The commercial use of high-forage-yielding tetraploid red clover (Trifolium pratense L.) is limited due to its low seed yield, but, surprisingly, the seed set potential of individual florets of diploid and tetraploid red clover cultivars is rarely studied. Hand pollination experiments were conducted in two diploid and one tetraploid red clover cultivars to investigate (1) the correlation between seed set and the number of pollinated florets per flower head, and (2) how seed set is influenced by the different sources and amounts of pollen deposited on the stigmas. The results demonstrate that an increase in the number of pollinated florets (from 0 to 80 florets) increased the number of harvested seeds. However, the seed set rate increased only up to approximately 20 pollinated florets. No significant difference in seed set was found between the florets pollinated with a low (on average 43 pollen grains) or high (on average 84 pollen grains) number of pollen grains, except for the red clover cultivar, which has the highest pollen viability. We suggest that the amount of viable pollen deposited on the stigma influences seed set in red clover. Furthermore, results showed that pollen from cultivars with different ploidy levels might disturb seed set. In conclusion, results indicate the importance of pollen quality (pollen viability and pollen source) to seed set of diploid and tetraploid red clover. We found hand pollination an excellent method/tool for investigating the underlying causes of low seed set in red clover, provided that at least 20 florets per flower head are pollinated.