Soft tissue pelvic sarcomas are rare mesenchymal malignancies, representing 5% of all soft tissue sarcomas [1]. The most common histologic subtypes of soft tissue pelvic sarcoma are liposarcoma and leiomyosarcoma. They account for 30-35% of all soft tissue sarcomas 1, 2. Frequently, these tumors are included in studies investigating sarcomas at other anatomical sites. However, soft tissue pelvic sarcomas have different clinical presentations, characteristics and patterns of recurrence and survival compared to other soft tissue sarcomas 3, 4. Moreover, soft tissue pelvic sarcoma survival outcomes data are mostly heterogeneous and come from small and historical cohorts 5, 6. To date, no consensus exists regarding what is the ideal tumor size cut-off in surgically treated non-metastatic soft tissue pelvic liposarcoma and leiomyosarcoma. Moreover, some studies suggested that historical tumor size categories may be poorly meaningful since, tumor size at time of diagnosis is often larger than traditional staging parameters 7, 8. In addition, no contemporary study formally tested for optimal tumor size cut-offs using the strictest statistical methodology that included multivariable analyses and AUC testing. Specifically, we tested whether an ideal tumor size cut-off, identified using a minimum p-value approach 9, 10, 11, 12, could result in improvement in cancer-specific survival (CSS) prediction. Additionally, other previously recommended tumor size cut-offs (5, 10 and 15 cm) were also tested 6, 13, 14, 15. We addressed these knowledge gaps in the current study, relying on the Surveillance, Epidemiology, and End Results (SEER) database (2004-2019) [16].