Introduction Research investigating psychometric properties of multi-joint upper body strength assessment tools for older adults is limited. This study aimed to assess the test–retest reliability and concurrent validity of novel clinical strength measures assessing functional concentric and eccentric pushing activities compared to other more traditional upper limb strength measures. Methods Seventeen participants (6 males and 11 females; 71 ± 10 years) were tested two days apart, performing three maximal repetitions of the novel measurements: vertical push-off test and dynamometer-controlled concentric and eccentric single-arm press. Three maximal repetitions of hand-grip dynamometry and isometric hand-held dynamometry for shoulder flexion, shoulder abduction and elbow extension were also collected. Results For all measures, strong test–retest reliability was shown (all ICC > 0.90, p < 0.001), root-mean-squared coefficient of variation percentage: 5–13.6%; standard error of mean: 0.17–1.15 Kg; and minimal detectable change (90%): 2.1–9.9. There were good to high significant correlations between the novel and traditional strength measures (all r > 0.8, p < 0.001). Discussion The push-off test and dynamometer-controlled concentric and eccentric single-arm press are reliable and valid strength measures feasible for testing multi-joint functional upper limb strength assessment in older adults. Higher precision error compared to traditional uni-planar measures warrants caution when completing comparative clinical assessments over time.