Some researchers have recently argued that sound symbolic requirements can cause phonological alternations, and others have claimed that the relationships between sounds and meanings can be likened to input-output relationships in phonology, both suggesting that sound symbolic patterns and phonological patterns may be governed by similar mechanisms. Against this theoretical development, this paper further addresses the question regarding how similar phonological systems and sound symbolic systems are, by focusing on their counting capacity. It has been known that phonological systems can count only up two segments. To examine whether a similar sort of restriction holds in sound symbolic patterns, we experimentally addressed the question of whether three segments of the same sort can cause stronger sound symbolic images than two segments. The results of three experiments using Pokémon names demonstrate that three segments can indeed cause stronger sound symbolic meanings than two segments. The overall results suggest that there is a non-negligible difference between phonological patterns and sound symbolic patterns.