Background: The association between smoking and blood pressure (BP) has been explored extensively, yet the results remain inconclusive. Using real-world evidence of a large Chinese population, we examine the effect of smoking on BP levels. Methods: We utilize half a million adults from the China Kadoorie Biobank (CKB) study with baseline sampling collected between 2004 and 2008. Multivariable linear regression analyses are used to estimate linear regression coefficients of smoking for systolic blood pressure (SBP) and diastolic blood pressure (DBP). Results: 459,815 participants (180,236 males and 279,579 females) are included in the analysis. Regular smoking is significantly associated with lower SBP (−0.57 mm Hg, p < 0.001) and DBP (−0.35 mm Hg, p < 0.001) when compared with non-smoking in men. Additionally, SBP and DBP decrease significantly among all groups of different smoking status in women (p < 0.001). Additionally, pack-years show negative associations with SBP and DBP in both men and women. Further analysis shows the interaction of smoking and alcohol consumption is associated with an increase of SBP and DBP (men: 2.38 mm Hg and 0.89 mm Hg; women: 5.21 mm Hg and 2.62 mm Hg) among co-regular smokers and regular drinkers when compared with regular smokers who are not exposed to alcohol consumption. Conclusions: A negative association between smoking and BP is observed. However, the interaction between smoking and alcohol consumption is associated with BP increase. The findings suggest the importance of considering smoking and alcohol consumption in BP control in addition to antihypertensive treatment in clinical and public health practice.