The optimal treatment of submacular hemorrhage (SMH) following neovascular age-related macular degeneration (nAMD) is controversial. This study aimed to compare visual outcomes of conservative versus active surgical treatment. Two hundred thirty-six eyes of 236 patients with SMH (≥ 1 disc diameter) were stratified into four groups: observation (n = 21); anti-vascular endothelial growth factor (VEGF) monotherapy (n = 161); non-surgical gas tamponade (n = 31); and subretinal surgery (n = 23). The primary outcome was best-corrected visual acuity (BCVA) at 12 months. The baseline BCVAs of the observation, anti-VEGF monotherapy, non-surgical gas tamponade, and subretinal surgery groups were 1.50 ± 0.70, 1.09 ± 0.70, 1.31 ± 0.83, and 1.62 ± 0.77 logarithm of minimal angle resolution (LogMAR), respectively. The mean BCVAs at 12 months were 1.39 ± 0.84, 0.90 ± 0.83, 1.35 ± 0.88, and 1.44 ± 0.91 LogMAR, respectively. After adjusting for age, baseline BCVA, SMH size, and the number of intravitreal anti-VEGF injections before SMH, the mean BCVA showed no significant difference among treatments at 12 months (P = 0.204). The anti-VEGF monotherapy group showed better mean BCVA significantly at 3 months (P < 0.001). Only baseline BCVA was associated with VA gain at 12 months (Odds ratio = 3.53, P < 0.001). This study demonstrated that there was no difference in 12 month visual outcomes among treatments and a better early visual outcome can be expected with anti-VEGF monotherapy.