ObjectiveThis study aims to elucidate the dose‒response relationship between 24-h activity behaviors and body fat percentage (BFP) in Chinese preschool children using a compositional isotemporal substitution model (ISM).MethodsIn a cross-sectional design, 881 children aged 3–6 from urban and rural areas of Jiangxi Province were sampled. Activity behaviors, including sedentary behavior (SB), low-intensity physical activity (LPA), and moderate- to high-intensity physical activity (MVPA), were measured using accelerometers. Sleep patterns were assessed through questionnaires, and BFP was determined by bioelectrical impedance analysis (BIA). The study employed compositional data analysis (CoDA) and ISM to estimate the impact of reallocating durations of different activity behaviors on BFP.ResultsHigher BFP was found in urban vs. rural children, decreasing with age. Overweight and obesity rates were 10.6% and 7.6%, respectively, above national averages. MVPA and LPA were negatively correlated with BFP, while SB was positively correlated.A 30-min MVPA reduction significantly increased zBFR, particularly in overweight children. Gender-specific nuances revealed that boys' MVPA negatively influenced zBFP (β = -0.155), P < 0.05), while girls' SB positively impacted zBFP (β = 0.636, P < 0.01). Isotemporal simulations emphasized amplified effects in overweight children, with boys' zBFR rising rapidly when MVPA was substituted and girls displaying a notable substitution effect between SB and LPA.ConclusionBFP is closely linked to 24-h activity behaviors, notably in overweight and obese preschoolers. ISM identified MVPA as a critical influencer, with a 30-min reduction substantially increasing BFP. Gender disparities were evident, implicating MVPA in boys and LPA and SB in girls.