In this study, a qualitative equivalence between the electrical percolation threshold and the effective thermal conductivity of composites filled with cylindrical nanofillers has been recognized. The two properties are qualitatively compared on a wide range of aspect ratios, from thin nanoplatelets to long nanotubes. Statistical continuum theory of strong-contrast is utilized to estimate the thermal conductivity of this type of heterogeneous medium, while the percolation threshold is simultaneously evaluated using the Monte Carlo simulations. Statistical two-point probability distribution functions are used as microstructure descriptors for implementing the statistical continuum approach. Monte Carlo simulations are carried out for calculating the two-point correlation functions of computer generated microstructures. Finally, the similarities between the effective conductivity properties and percolation threshold are discussed.