Molecular phylogenetics elucidates the evolution and divergence of green plants by analyzing sequence data from diverse sources. Notably, phylogenetic reconstruction based on mitochondrial genes often shows incongruence with results from nuclear and chloroplast genes. Although the uniparental inheritance and conservatively retained protein-coding genes of mitochondrial genomes inherently exclude certain confounding factors that affect phylogenetic reconstruction-such as hybridization and gene loss-the use of mitochondrial genomes for phylogeny and divergence-time estimation has remained limited. Here, we assembled a comprehensive dataset of 565 mitochondrial genomes representing all major lineages of green plants. Applying multiple partitions and phylogenetic models, our mitochondrial-based phylogenies support paraphyly in both bryophytes and charophytes, place hornworts (Anthocerotaceae) as sister to all tracheophytes, and recover stoneworts (Charophyceae) as sister to land plants. We systematically evaluated the influence of factors in mitochondrial coding sequences, including GC-content heterogeneity and codon-usage bias. Furthermore, by rigorously testing seven dating strategies, we assessed the impact of confounding elements affecting divergence-time estimates, such as fossil calibration number and prior settings, as well as rate heterogeneity among sites and across lineages. Our dating analyses support a Neoproterozoic origin (crown age) of land plants and a Triassic origin of angiosperms, consistent with nuclear evidence. In conclusion, we emphasize the importance of exploring alternative partitioning strategies and addressing among-lineage heterogeneity in both phylogenetic and dating analyses, with extended sampling and careful data pruning to minimize systematic error in phylogenetic inference.