Little research has been conducted on the effect of beef feedlot manure amendments on runoff hydrology and soil loss during the transition between continuous and legacy application phases. We conducted a 6 yr (2013–2018) study and utilized surface (0–5 cm) soil collected from a long-term (since 1998) field experiment on a clay loam soil in southern Alberta, Canada. The treatments were stockpiled or composted feedlot manure with straw (ST) or wood-chip (WD) bedding applied at 13, 39, and 77 Mg ha^-1 (dry weight). There was also an unamended control and inorganic fertilized treatment (IN). Disturbed surface soil was collected from all treatments after 15 (C15), 16 (C16), and 17 (C17) continual annual applications (2013–2015), and then 1 (L1), 2 (L2), and 3 (L3) yr (2016–2018) after manure applications were discontinued in 2015. The soil was packed into runoff trays, and a Guelph rainfall simulator (70 mm h^-1) was used to generate 30 min of runoff. The time to runoff, total runoff depth, and flow-weighted mean concentrations (FWMCs) and mass loads of total suspended solids (TSS) were determined. Mean values of runoff variables were significantly greater for ST than WD in certain years for time to runoff (by 28%–127%), runoff depth (44%), TSS FWMCs (58%–137%), and FWMC loads (28%–230%). Mean TSS was generally significantly greater at the two lower than highest manure rates for FWMC (48%–135%) and loads (21%–253%). Overall, choice of bedding material showed the most potential to maximize infiltration and reduce water and sediment loss.