We discuss whether electron transfer from a photoexcited polymer donor to a fullerene acceptor in an organic solar cell is tractable in terms of Marcus theory, and whether the driving force ΔG0 is crucial in this process. Considering that Marcus rates are presumed to be thermally activated, we measured the appearance time of the polaron (i.e., the radical-cation) signal between 12 and 295 K for the representative donor polymers PTB7, PCPDTBT, and Me-LPPP in a blend with PCBM as acceptor. In all cases, the dissociation process was completed within the temporal resolution of our experimental setup (220–400 fs), suggesting that the charge transfer is independent of ΔG0. We find that for the PCPDTBT:PCBM (ΔG0 ≈ −0.2 eV) and PTB7:PCBM (ΔG0 ≈ −0.3 eV) the data is mathematically consistent with Marcus theory, yet the condition of thermal equilibrium is not satisfied. For MeLPPP:PCBM, for which electron transfer occurs in the inverted regime (ΔG0 ≈ −1.1 eV), the dissociation rate is inconsistent with Marcus theory but formally tractable using the Marcus–Levich–Jortner tunneling formalism which also requires thermal equilibrium. This is inconsistent with the short transfer times we observed and implies that coherent effects need to be considered. Our results imply that any dependence of the total yield of the photogeneration process must be ascribed to the secondary escape of the initially generated charge transfer state from its Coulomb potential.