The fractional flow reserve (FFR) is considered to be a reliable index for the assessment of clinically relevant coronary artery stenosis. However, mismatch in assessing the severity of coronary stenosis between coronary angiography and the FFR has been pointed out. The cardiovascular intervention therapeutics (CVIT)-DEFER registry is a prospective multicenter registry study that has enrolled 3,228 consecutive patients among 3,804 patients with angiographically moderate coronary artery lesions in whom FFR analysis was clinically indicated. The demographic and angiographic parameters associated with an FFR ≤0.8 were analyzed, and the incidence of discrepancy between the angiographic severity of coronary stenosis and the FFR was assessed. Based on the visual assessment, 1,609 (42.9%) lesions were categorized as showing 50% stenosis, 1,882 lesions (50.2%) as 75% stenosis, and 257 lesions (6.9%) as 90% stenosis. Angiographic-FFR "mismatch," which was defined as visual stenosis ≥75% with FFR >0.80, was found in 43.4% of lesions, while reverse angiographic mismatch (visual stenosis <75% with FFR ≤0.8) was found in 23.2%. The independent predictors for "angiographic-FFR mismatch" were the presence of percutaneous coronary intervention (PCI) history, one-vessel disease, non-left anterior descending artery (LAD) location, non-diffuse lesion, non-ostial lesion, and non-tandem lesion. Conversely, "reverse angiographic mismatch" was independently associated with the multivessel disease, LAD location, and diffuse lesion. The FFR is not only influenced by luminal stenosis but also by coronary artery morphology and the amount of jeopardized myocardium. Angiographic-FFR mismatch is frequent in patients with moderate coronary stenosis, which suggests the clinical importance of using physiological assessment to guide PCI.