Objectives/Hypothesis Use of the Muller maneuver (MM) in the evaluation of patients with obstructive sleep apnea is controversial. One criticism of this test is that it is somewhat subjective. Our objective is to explore the reliability of this technique and its association with sleep‐disordered breathing.Study Design Prospective study performed in an academic tertiary care center.Methods An analysis of MM scores from 180 consecutive patients obtained independently by two examiners was completed. These scores were compared with each other and with the apnea‐hypopnea index (AHI) obtained from polysomnographic studies.Results Collapse of the soft palate (PAL), lateral pharyngeal wall (LPW), and base of the tongue (BOT) was rated on a five‐point scale (0–4). The mean scores determined by the faculty examiner were 2.47, 2.06, and 1.58, respectively; the mean scores determined by the resident examiner were 2.34, 2.25, and 1.48, respectively. The scores of the two examiners correlated to within ±1 unit 83.9% of the time at the PAL, 91.1% at the LPW, and 85.0% at the BOT. The degree of correlation was not influenced by year of training of the resident. When the AHI was converted to a four‐point scale based on severity, the score correlated within ±1 of the average MM score 72.1% of the time.Conclusions Despite the subjective nature of the MM, the five‐point scale can be used by independent examiners to achieve an evaluation of the upper airway that is reproducible. The preoperative severity of sleep‐disordered breathing based on the AHI is moderately correlated with the MM score.