Ships designed for operation in Polar waters must be approved in accordance with the International Code for Ships Operating in Polar Waters (Polar Code), adopted by the International Maritime Organization (IMO). To account for ice loading on ships, the Polar Code includes references to the International Association of Classification Societies’ (IACS) Polar Class (PC) standards. For the determination of design ice loads, the PC standards rely upon a method applying the principle of the conservation of momentum and energy in collisions. The method, which is known as the Popov Method, is fundamentally analytical, but because the ship–ice interaction process is complex and not fully understood, its practical applications, including the PC standards, rely upon multiple assumptions. In this study, to help naval architects make better-informed decisions in the design of Arctic ships, and to support progress towards goal-based design, we analyse the effect of the assumptions behind the Popov Method by comparing ice load predictions, calculated by the Method with corresponding full-scale ice load measurements. Our findings indicate that assumptions concerning the modelling of the ship–ice collision scenario, the ship–ice contact geometry and the ice conditions, among others, significantly affect how well the ice load prediction agrees with the measurements.