Background The thumb carpometacarpal (CMC) joint orchestrates pinch in its various positions, and thumb CMC osteoarthritis (OA) is a major source of orthopaedic morbidity. Self-reported pain, weakness, and physical examination may not correspond to radiographic findings when diagnosing early thumb CMC OA. Weakness is a prominent feature of the disease, but little evidence exists to quantify self-reported loss of strength with time, or to compare weakness with that of a nonarthritic population during early disease. Questions/purposes We asked: (1) Is pinch strength in subjects with early thumb CMC OA less than that in asymptomatic control subjects; and (2) weakness in which pinch position (key, tripod, or tip pinch) has the strongest association with early OA diagnosis? Methods For this case-control study, we recruited 23 subjects who were asymptomatic and 91 with early OA for comprehensive history, physical examination, strength measurements, pain surveys, and radiographic evaluation. We used multivariate logistic regression to quantify the association between declining pinch strength and early OA diagnosis after controlling for age, sex, and BMI. This analysis was performed for three different pinch positions (key, tripod, and tip pinch) to evaluate which measurement was most closely associated with diagnosis. Results Pinch strength was less in patients with early thumb CMC OA. Key pinch had the most robust association with OA diagnosis, in which a 20% decrease in key pinch strength from the control subjects' baseline was associated with a 10% increase in the OA diagnosis (95% CI, 3%-16%; p = 0.004). This had a stronger association with OA diagnosis than tip pinch, in which a 20% decrease in strength was associated with a 6% increase in early CMC OA (95% CI, 1.0%-11%; p = 0.031). Tripod pinch also was associated to a lesser extent; a 20% reduction in tripod pinch led to a 5% increase in OA (95% CI, 1.3%-9%; p = 0.048). Conclusions Decreasing pinch strength, especially key pinch, is associated with early CMC arthritis before the development of extensive radiographic disease.