Background There is no consensus regarding the management of carotid free-floating thrombi in patients with acute ischemic stroke and transient ischemic attack. Aims This systematic review aims to (1) describe the current imaging methods for diagnosis of carotid free-floating thrombi and its associated risk factors, (2) estimate the proportion of carotid free-floating thrombi diagnosed in stroke patients, (3) estimate the proportion of carotid free-floating thrombi patients treated medically and surgically, and (4) evaluate 30-day outcomes. Methods We searched MEDLINE, EMBASE, and manually for references reporting carotid free-floating thrombi from 1960 until June 2017. We estimated the frequency of carotid free-floating thrombi and evaluated 30-day outcomes using Cox regression. We defined the timing of surgical intervention as early (less than 72 h) and delayed (more or equal to 72 h). Summary of review We retrieved 525 carotid free-floating thrombi cases from 58 case series and 83 case reports. Carotid free-floating thrombi were present in 1.53% of stroke patients. Carotid free-floating thrombi diagnosis was made by digital subtraction angiography (38.1%), carotid duplex ultrasound (29.5%), and computed tomography angiography (29.5%). The 30-day risk of transient ischemic attack, silent brain ischemia, any stroke or death was 17.1%. In multivariate analyses, there were no differences in outcome for any anticoagulation regime or timing of revascularization procedure. Conclusions The diagnosis of carotid free-floating thrombi is more common due to the increased use of early noninvasive vascular imaging in transient ischemic attack and stroke. It poses a high short-term risk of stroke and death, but there is as yet no established treatment. The low quality of evidence in the carotid free-floating thrombi literature limits the interpretation of our results and warrants a large-scale prospective cohort study in carotid free-floating thrombi.