, which may be biased by factors such as grade level (first-year only), gender (female only), or club membership (Math Club).