Objectives: As diagnosis and treatment guidelines for bone sarcomas continue updating, it is important to examine whether, when, and which kinds of patients have had a survival improvement over the last four decades. Methods: This cohort study included 9178 patients with primary bone and joint sarcomas from 1 January 1980 to 31 December 2018 using data from Surveillance, Epidemiology and End Results (SEER)-9 Registries. The follow-up period was extended to November 2020. Patients were divided by decade into four time periods: 1980–1989, 1990–1999, 2000–2009, and 2010–2018. The primary endpoint was bone sarcomas-specific mortality (CSM). The 5-year bone sarcomas-specific survival (CSS) rate was determined stratified by demographic, neoplastic, temporal, economic, and geographic categories. The associations between time periods and CSM were examined using a multivariable Cox regression model, with reported hazard ratio (HR) and 95% confidence interval (CI). Results: The 5-year CSS rate for bone sarcomas was 58.7%, 69.9%, 71.0%, and 69.2%, in the 1980s, 1990s, 2000s, and 2010s, respectively. Older age, male gender, tumor sites at pelvic bones, sacrum, coccyx and associated joints, as well as vertebral column, osteosarcoma and Ewing tumor, and residence in non-metropolitan areas were independently associated with higher CSM risk. After adjusting for the covariates above, patients in the 1990s (HR = 0.74, 95% CI = 0.68–0.82), 2000s (HR = 0.71, 95% CI = 0.65–0.78), and 2010s (HR = 0.68, 95% CI = 0.62–0.76) had significantly lower CSM risks than patients in the 1980s. However, patients in the 2000s and 2010s did not have lower CSM risks than those in the 1990s (both p > 0.05). Conclusions: Although bone sarcomas survival has significantly improved since 1990, it almost halted over the next three decades. Bone sarcomas survival should improve over time, similar to common cancers. New diagnostic and therapeutic strategies such as emerging immune and targeted agents are warranted to overcome this survival stalemate.