Seafarer sleepiness jeopardizes safety at sea and has been documented as a direct or contributing factor in many maritime accidents. This study investigates sleep, sleepiness, and neurobehavioral performance in a simulated 4 h on/8 h off watch system as well as the effects of a single free watch disturbance, simulating a condition of overtime work, resulting in 16 h of work in a row and a missed sleep opportunity. Thirty bridge officers (age 30 ± 6 yrs; 29 men) participated in bridge simulator trials on an identical 1-wk voyage in the North Sea and English Channel. The three watch teams started respectively with the 00-04, the 04-08, and the 08-12 watches. Participants rated their sleepiness every hour (Karolinska Sleepiness Scale [KSS]) and carried out a 5-min psychomotor vigilance test (PVT) test at the start and end of every watch. Polysomnography (PSG) was recorded during 6 watches in the first and the second half of the week. KSS was higher during the first (mean ± SD: 4.0 ± 0.2) compared with the second (3.3 ± 0.2) watch of the day (p < 0.001). In addition, it increased with hours on watch (p < 0.001), peaking at the end of watch (4.1 ± 0.2). The free watch disturbance increased KSS profoundly (p < 0.001): from 4.2 ± 0.2 to 6.5 ± 0.3. PVT reaction times were slower during the first (290 ± 6 ms) compared with the second (280 ± 6 ms) watch of the day (p < 0.001) as well as at the end of the watch (289 ± 6 ms) compared with the start (281 ± 6 ms; p = 0.001). The free watch disturbance increased reaction times (p < 0.001) from 283 ± 5 to 306 ± 7 ms. Similar effects were observed for PVT lapses. One third of all participants slept during at least one of the PSG watches. Sleep on watch was most abundant in the team working 00-04 and it increased following the free watch disturbance. This study reveals that-within a 4 h on/8 h off shift system-subjective and objective sleepiness peak during the night and early morning watches, coinciding with a time frame in which relatively many maritime accidents occur. In addition, we showed that overtime work strongly increases sleepiness. Finally, a striking amount of participants fell asleep while on duty.