Objective: We investigated the effect of two key timings for basal insulin rate reduction on exercise-induced glucose changes and explored the association between circulating insulin concentrations and muscle vasoreactivity. Research Design and Methods: Twenty adults and adolescents performed 60-min exercise sessions (ergocycle) at 60% VO2peak, 240 min after a standardized lunch. In a randomized order, we compared an 80% basal insulin reduction applied 40 min (T-40) or 90 min (T-90) before exercise onset. Near-infrared spectroscopy was used to investigate muscle hemodynamics at vastus lateralis. Glucose and insulin plasma concentrations were measured. Results: Reduction in plasma glucose (PG) level during exercise was attenuated during T-90 versus T-40 strategy (−0.89 ± 1.89 mmol/L vs. −2.17 ± 2.49 mmol/L, respectively; P = 0.09). Linear mixed model analysis showed that PG dropped by an additional 0.01 mM per minute in T-40 versus T-90 (time × strategy interaction, P < 0.05). The absolute number of hypoglycemic events was not different between the two strategies, but they occurred later with T-90. Free insulin tends to decrease more during the pre-exercise period in the T-90 strategy (P = 0.08). Although local muscle vasodilatation (ΔTHb) was comparable between the two strategies, we found that PG dropped more in cases of higher exercise-induced skeletal muscle vasodilatation (ΔTHb × time interaction P < 0.005, e: −0.0086 mM/min and additional mM of ΔTHb). Conclusion: T-90 timing reduced exercise-induced drop in PG and delayed the occurrence of hypoglycemic episodes compared with T-40 timing without a significant reduction in the number of events requiring treatment.