Extended periods of dark doldrums, when the wind isn't blowing and the sun not shining, can pose reliability risks on future energy systems with high levels of variable wind and solar electricity generation. Better understanding these risks and applying methods to account for them in electricity system planning models is important to achieve a successful energy transition. This research analyzes an extended history of weather data to provide insight on the likelihood and characteristics of dark doldrums risks for electricity system planners. It also demonstrates planning methods to optimize low-carbon electricity portfolios that incorporate dark doldrums risk while maintaining reliability. The characteristics and frequency of dark doldrum events identified in the historical record depend significantly on how they are defined. It is found that future system reliability risks with dark doldrums occur during morning or evening hours on days with low wind and relatively high demand, particularly during extended wind droughts lasting multiple days. Promising areas for future research include: 1) extending historical weather analysis to national- international, and/or global-levels to understand regional variation in doldrums characteristics; 2) Demonstrate capacity expansion scenarios to identify portfolios robust to wider ranges of doldrums uncertainties, 3) Fine-tune and generalize “extreme week” capacity expansion chronology methods demonstrated in this study and 4) Explore more spatially-granular siting decisions to optimize investments with uncorrelated resources and potentially mitigate doldrums impacts.