In this paper, we examine whether the short-term variation in the size and value premium in the Japanese stock market is sufficiently predictable to be exploited by a tactical timing strategy. In the spirit of Pesaran and Timmermann (1995), we employ a dynamic modeling approach in which we explicitly allow for permutations among the determinants in order to mitigate typical data-snooping biases. Using a base set of candidate regressors, we first perform an in-sample estimation of all economically sensible models in a rolling window framework. Subsequently, we generate out-of-sample forecasts for all models and evaluate the performance in a model training period. Model selection is based on three well-known selection criteria: hit ratio (% correct sign), the mean return of the strategy and the realized information ratio. We then implement our investment strategy in a second stage out-of-sample period: the trading period. The empirical results confirm sufficient predictability under low levels of transactions costs. Under higher transaction costs scenarios we find it difficult to obtain incremental benefits from style rotation strategies. Moreover, the results are entangled with the choice of forecast horizon and model selection procedure.