PURPOSE: Television (TV) watching is the most prevalent sedentary leisure time activity in the United Kingdom. We examined associations between TV viewing time, measured over 10 yr, and two objective measures of physical capability, usual walking speed (UWS) and grip strength. METHODS: Community-based participants (n = 8623; 48-92 yr old) enrolled in the European Prospective Investigation of Cancer-Norfolk study attended a third health examination (3HC, 2006-2011) for measurement of maximum grip strength (Smedley dynamometer) and UWS. TV viewing time was estimated using a validated questionnaire (n = 6086) administered during two periods (3HC, 2006-2007; 2HC, 1998-2000). Associations between physical capability and TV viewing time category (<2, 2 < 3, 3 < 4, and ≥4 h·d) at the 3HC, 2HC, and using an average of the two measures were explored. Sex-stratified analyses were adjusted for age, physical activity, anthropometry, wealth, comorbidity, smoking, and alcohol intake and combined if no sex-TV viewing time interactions were identified. RESULTS: Men and women who watched the least TV at the 2HC or 3HC walked at a faster usual pace than those who watched the most TV. There was no evidence of effect modification by sex (Pinteraction = 0.09), and in combined analyses, participants who watched for <2 h·d on average walked 4.29 cm·s (95% confidence interval, 2.56-6.03) faster than those who watched for ≥4 h·d, with evidence of a dose-response association (Ptrend < 0.001). However, no strong associations with grip strength were found. CONCLUSIONS: TV viewing time predicted UWS in older adults. More research is needed to inform public health policy and prospective associations between other measures of sedentariness, such as total sitting time or objectively measured sedentary time, and physical capability should be explored.