Core Ideas: Miscanthus and switchgrass are often touted as biomass crop choices, but have rarely been compared side‐by‐side. Miscanthus was slower to establish than switchgrass cultivars, but out‐yielded switchgrass in the latter 3 yr of the trial. Phosphorus was the soil nutrient most likely to be drawn down to levels requiring fertilization, after N. Comparative performance tests of perennial grasses for biomass yield, quality, and soil nutrient removal are needed to guide decisions toward meeting European Union targets for renewable energy production. We compared hybrid miscanthus (Miscanthus × giganteus Greef and Deuter ex Hodkinson and Renvoize) and switchgrass (Panicum virgatum L.) cultivars Cave‐in‐Rock, (upland type), Alamo, and Kanlow (lowland types) for biomass yield and changes in soil macronutrient levels and removal rates in a humid Spanish environment. Soil and plant nutrient and C levels were measured after each annual biomass harvest for 4 yr. Plant nutrient concentrations were multiplied by biomass yield to express nutrient removal. Yield ranking (4‐yr mean) was miscanthus (17.6 Mg ha⁻¹) > Kanlow (13.3 Mg ha⁻¹) = Alamo (13.0 Mg ha⁻¹) > Cave‐in‐Rock (7.7 Mg ha⁻¹). Miscanthus biomass yield peaked in the third year at 27.0 Mg ha⁻¹. The high yield of miscanthus together with its relatively low macronutrient concentrations and intermediate removal rates, indicate its advantages over switchgrass as a biomass crop choice. Alamo and Kanlow usually removed more macronutrients than Cave‐in‐Rock, suggesting a greater long‐term fertilizer requirement for the lowland types. Soil C stocks increased by a mean of 2850 kg ha⁻¹ over 4 yr at 0‐ to 20‐cm depth. Phosphorus was the macronutrient most likely to become deficient after repeated harvests. Miscanthus and lowland switchgrass cultivars performed well in the Atlantic maritime region of Spain, therefore the choice of crop would rely on economics of establishment and long‐term stand maintenance.