Biases in estimates of black hole kicks from the spin distribution of binary black holes

Abstract: A population of more than 50 binary black hole mergers has now been observed by the LIGO and Virgo gravitational-wave observatories. While neutron stars are known to have large velocities associated with impulsive kicks imparted to them at birth in supernovae, whether black holes receive similar kicks, and of what magnitude, remains an open question. Recently, Callister et al. (2021) analysed the binary black hole population under the hypothesis that they were all formed through isolated binary evolution and claimed that large black hole kicks (greater than 260 km/s at 99% confidence) were required for the spin distribution of merging binary black holes to match observations. Here we highlight that a key assumption made by Callister et al. (2021) -- that all secondary black holes can be tidally spun up -- is not motivated by physical models, and may lead to a bias in their estimate of the magnitudes of black hole kicks. We make only minor changes to the Callister et al. (2021) model, accounting for a population of wider merging binaries where tidal synchronisation is ineffective. We show that this naturally produces a bimodal spin distribution for secondary black holes, and that the spin-orbit misalignments observed in the binary black hole population can be explained by more typical black hole kicks of order 100 km/s, consistent with kicks inferred from Galactic X-ray binaries containing black holes. We conclude that the majority of the binary black hole population is consistent with forming through isolated binary evolution.