As artificial intelligence is increasingly being utilized for crucial image and video analysis tasks, football video analysis is also growing to become one of the interesting areas where several tasks have challenges associated with it. This proposal brings forth a total football video analysis system containing tasks such as object detection and tracking, position mapping, and team classification for tactical behavior analysis tasks. YOLOv8x and YOLOv8-Pose are harnessed for detecting players, balls, and key points on the pitch, while ByteTrack is used for real-time tracking of players for all their positions on the ground. The homography transformation technique is also used to position objects from the video frames to points on a 2D pitch map, while K-means clustering is also utilized for classifying players to their respective team colors depicted by their jerseys. The system also follows the ball’s motion to help strategists analyze motions associated with passing, shooting, and scoring goals. The experiment conducted shows high performance capabilities of all systems to have mAP = 0.5 scores of 0.993 for players’ detection tasks, 0.995 for key point detection tasks on the pitch areas, and 0.904 for all ball detection tasks. This approach is not only helpful for effective tactical analysis but also provides significant prospective functionality for interactive sports broadcasting and analysis systems for coaches and match analysis platforms.