Abstract. Rips currents and other surf hazards are an emerging public health issue globally. Lifeguards, warning flags and signs are important and to varying degrees they are effective strategies to minimize risk. In the United States and other jurisdictions around the world, lifeguards use coloured flags (green, yellow and red) to indicate whether the danger posed by the surf and rip hazard is low, moderate, or high respectively. The choice of flag depends on the lifeguard monitoring the changing surf conditions along the beach and over the course of the day using both regional surf forecasts and careful observation. There is a potential that the chosen flag does not accurately reflect the potential risk, which may increase the potential for rescues or drownings. In this study, machine learning used to determine the potential for error in the flags used at Pensacola Beach, and the impact of that error on the number of rescues. A decision tree analysis suggests that the wrong flag was flown on ~ 35 % of days between 2004 and 2008 (n = 396/1125), and that those differences account for only 17 % of all rescue days and ~ 60 % of the total number of rescues. Further analysis reveals that the largest number of rescue days and total number of rescues is associated with days where the flag deployed over-estimated the surf and hazard risk, such as a red or yellow flag flying when the model would suggest a green flag would be more appropriate based on the wind and wave forcing. Regardless whether this is a result of the lifeguards being overly cautious or the rip and surf hazard is associated with weak rips forced by a transverse-bar and rip morphology, the results suggest that beach users are discounting the lifeguard warnings if it isn't consistent with how they perceive the surf hazard. Results suggest that machine learning techniques have the potential to support lifeguard and thereby reduce the number of rescues and drownings.