Objective Acute myocardial damage is detected in a significant portion of patients with coronavirus 2019 disease (COVID-19) infection, with a reported prevalence of 7–28%. The aim of this study was to investigate the relationship between electrocardiographic findings and the indicators of the severity of COVID-19 detected on electrocardiography (ECG). Methods A total of 219 patients that were hospitalized due to COVID-19 between April 15 and May 5, 2020 were enrolled in this study. Patients were divided into two groups according to the severity of COVID-19 infection: severe ( n = 95) and non-severe ( n = 124). ECG findings at the time of admission were recorded for each patient. Clinical characteristics and laboratory findings were retrieved from electronic medical records. Results Mean age was 65.2 ± 13.8 years in the severe group and was 57.9 ± 16.0 years in the non-severe group. ST depression (28% vs. 14%), T-wave inversion (29% vs. 16%), ST-T changes (36% vs. 21%), and the presence of fragmented QRS (fQRS) (17% vs. 7%) were more frequent in the severe group compared to the non-severe group. Multivariate analysis revealed that hypertension (odds ratio [OR]: 2.42, 95% confidence interval [CI]:1.03–5.67; p = 0.041), the severity of COVID-19 infection (OR: 1.87, 95% CI: 1.09–2.65; p = 0.026), presence of cardiac injury (OR: 3.32, 95% CI: 1.45–7.60; p = 0.004), and d-dimer (OR: 3.60, 95% CI: 1.29–10.06; p = 0.014) were independent predictors of ST-T changes on ECG. Conclusion ST depression, T-wave inversion, ST-T changes, and the presence of fQRS on admission ECG are closely associated with the severity of COVID-19 infection.