A year ago, I explained that prediction accuracy influences optimal portfolio construction, with better prediction accuracy resulting in more concentrated portfolios, higher turnover, higher position limits and higher returns and information ratios.

An investor with perfect foresight would logically construct a highly-concentrated, high-returning portfolio, whereas an investor with no foresight would build a highly-diversified portfolio with no active bets. In between these two extremes, portfolio characteristics should reflect the predictability of the underlying market. Based on q...