I like to include trends into the handicapping process because they occasionally give insight into a teams philosophy (preparation for con vs. non con games, etc.). I was wondering if there are any standard sample sizes needed for an outcome to be included in category. Suppose Statfox claims"

TEAM A is on a 13-5 ATS run at home vs. very good teams (teams with win % greater than 75%)

Suppose this team had a home game in the 2nd week of the season against Troy who was coming off a win against Western Kentucky, would that game be included?

I realize this is a basic question but I appreciate any insight. I imagine the reason all of this information is so readily available is similar to many money managers objectives. While it is great to have an opinion on direction, activity is the primary goal.