The paper investigates the discrepancy between the number of observed and expected satellite galaxies of the Milky Way under the ΛCDM model. It shows that dissipationless ΛCDM simulations predict that most of the most massive subhaloes of the Milky Way are too dense to host any of its bright satellites (L_V > 10^5 L_⊙). These subhaloes have peak circular velocities at infall of 30-70 km/s and infall masses of [0.2-4] × 10^10 M_⊙. This suggests that galaxy formation becomes stochastic at these masses, in contrast to the monotonic relation between galaxy luminosity and halo circular velocity for more massive haloes. The paper also shows that at least two of these massive dark subhaloes are expected to produce a larger dark matter annihilation flux than Draco. The authors suggest that baryonic feedback or different dark matter physics could reduce the central densities of these subhaloes. The paper concludes that if massive dark subhaloes exist in the Milky Way, they would have important implications for indirect dark matter searches. If they do not exist, it may indicate that the Milky Way's subhalo population differs from the average ΛCDM expectation, or that baryonic processes or different dark matter physics strongly affect the abundance or structure of massive subhaloes.The paper investigates the discrepancy between the number of observed and expected satellite galaxies of the Milky Way under the ΛCDM model. It shows that dissipationless ΛCDM simulations predict that most of the most massive subhaloes of the Milky Way are too dense to host any of its bright satellites (L_V > 10^5 L_⊙). These subhaloes have peak circular velocities at infall of 30-70 km/s and infall masses of [0.2-4] × 10^10 M_⊙. This suggests that galaxy formation becomes stochastic at these masses, in contrast to the monotonic relation between galaxy luminosity and halo circular velocity for more massive haloes. The paper also shows that at least two of these massive dark subhaloes are expected to produce a larger dark matter annihilation flux than Draco. The authors suggest that baryonic feedback or different dark matter physics could reduce the central densities of these subhaloes. The paper concludes that if massive dark subhaloes exist in the Milky Way, they would have important implications for indirect dark matter searches. If they do not exist, it may indicate that the Milky Way's subhalo population differs from the average ΛCDM expectation, or that baryonic processes or different dark matter physics strongly affect the abundance or structure of massive subhaloes.