Handgrip strength (HGS), a simple tool for the evaluation of muscular strength, is independently associated with negative prognosis in many diseases. It is unknown whether HGS is prognostically relevant in COVID-19. We evaluated the ability of HGS to predict clinical outcomes in people with COVID-19-related pneumonia. 118 patients (66% men, 63 ± 12 years), consecutively hospitalized to the “Santa Maria” Terni University Hospital for COVID-19-related pneumonia and respiratory failure, underwent HGS measurement (Jamar hand-dynamometer) at ward admission. HGS was normalized to weight2/3 (nHGS) The main end-point was the first occurrence of death and/or endotracheal intubation at 14 days. Twenty-two patients reached the main end-point. In the Kaplan–Meyer analysis, the Log rank test showed significant differences between subjects with lower than mean HGS normalized to weight2/3 (nHGS) (< 1.32 kg/Kg2/3) vs subjects with higher than mean nHGS. (p = 0.03). In a Cox-proportional hazard model, nHGS inversely predicted the main end-point (hazard ratio, HR = 1.99 each 0.5 kg/Kg2/3 decrease, p = 0.03), independently from age, sex, body mass index, ratio of partial pressure arterial oxygen and fraction of inspired oxygen (PaO2/FiO2 ratio), hypertension, diabetes, estimated glomerular filtration rate and history of previous cardiovascular cardiovascular disease. These two latter also showed independent association with the main end-point (HR 1.30, p = 0.03 and 3.89, p < 0.01, respectively). In conclusion, nHGS measured at hospital admission, independently and inversely predicts the risk of poor outcomes in people with COVID-19-related pneumonia. The evaluation of HGS may be useful in early stratifying the risk of adverse prognosis in COVID-19.