Chain and barrier pillar design for longwall mining and production pillar design for room-and pillar retreat mining have tended to rely on simplistic abutment angle concepts for the estimation of pillar stress increases during and subsequent to extraction. Historically, the underpinning database of monitored abutment loading has been small and displayed considerable variation, leading to the application of a number of mine site-specific approximations and often necessarily conservative assumptions. Also, over the last decade, the trend towards wider longwall faces and narrower room-and-pillar sections in deeper areas has challenged established design practices. However, in recent years, considerable effort has been made both in the US and Australia with regard to expanding the abutment loading database and developing an improved understanding of the pillar loading environment. This paper examines some of the progress made and the implications thereof, with a focus on the derivation of formula for abutment angle prediction.