Background Contradictory data have been reported about the association between testosterone levels and the levels of low‐density lipoprotein cholesterol (LDL). Objectives The aim of this study was to elucidate the association between testosterone and LDL levels. Materials and methods A cross‐sectional study was conducted that included 7268 men who had participated in a health examination. Men who took agents that influence serum lipid profiles within the previous 6 months were excluded. A full metabolic work‐up and serum testosterone level checks were performed. The main outcome measures included the testosterone level and testosterone <3.5 ng/ml prevalence of each decile of LDL and their polynomial trendlines and the odds ratio (OR) of testosterone <3.5 ng/ml according to the LDL level. Results The polynomial trendline suggests the inverse association of testosterone with the deciles of triglycerides, the positive association of testosterone with the deciles of high‐density lipoprotein cholesterol and the inverted U‐shaped association between testosterone and the deciles of LDL. The polynomial trendline suggests a U‐shaped curvilinear relationship between the deciles of LDL and the prevalence of testosterone <3.5 ng/ml. The adjusted ORs of testosterone <3.5 ng/ml in men in the lowest and highest deciles were significantly higher than those of men in the 10th–90th deciles of LDL (OR [95% CI], < 10th LDL: 1.4[1.2–1.8]; 10th – 90th LDL: 1.0 [reference]; >90th LDL: 1.3[1.0–1.6]), which reinforces the U‐shaped curvilinear relationship between LDL levels and the prevalence of testosterone <3.5 ng/ml. Discussion and conclusion Further research is needed to elucidate the association of very low or very high LDL concentrations with circulating testosterone in men.