Collateral perfusion has a major effect on clinical outcome in carotid steno-occlusive disease. Arterial spin labeling (ASL) may provide a biomarker of collateral perfusion by measuring arterial transit time (ATT) in addition to cerebral blood flow (CBF). However, the concomitant measurement of ATT and CBF is not feasible for clinical applications. As an alternative, ATT can be estimated from single post-labeling delay ASL image using the spatial coefficient of variance (CoV). In this study, we investigate whether the spatial CoV lateralization through collateral perfusion can predict the side of carotid occlusion as a proof-of-principle. In addition to spatial CoV, we also investigated whether temporal variance (tVAR) of the ASL signal could be used as a predictor of occlusion versus stenosis.