In this article, the author investigates expected return forecasting methodologies and their application in an asset allocation context. Although present value model–related methods are popular in practice, little is known about their performance when used for asset allocation. An intuitive and traceable carry-based method is developed by the author and tested and benchmarked against competing alternatives. The results are evaluated from different perspectives, and the obtained returns are regressed on well-known risk factors. The proposed methodology outperformed other return forecasting variants on various metrics and generated significant alphas regardless of the weight determination approach used. The methodology can be extended to further asset classes and geographic regions and provides a framework for allocating assets strategically.