We present algorithms for a new self-adjusting binary search tree, which we call ashuffle tree. The tree is easy to implement and does not require parent pointers orbalancing information to be stored in tree nodes. A maximum of one rotation isapplied randomly during each traversal, which keeps the cost of balancingactivity low.