The pack ranks itself alpha/beta/delta by current distance to the elk each decision tick, then every wolf's next position is the average of three encircling estimates pulled from those three leaders — the same alpha-beta-delta encircling model used in the Grey Wolf Optimizer algorithm (Mirjalili et al., 2014), here driving an actual hunt instead of a numeric optimizer. The coefficient "a" shrinks from 2 to 0 over each hunt, so the pack drifts from wide searching to a tight, coordinated encirclement as the chase goes on.