Technical Abstract: Molecular markers in perennial forages are becoming ubiquitous for marker assisted selection (MAS). Nevertheless, widespread implementation of MAS in perennial forage breeding programs for highly quantitative traits has not yet occurred. A primary reason is likely the cost associated with genotyping plants using molecular markers. Compared to animal breeding, as well as plant breeding systems involving inbred line development, phenotyping costs in outbred forage species are comparatively much lower. Therefore, many MAS schemes are usually not cost competitive, especially in cases where many molecular markers are required. Presented are cost-effective MAS strategies that are immediately implementable in most diploid or polyploidy perennial forage breeding programs. Breeding methods developed during the pre-MAS era maximize the ratio between additive variance and the square root of the phenotypic variance (i.e. hsA). With the advent of molecular markers, many new breeding methods based primarily on correlated selection responses between molecular markers and quantitative traits have been proposed (i.e. rsA). The MAS strategies presented here use molecular markers to improve hsA by enhancing traditional breeding methods. A selection strategy pyramid is envisioned with traditional maternal halfsib selection serving as the base followed by marker assisted paternity selection. With no additional cost within maternal and paternal maximum linkage, disequilibrium based MAS strategies can be added. Finally, as the total cost to genotype one individual decreases, residual linkage disequilibrium based MAS strategies can be added to the above mentioned strategies or at some price point supplant these strategies. Using this strategy, the greatest amount of information per molecular marker will be obtained from an initial set of markers. However, as the number of markers utilized is increased, the amount of information per marker at some point will decrease. Furthermore, it is unnecessary to have an unlimited number of molecular markers for selection among a smaller number of genotypes since only a certain amount of molecular marker information will be necessary to rank selection units, with less information needed for accurate ranking among fewer genotypes. Therefore, the number of markers utilized in a selection strategy should be optimized so the maximum amount of resources can be applied to increase the number of selection units under evaluation. Even under the perfect correlated selection response scheme (i.e. ksA), selection intensity and the amount of additive variance present are still paramount drivers of selection gain.