The study compares the performance of alternative implementations of both time-series and cross-sectional momentum strategies across 24 markets. We find that over our sample period, both types of momentum strategies generate positive returns under the majority of implementations evaluated but that time-series momentum is clearly superior. An important difference between the two momentum strategies is that with time-series momentum, the number of stocks included in the winner and loser portfolios vary with the state of the market. As a consequence, cross-sectional momentum digs deeper to select winning stocks when markets are weak and deeper to select losing stocks when markets are strong. As the information in the momentum signals is concentrated in the tails of the return distribution, it is not that surprising that momentum is best implemented using time-series momentum.