Objective: To provide contemporaneous epidemiological estimates for point prevalence, birth prevalence, and cumulative diagnostic incidence of males with XLMTM in the US. Methods: Using a retrospective cohort study design, genetically confirmed XLMTM cases identified from a patient registry at Boston Children’s Hospital were tokenized and matched to cases in a large open claims database. Demographic and clinical characteristics of confirmed cases were used to develop a machine learning algorithm for XLMTM case identification. Low, moderate, and high estimates of predicted patient numbers were calculated using predictive score cut-offs. Prevalence and incidence estimates were derived from numbers of predicted and confirmed patients in each cohort. These numbers were projected to the annual US male census population to estimate XLMTM point prevalence, birth prevalence, and cumulative diagnostic incidence. Results: Of 133 tokenized patients, 78 were matched to patients in the open claims database. Point prevalence estimates were 220 (low), 489 (moderate), and 552 (high) living US cases in 2025, reflecting 5.88 (low), 13.07 (moderate), and 14.76 (high) estimated XLMTM cases/1,000,000 US males aged <18 years. The highest US birth prevalence estimates were in 2018 (low: 22.2; moderate: 43.7; high: 47.6 per 1,000,000 male births); estimates were lowest for 2021–2025. Cumulative diagnostic incidence estimates in the full US pediatric population were 325 (low) and 707 (high) for 2017–2021, and 8 (low) and 124 (high) for 2021–2025; these latter values may be underestimates, reflecting limitations of the claims-based model. Conclusions: This study is the first to estimate XLMTM birth prevalence and point prevalence based on a machine learning model. While the conservative low birth prevalence estimate in 2018 is similar to previous reports, moderate and high estimates are higher, suggesting XLMTM may be more common than previously reported. This innovative methodology could inform epidemiologic studies in other rare and ultra-rare diseases.