Previous studies focusing on assessing the effects of remnant cholesterol (RC) and low-density lipoprotein cholesterol (LDL-C) on stroke may not consider their mutual influence. We aimed to explore the associations of RC and discordant high RC with LDL-C with stroke, ischemic stroke (IS), and hemorrhagic stroke. This prospective cohort study was conducted based on 3 cohorts of the China-PAR (Prediction for Atherosclerotic Cardiovascular Disease Risk in China) project. RC was calculated as non-high-density lipoprotein cholesterol minus LDL-C estimated by Martin/Hopkins equations. Concordant/discordant categories for RC versus LDL-C were determined based on cut-points of 130 mg/dL for LDL-C and equivalent percentile (32.50 mg/dL) for RC. Cox models were used to estimate adjusted hazard ratios and 95% CIs for incident stroke. Among 113 448 participants recruited at baseline, a total of 98 967 participants were eligible for the final analysis (mean age of 51.44 years; 40.45% were men). During 728 776.87 person-years of follow-up, 2859 stroke cases, 1811 IS cases, and 849 hemorrhagic stroke cases were observed. RC was positively associated with stroke and IS, but not hemorrhagic stroke, with adjusted hazard ratios (95% CIs) of 1.06 (1.02-1.10), 1.09 (1.04-1.13), and 0.95 (0.88-1.03) for per SD increase in RC. Compared with low LDL-C/low RC group, low LDL-C/high RC group had higher risks of stroke (adjusted hazard ratio, 1.15 [95% CI, 1.02-1.30]) and IS (1.19, 1.03-1.38), while high LDL-C/low RC group had no increased risk of stroke (1.07 [0.95-1.20]) and IS (1.09 [0.94-1.25]). Higher RC was associated with increased risks of stroke and IS but not hemorrhagic stroke. Discordantly high RC, not discordantly high LDL-C, conferred higher risks of stroke and IS. Our findings support further lowering RC by interventions to reduce residual IS risk.