Existing design codes and most methods for ice load calculation for conical structures do not take velocity effects into account. They were developed as an upper bound estimate for the load from slow moving ice which fails in bending against the cone. Velocity effects can be ignored when the structure is designed for an area with slow ice movement, for example, the nearshore Beaufort Sea. Sakhalin structures will be exposed to ice moving at velocities up to about 1.5 m/sec. Model tests show that quasi-static methods may underestimate the ice load on a steep cone when the interaction velocity is that high. The present paper summarizes results of published model tests with conical structures that show a velocity effect. An empirical correction factor to the Ralston method is developed to account for the increase in cone load with ice velocity. The paper also discusses velocity effects on ice failure length and possible transition from bending failure to an alternative failure mode when the ice velocity is high.