Background/Aim: Patients with fecal incontinence (FI) have lower anal resting (MRP) and squeeze (MSP) pressure and an impaired sensitivity compared to healthy people. However, whether anorectal manometry (ARM) can separate precisely between health and disease is discussed controversially. The aim was to evaluate the accuracy of ARM in a huge cohort of patients and controls. Methods: ARM was obtained in 144 controls and in 559 FI patients. MRP, MSP, and balloon volume at first perception (BVP) and urge sensation (BVU) were determined. Receiver operating curve analysis was used to determine optimal cut-offs and sensitivity, specificity and accuracy calculated. Results: FI patients showed lower MRP, MSP, BVU (p < 0.001) and a higher BVP (p = 0.007). Deterioration of the ARM parameter increased with FI severity. ARM demonstrated an excellent sensitivity (91.4%) and accuracy (85.8%), but only a moderate specificity (62.5%). The sensitivity of ARM rose with FI severity. The pressure data showed higher sensitivity and accuracy than the sensory data despite comparable specificity. Conclusions: Sensitivity and accuracy of single ARM parameters is only moderate for the pressure data and poor for the sensory data. In contrast, ARM demonstrated an excellent sensitivity, a moderate specificity, and a convincing accuracy justifying its use in clinical routine.