We develop a renewable energy mix optimization program. The main novelty is to take into account and evaluate the impact of climate variability, from multi-decadal to hourly time scales, on the mix. This impact is resolved by plugging regional climate simulations over the 1989–2012 period into renewable energy production and demand models at regional scale. The optimal mix is then inferred from a mean-variance analysis with as objectives both to maximize the mean of the total renewable penetration and to minimize the variance, or risk, of the latter. We consider two cases: in the first one, the analysis takes cross-region correlations in the production and the demand into account. In the second one, synergies between regions are ignored. As a first case study, we apply the numerical model to Italy, a country with a relatively high share of renewables. We focus on the mix maximizing the ratio of the mean renewable penetration over the risk for the same renewable capacity as installed in 2015. Our main findings are that (i) the optimal technological and spatial capacity distribution — with about two thirds wind and one third solar — differs significantly from the actual mix (with a substantial overinvestment in photovoltaics); and (ii) ignoring the impact of interannual and intraday climate variability on the risk strongly impacts the optimal mix. This study shows the importance of basing renewable energy development incentives on optimal strategies taking into account both the risk and the impact of climate variability on the latter.