Cover crops are increasingly being used for weed management, and planting them as diverse mixtures has become an increasingly popular strategy for their implementation. While ecological theory suggests that cover crop mixtures should be more weed suppressive than cover crop monocultures, few experiments have explicitly tested this for more than a single temporal niche. We assessed the effects of cover crop mixtures (5- or 6-species and 14-species mixtures) and monocultures on weed abundance (weed biomass) and weed suppression at the time of cover crop termination. Separate experiments were conducted in Madbury, NH, from 2014 to 2017 for each of three temporal cover-cropping niches: summer (spring planting–summer termination), fall (summer planting–fall termination), and spring (fall planting–subsequent spring termination). Regardless of temporal niche, mixtures were never more weed suppressive than the most weed-suppressive cover crop grown as a monoculture, and the more diverse mixture (14 species) never outperformed the less diverse mixture. Mean weed-suppression levels of the best-performing monocultures in each temporal niche ranged from 97% to 98% for buckwheat (Fagopyrum esculentum Moench) in the summer niche and forage radish (Raphanus sativus L. var. niger J. Kern.) in the fall niche, and 83% to 100% for triticale (×Triticosecale Wittm. ex A. Camus [Secale × Triticum]) in the winter–spring niche. In comparison, weed-suppression levels for the mixtures ranged from 66% to 97%, 70% to 90%, and 67% to 99% in the summer, fall, and spring niches, respectively. Stability of weed suppression, measured as the coefficient of variation, was two to six times greater in the best-performing monoculture compared with the most stable mixture, depending on the temporal niche. Results of this study suggest that when weed suppression is the sole objective, farmers are more likely to achieve better results planting the most weed-suppressive cover crop as a monoculture than a mixture.