Background “EMS provider judgment” was recently added as a field triage criterion to the national guidelines, yet its predictive value and real-world application remain unclear. We examine the use and independent predictive value of EMS provider judgment in identifying seriously injured persons. Methods We analyzed a population-based retrospective cohort, supplemented by qualitative analysis, of injured children and adults evaluated and transported by 47 EMS agencies to 94 hospitals in 5 regions across the Western U.S. from 2006–2008. We used logistic regression models to evaluate the independent predictive value of EMS provider judgment for ISS ≥ 16. EMS narratives were analyzed using qualitative methods to assess and compare common themes for each step in the triage algorithm, plus EMS provider judgment. Results 213,869 injured patients were evaluated and transported by EMS over the 3-year period, of whom 41,191 (19.3%) met at least one of the field triage criteria. EMS provider judgment was the most commonly used triage criterion (40.0% of all triage-positive patients; sole criterion in 21.4%). After accounting for other triage criteria and confounders, the adjusted odds ratio (OR) of ISS ≥ 16 for EMS provider judgment was 1.23 (95%CI 1.03-1.47), although there was variability in predictive value across sites. Patients meeting EMS provider judgment had concerning clinical presentations qualitatively similar to those meeting mechanistic and other special considerations criteria. Conclusions Among this multi-site cohort of trauma patients, EMS provider judgment was the most commonly used field trauma triage criterion, was independently associated with serious injury and was useful in identifying high risk patients missed by other criteria. However, there was variability in predictive value between sites.