Buzz, regarding (4): "Stellar evolution models, chemical enrichment by supernovae within the Milky Way, and gravitational microlensing events all indicate that a population of about ∼108 -109 stellar-mass black holes resides in our Galaxy (Shapiro & Teukolsky 1983; van den Heuvel 1992; Brown & Bethe 1994; Samland 1998; Agol et al. 2002). And yet, despite such high estimates, fewer than fifty stellar black hole candidates have been studied and confirmed,
all of which are in X-ray binary systems." re;;
https://arxiv.org/pdf/1704.03455.pdf ,Hunting Black Holes with Gaia]
Regarding the other flip floppy sounding mass fractions, I defer to;
https://arxiv.org/pdf/1704.03455.pdf, Measuring the binary fraction of massive stars using young star clusters: the resolved cluster RSGC1, "It has long since been known that a substantial fraction (∼ >50%) of massive stars are in binaries, and that the companion mass distribution is skewed to higher masses than would be expected if it were randomly sampled from the initial mass function (IMF) (e.g. Abt & Levy 1978; Garmany et al. 1980; Gies 1987; Mason et al. 1998; Garc´ıa & Mermilliod 2001; Kobulnicky & Fryer 2007). In terms of the impact of binarity on stellar evolution, the important quantity is not what fraction of massive stars are in binaries, but what fraction will interact during their lifetime. The most recent measurements indicate that the interacting binary fraction for massive stars may be very high, 50 − 70% (Sana et al. 2012, 2013). These results imply that effects of binarity on the evolution of massive stars cannot be neglected, and therefore call into question all predictions for massive stellar evolution which have their foundation in single-star models." The paper further asserts "Formally, the derived binary fraction for the stars of the same mass as the RSGs, in this case ∼18M, is fbin=37+27 −23. ...This result is somewhat at odds with other recent estimates of binary fractions of massive stars as measured from radial velocity surveys in massive your clusters, which have yielded values of fbin in the region of 50-80%, somewhat lower than other recent estimates of >50%." Further confounding matters is this paper;
https://arxiv.org/abs/1306.1811, The Multiplicity of High-Mass Stars, "We report about an ongoing photometric and spectroscopic monitoring survey of about 250 O- and 540 B-type stars in the southern Milky Way with the aim to determine the fraction of close binary systems as a function of mass and to determine the physical parameters of the individual components in the multiple systems. Preliminary results suggest that the multiplicity rate drops from 80% for the highest masses to 20% for stars of 3 solar masses. Our analysis indicates that the binary systems often contain close pairs with components of similar mass. This coincidence cannot originate from a random tidal capture in a dense cluster but is likely due to a particular formation process for high-mass stars. "