Until recently, the cyclopolymerization (CP) of 1,6-heptadiyne derivatives using Grubbs catalysts had been unsuccessful, leading to the misbelief that these catalysts were inactive in these circumstances. However, a recent breakthrough has changed this previous perspective of CP, where a successful living CP was reported using a third-generation Grubbs catalyst with the aid of weakly coordinating ligands. Although it became clear that weakly coordinating ligands greatly enhanced the efficiency of CP by suppressing the decomposition of the propagating carbene, it was still unclear as to what was actually occurring during the previous attempts at CP using ligand-free conditions, especially in the case of the Hoveyda-Grubbs catalyst. Here, we have found that second-generation Grubbs or Hoveyda-Grubbs catalysts in dichloromethane (DCM) formed predominantly side products, i.e., dimers and trimers of 1,6-heptadiyne derivatives, instead of producing the desired conjugated polymers. Further mechanistic studies disclosed that [2 + 2 + 2] cycloaddition reactions by the decomposed Grubbs catalyst were responsible for these side products, not the commonly presumed olefin metathesis pathway. Furthermore, a control experiment revealed that pyridine not only stabilized the propagating carbene but also suppressed the dimer formation by poisoning the newly generated catalytic species that would have promoted [2 + 2 + 2] cycloaddition. This observation explained why the third-generation Grubbs catalyst successfully and selectively cyclopolymerized 1,6-heptadiyne monomers. Another significant observation was that depending on the nature of the substituents of the 1,6-heptadiynes, different ratios of polymers and side-products were obtained as a result of competition between CP and cycloaddition. Monomers containing more coordinating substituents favored the undesired cycloaddition products owing to slower polymerization and faster decomposition of the carbene, while weakly chelating monomers strongly favored CP. Finally, with this new mechanistic understanding of the factors that contribute to CP propagation and decomposition of the Grubbs catalysts, we could maximize the efficiency of CP by modifying the monomer structure, lowering the reaction temperature, or adding stabilizing ligands. This report demonstrates a successful result of how mechanistic investigation has turned a previously unattainable polymerization into an efficient one.