Based on the graph 400, the inventors arrived at additional conclusions with respect to viewing habits of viewers and the relationship between those viewing habits and content similarity. Like the graph 300 in FIG. 3, the graph 400 includes respective weekly peaks 406 (collectively) and respective daily peaks 408 (collectively) for the similarity scores, S. Again for the sake of clarity, not all daily peaks 408 have been labeled in FIG. 4. The daily peaks 408 on the graph 400 indicate that video content viewed around the same time each day (e.g., some multiple of 24 hours apart) is relatively more similar than content viewed for other time differences (e.g., 10, 12, 18, 20, etc., hours apart). The weekly peaks 406 similarly indicate that video content viewed around the same time each week (e.g., some multiple of 168 hours apart) is also relatively more similar than content viewed for other time differences (e.g., 158, 178, 326, 346 494, 514, etc., hours apart). The graph 400 further indicates that video content viewed around the same time each week is relatively more similar than video content viewed around the same time each day. As also seen in the graph 400 of FIG. 4, the similarity scores associated with the weekly peaks 406 are relatively higher than the similarity scores associated with the daily peaks 408, and the similarity scores for the weekly/daily programs were higher than other program pairings that were not weekly or daily.