Objectives Despite limited supporting data, hospitals continue to apply ambulance diversion (AD). Thus, we examined the impact of three different diversion policies on diversion hours, transport time (TT; leaving scene to arrival at the hospital), and ambulance patient offload time (APOT; arrival at the hospital to patient turnover to hospital staff) for 9-1-1 transports in a 22-hospital county Emergency Medical Services (EMS) system. Methods This retrospective study evaluated metrics during periods of three AD policies, each 27 days long: hospital-initiated (Period 1), complete suspension (Period 2), and County EMS-initiated (Period 3). We described the median transports and diversion hours, and compared the daily average and daily 90th percentile TT and APOT during the three study periods. Results Over the study period, there were 50,992 total transports in the county; Period 3 had fewer median transports per day than Period 1 (581 vs 623, p < 0.001), while Period 2 was similar to Period 1 (606 vs 623, p = 0.108). Median average daily diversion hours decreased from 98.1 h during Period 1 to zero hours during both Periods 2 (p < 0.001) and 3 (p < 0.001). Median daily average TT decreased from 18.3 min in Period 1 to 16.9 min in both Periods 2 (p < 0.001) and 3 (p < 0.001). Median daily 90th percentile TT showed a similar decrease from 30.2 min in Period 1 to 27.5 in Period 2 (p < 0.001), and to 28.1 in Period 3 (p = 0.001). Median average daily APOT was 26.0 min during Period 1, similar at 25.2 min during Period 2 (p = 0.826) and decreased to 20.4 min during Period 3 (p < 0.001). The median daily 90th percentile APOT was 53.9 min during Period 1, similar at 51.7 min during Period 2 (p = 0.553) and decreased to 40.3 min during Period 3 (p < 0.001). Conclusions Compared to hospital-initiated AD, enacting no AD or County EMS-initiated AD was associated with less diversion time; TT and APOT showed statistically significant improvement without hospital-initiated AD but were of unclear clinical significance. EMS-initiated AD was difficult to interpret as that period had significantly fewer transports. EMS systems should consider these findings when developing strategies to improve TT, APOT, and system use of diversion.