This study examined postural variabilities based on the self-perceived most comfortable postures of 12 participants (six men and six women) when sitting on three commonly used types of chairs (a stool, computer chair, and gaming chair). Participants’ global joint angles were recorded and analyzed. Of the chairs studied, the stool was not adjustable, but the computer and gaming chairs were moderately and highly adjustable, respectively. During the test, participants were encouraged to adjust the chairs until they perceived that the most comfortable posture had been reached. The results demonstrated that in a sitting position perceived to be comfortable, the participants’ postural variabilities with respect to global joint angle, calculated from five repetitions, were unexpectedly high for all three chair types, at approximately 9.4, 10.2, and 11.1° for head inclination, trunk angle, and knee angle, respectively. The average differences in range for each joint angle among the three chair types were relatively low, with all values within 3°. The result also showed that gender (p < 0.01) and chair type (p < 0.001) significantly affected trunk angle, whereas these variables did not affect head inclination or knee angle (p > 0.05). The preliminary results observed unexpectedly high variabilities in sitting posture when the participants sat at a posture that they perceived to be the most comfortable. The findings also indicated an inherent difference in comfortable sitting posture between genders; women tend to extend their trunk backward more than men. For permanent use with only an initial adjustment and memory-aided seat design, designers should minimize the loads that are borne by body parts over a prolonged period due to an unchanging sitting posture.