Purpose: To determine the discriminating ability of some parameters provided by Humphrey Matrix perimetry for early glaucoma detection. Methods: A prospective cross-sectional study was performed. Sixty-five primary open-angle glaucoma patients with early-stage visual field defects on standard automated perimetry (mean deviation = –1.98 ± 1.93 dB) and 56 healthy subjects were included. All subjects performed the Humphrey Matrix perimetry with a threshold 30-2 strategy. The receiver operating characteristic (ROC) curve was constructed for each parameter and calculated the area under the ROC curve (AUC) to seek the best discriminating parameter for early glaucoma detection. Results: The AUCs of Humphrey Matrix perimetry with the mean deviation, pattern standard deviation, Glaucoma Hemifield Test, number of points that have a p < 5% in pattern deviation plot (PDP) and number of points that have a p < 1% in PDP were 0.795, 0.808, 0.689, 0.985 and 0.946, respectively. Conclusion: Humphrey Matrix perimetry allowed accurate discrimination between normal and early glaucomatous eyes. The number of points that had a p < 5% in PDP was the best discriminating parameter for early glaucoma detection using Humphrey Matrix perimetry.