OBJECTIVE It is unclear whether endoscopic mucosal resection (EMR) or hot snare resection is better for resecting large nonpedunculated polyps. The aim of this study was to determine a cutoff size of nonpedunculated neoplastic colorectal polyps at which the risk of incomplete resection differed between EMR and hot snare resection. METHODS Patients with nonpedunculated neoplastic polyps (10 - 25 mm in diameter) were randomly assigned to undergo endoscopic resection using EMR (52 patients with 63 polyps) or hot snare resection (52 patients with 62 polyps). EMR included submucosal injection of saline before resection. The primary outcome measure was the proportion with complete polyp resection determined by histopathology. The secondary outcome was total procedure time. RESULTS Patient characteristics were similar between groups. EMR achieved complete resection more frequently than hot snare resection (89 % vs. 73 %; P = 0.02), particularly for polyps ≥ 20 mm (75 % [9 /12] vs. 18 % [2 /11]; P = 0.006). A complete resection rate of > 90 % was achieved for polyps of size < 19 mm with EMR, and for polyps of size ≤ 14 mm with hot snare resection. In multivariate analysis, incomplete resection was associated with hot snare resection (odds ratio [OR] 2.8, 95 % confidence interval (95 %CI) 1.0 - 8.3; P = 0.04) and polyp size ≥ 15 mm (OR 4.0, 95 %CI 1.3 - 14; P = 0.01). Total procedure time was shorter with hot snare resection than with EMR (mean 14.8 min vs. 17.2 min; P < 0.001). CONCLUSIONS EMR and hot snare resection appear to achieve similar complete resection rates for polyps up to 14 mm; however, EMR may be superior for larger polyps, particularly for those ≥ 20 mm.Registered at Clinicaltrials.gov: NCT 01950117.