The 72-km-long Teton normal fault bounds the eastern base of the Teton Range in northwestern Wyoming, USA. Although geomorphic surfaces along the fault record latest Pleistocene to Holocene fault movement, the postglacial earthquake history of the fault has remained enigmatic. We excavated a paleoseismic trench at the Buffalo Bowl site along the southernmost part of the fault to determine its Holocene rupture history and slip rate. At the site, ∼6.3 m of displacement postdates an early Holocene (ca. 10.5 ka) alluvial-fan surface. We document evidence of three surface-faulting earthquakes based on packages of scarp-derived colluvium that postdate the alluvial-fan units. Bayesian modeling of radiocarbon and luminescence ages yields earthquake times of ca. 9.9 ka, ca. 7.1 ka, and ca. 4.6 ka, forming the longest, most complete paleoseismic record of the Teton fault. We integrate these data with a displaced deglacial surface 4 km NE at Granite Canyon to calculate a postglacial to mid-Holocene (14.4–4.6 ka) slip rate of ∼1.1 mm/yr. Our analysis also suggests that the postglacial to early Holocene (14.4–9.9 ka) slip rate exceeds the Holocene (9.9–4.6 ka) rate by a factor of ∼2 (maximum of 3); however, a uniform rate for the fault is possible considering the 95% slip-rate errors. The ∼5 k.y. elapsed time since the last rupture of the southernmost Teton fault implies a current slip deficit of ∼4–5 m, which is possibly explained by spatially/temporally incomplete paleoseismic data, irregular earthquake recurrence, and/or variable per-event displacement. Our study emphasizes the importance of minimizing slip-rate uncertainties by integrating paleoseismic and geomorphic data sets and capturing multiple earthquake cycles.