We examined the longitudinal association between blood pressure (BP) and stroke incidence in young and middle-aged adults. BP measured during 9 examinations of the CARDIA study (Coronary Artery Risk Development in Young Adults) from 1985-1986 to 2015-2016 was used to classify participants (n=5079) according to the 2017 Hypertension Clinical Practice Guidelines. We used the highest BP obtained through the third examination (1990-1991) to define baseline BP categories; time-dependent categories (accounting for change in BP over time) were determined incorporating follow-up measurements. BP groups at ages 30 and 40 years were also defined. Stroke events were adjudicated until 2018. Mean age at baseline was 29.8 years. Stroke occurred in 100 participants. Stroke incidence (per 100 000 person-years) was higher (P<0.001) in Black (120 [95% CI, 95-149]) versus White (29 [95% CI, 18-46]) participants. After adjustment with Cox models for sociodemographic and cardiovascular risk factors, stage 2 hypertension was associated with a higher risk of stroke at baseline (hazard ratio, 3.72 [95% CI, 2.12-6.54]), as a time-dependent variable (hazard ratio, 5.84 [95% CI, 3.43-9.95]), at age 30 (hazard ratio, 4.14 [95% CI, 2.19-7.82]) and at age 40 (hazard ratio, 5.59 [95% CI, 3.35-9.31]), compared with normal BP. Elevated BP and stage 1 hypertension showed more modest increases in risk. As a continuous variable, systolic BP ≥90 mm Hg at age 40 was directly associated with stroke risk. These findings call for primordial prevention strategies to reduce population BP levels among young and middle-aged adults, particularly in Black young adults given ≈4-fold higher stroke incidence, including within values traditionally considered to be normal.