The Hellenic subduction zone displays well-defined temporal and spatial variations in subduction rate and offers an excellent natural laboratory for studying the interaction among slab buoyancy, subduction rate, and tectonic deformation. In space, the active Hellenic subduction front is dextrally offset by 100–120 km across the Kephalonia Transform Zone, coinciding with the junction of a slowly subducting Adriatic continental lithosphere in the north (5–10 mm/yr) and a rapidly subducting Ionian oceanic lithosphere in the south (∼35 mm/yr). Subduction rates can be shown to have decreased from late Eocene time onward, reaching 5–12 mm/yr by late Miocene time, before increasing again along the southern portion of the subduction system. Geodynamic modeling demonstrates that the differing rates of subduction and the resultant trench offset arise naturally from subduction of oceanic (Pindos) lithosphere until late Eocene time, followed by subduction of a broad tract of continental or transitional lithosphere (Hellenic external carbonate platform) and then by Miocene entry of high-density oceanic (Ionian) lithosphere into the southern Hellenic trench. Model results yield an initiation age for the Kephalonia Transform of 6–8 Ma, in good agreement with observations. Consistency between geodynamic model results and geologic observations suggest that the middle Miocene and younger deformation of the Hellenic upper plate, including formation of the Central Hellenic Shear Zone, can be quantitatively understood as the result of spatial variations in the buoyancy of the subducting slab. Using this assumption, we make late Eocene, middle Miocene, and Pliocene reconstructions of the Hellenic system that include quantitative constraints from subduction modeling and geologic constraints on the timing and mode of upper plate deformation.