A paper finally comes along listening to reason. Said paper is “Avian ancestors: A review of the phylogenetic relationships of the theropods Unenlagiidae, Microraptoria, Anchiornis and Scansoriopterygidae” (Federico L. Agnolín and Fernando E. Novas 2013)., which not only expresses the now pratically omnipresent tendencies to consider maniraptor phylogeny to be in serious need for a revamp, but also points out exactly why Microraptor and kin were most likely flying animals.
The primary point of interest is that the paper highlights one thing pretty much nobody had taken into consideration previously: the microraptoran shoulder girdle is uniquely specialised for powered flight. It has a socket curvature that places the shoulder much higher than in the average theropod, allowing for a full vertical upstroke; combined with previous assessments that archosaur limb range was far more flexible than previously thought, this pretty much means that the notion that maniraptors couldn’t raise their arms significantly is pretty much dead.
In addition, the paper also highlights the unique adaptations for powered flight seen in Microraptor, like the presence of an allula, the presence of a propatagium and the already noted well developed deltoideus complexes and tail rods.
The paper, as the name implies, also restructures deinonychosaur phylogeny. In particular, it renders it polyphyletic, with microraptorans and unenlagiines being closer to birds than to other deinonychosaurs, and within these some taxa are moved closer to Aves (i.e. Rahonavis).
This pretty much suggests one of two things: that powered flight was probably acquired in the last common ancestor between Passer and Microraptor, to the exclusion of several “classical deinonychosaurs”; or that flight was in fact acquired and lost multiple times among dinosaurs (an idea supported if Xiaotingia+Anchiornis is closer to Aves than the other taxa examined, as well as if Pelecanimimus‘ infamous flight-y traits still apply).
Considering that eudromaeosaurs still possess tail rods and quill knobs, that their common ancestor evolved flight independently also becomes a serious possibility.