Background and Objectives: Thoracic surgery is associated with postoperative pain and delayed recovery. Erector spinae plane block (ESPB) is increasingly used in multimodal analgesia, but its impact on postoperative recovery remains unclear. This study compared recovery outcomes between patients receiving ESPB and those without regional block. Materials and Methods: This single-center retrospective observational cohort study analyzed 308 patients undergoing thoracic surgery between December 2021 and July 2025. Patients underwent video-assisted thoracoscopic surgery (VATS) or thoracotomy. The primary outcome was hospital length of stay; secondary outcomes included operation duration, intensive care unit (ICU) stay, and chest drain-related measures. Multivariable linear regression adjusted for age, sex, and surgical approach. An exploratory interaction between ESPB and surgical approach was assessed. Results: Baseline demographic characteristics were comparable between groups. OR time was longer in the Block group (169.05 ± 131.74 vs. 129.85 ± 100.22 min, p = 0.005). Unadjusted hospital length of stay distributions differed significantly (Mann–Whitney U test, p = 0.011), but ESPB was not significantly associated with hospitalization after adjustment (β = 0.003 days, 95% CI: −0.532 to 0.537, p = 0.992; sensitivity model p = 0.624). Thoracotomy was associated with longer hospitalization than VATS (β = 1.960 days, p < 0.001). The exploratory ESPB × surgical approach interaction was significant (p = 0.021), but the adjusted association within VATS was not (p = 0.207). ICU stay and chest drain-related outcomes did not differ significantly. Conclusions: ESPB was not significantly associated with shorter hospitalization after adjustment for age, sex, and surgical approach, despite differences in unadjusted comparisons. Exploratory findings suggested variation by surgical approach. Prospective studies are needed to clarify the relationship between ESPB and postoperative recovery.