Punch speed is an important factor in the sport of boxing, and its measurement has important implications for monitoring training progression and outcomes. The aim of the current study was to establish the concurrent validity and reliability of a linear positional transducer and an accelerometer for the quantification of punch characteristics in untrained adults. Men and women aged 18–30 years with no previous boxing experience and no upper-limb musculoskeletal injuries were recruited. Participants performed 6 straight right punches; 3 at a self-determined 50% effort; and 3 at maximum effort. An accelerometer (Crossbow) and a linear positional transducer (GymAware) were used to examine peak velocity and acceleration of each punch. Validity was examined using Pearson's correlation analyses and by calculating mean bias and limits of agreement between measures from each device, whereas reliability was established using intraclass correlation coefficients (ICCs). Forty-four healthy young adults (28M and 16F; age 22.2 ± 2.9 years) participated. Moderate-to-strong positive associations were observed for both devices at 50% effort for velocity (r = 0.572–0.696) and acceleration (r = 0.867–0.921) and at maximum effort for velocity (r = 0.748–0.781) and acceleration (r = 0.897–0.946). High levels of reliability were observed with maximum punches for both devices (ICC = 0.922–0.981). Overall, moderate-strong measurement validity and reliability for punch speed was observed between the accelerometer and GymAware. Thus, the GymAware linear positional transducer is an acceptable measurement tool for the quantification of punch speed for straight punches in untrained adults.