© 2018 Elsevier B.V. The contribution of this paper is to enable solid conclusions to be drawn about the existence of momentum effects in China as the current evidence is unsatisfactory. We review and analyse the existing empirical studies on momentum and contrarian strategies in China and show that many of the findings in these studies appear inconsistent, if not actually contradictory. To clarify this confused situation we initially identify common findings in the diverse and seemingly contradictory body of existing empirical evidence. Subsequently, we systematically assess how the design of empirical studies affects the results of investigations in this area. We do this by conducting an empirical analysis of monthly data on Chinese A shares, varying one factor in the research design at a time (sample period, equally or value-weighed portfolios, skipping a period between portfolio formation and holding periods, and exclusion of post-IPO observations). This allows us to pinpoint directly how each of these factors affects momentum profits and thus when these profits are likely to be observed. It also indicates why studies using different designs might have arrived at seemingly inconsistent conclusions. Overall, we draw a number of conclusions: there appear to exist medium- and longer-term reversals in the pre-2001 period and short-term reversals and longer-term momentum effects thereafter; there is substantial time-variation in the profits to momentum strategies; small stocks exhibit stronger reversals than their larger counterparts; a large fraction of portfolio returns occur in the first month after formation; there is evidence of post-IPO price drifts. In summary, this study reconciles and explains the inconsistent evidence on the existence of momentum and contrarian effects in China allowing clear conclusions to be drawn.