Objective To explore the association between triglyceride glucose (TyG) index and TyG-body mass index (TyG-BMI) and the prevalence of metabolic associated fatty liver disease (MAFLD) in the elderly men. Methods Totally 2290 elderly men were selected from January to December in 2021 in the Second Medical Center of Chinese PLA General Hospital, and divided into MAFLD group (n=1322) and non-MAFLD group (n=968). Multivariate logistic regression was used to analyze the association between TyG index, TyG-BMI and MAFLD. The receiver operating characteristic (ROC) curve was drawn to explore the predictive value of TyG index and TyG-BMI with MAFLD in the elderly men. Results Two thousand two hundred and ninety elderly men were (74.3±10.1) years old, and an average BMI of (24.63±2.70) kg/m². BMI, γ‑glutamyl transaminase (γ‑GT), alanine aminotransferase (ALT), aspartate aminotransferase (AST), serum creatinine (Scr), thyroid stimulating hormone (TSH), free triiodothyronine (FT₃), the rate of smoking and drinking, and the prevalence of hypertension, diabetes, hyperuricemia, high triglyceride (TG), low high density lipoprotein cholesterol (HDL-C), hyperuricemia, thyroid nodules and cholelithiasis were all significantly higher in non-MAFLD group than those in MAFLD group (P<0.05), while the age of MAFLD group was lower than that of non-MAFLD group (P=0.003). Multivariate logistic regression analysis showed that the risk of MAFLD in patients of TyG quartile groups Q₂, Q₃, Q₄ was 1.667 (95%CI 1.257-2.236, P<0.001), 2.004 (95%CI 1.482-2.710, P<0.001) and 5.420 (95%CI 3.266-8.995, P<0.001) times higher than that of TyG Q₁, respectively. The risk of MAFLD in patients of TyG-BMI Q₂, Q₃, Q₄ was 2.215 (95%CI 1.549-3.167, P<0.001), 2.809 (95%CI 1.723-4.580, P<0.001) and 2.513 (95%CI 1.253-5.040, P=0.009) times higher than that of TyG-BMI Q₁, respectively. The ROC curve showed that areas under the curve (AUC) of MAFLD predicted by TyG index and TyG-BMI were 0.717 (95%CI 0.696-0.738) and 0.840 (95%CI 0.823-0.856), and the best cut-off values were 8.63 and 205.20, respectively. Moreover, the ROC curve showed that AUC of MAFLD in the elderly men without hyperlipidemia or diabetes predicted by TyG index and TyG-BMI were 0.653 (95%CI 0.622-0.684) and 0.840 (95%CI 0.818-0.862), and the best cut-off values were 8.42 and 202.66, respectively. In addition, AUC, accuracy, specificity, sensitivity, positive predictive value and negative predictive value predicted by TyG-BMI were higher than those by TyG index. Conclusions TyG index and TyG-BMI are significantly associated with MAFLD in the elderly men. Both TyG index and TyG-BMI have certain predictive value for the prevalence of MAFLD in the elderly men, and TyG-BMI may be better.