Sleep is increasingly recognised as an important risk factor for metabolic disease, and as an important influence on severity in established metabolic disease. Recent evidence suggests that sleep timing variability (the day-to-day fluctuations of sleep timing) may be an important factor in metabolic diseases such as type 2 diabetes mellitus. In the current study, we explore the associations between measures of sleep timing variability and cardiometabolic measures in a group of healthy middle-aged adults with and without type 2 diabetes mellitus. Healthy controls (N = 27) and adults with well-controlled uncomplicated type 2 diabetes mellitus (N = 30) wore actiwatches for an average of 9 days for objective assessment of sleep timing parameters, and also underwent a detailed clinical assessment. We found greater self-reported social jetlag in the diabetes group, but no groupwise differences in measures of sleep timing variability. In the diabetes patients, HbA1c levels were inversely correlated with variability in the time of sleep onset and midsleep, and with sleep duration. HOMA-IR did not correlate with any sleep timing variability measure, nor were there associations between sleep timing variability and other metabolic biomarkers (cholesterol, LDL, HDL, triglycerides and uric acid). Systolic blood pressure was inversely correlated with actigraphically defined social jetlag in both the control and diabetes groups. The results of this study indicate associations between sleep timing variability and HbA1c, but the direction of these relationships is at variance with some other recent reports. Our results indicate a need for future hypothesis-testing studies to further explore the impact of sleep timing variance on metabolic health.