The holotype of Deltadromeus agilis is a partial skeleton of an animal which is estimated to have measured 8 meters long. The weight of the living animal was estimated to have been around 1050 kilograms, slightly more than an imperial ton. , speculatively restored as a noasaurid, compared to a human A number of specimens were originally considered by Ernst Stromer to be conspecific with Bahariasaurus, but were referred to Deltadromeus by Paul Sereno in 1996. They were thought to come from a much larger individual, with a femur length of 1.22 meters, compared to the 0.74 meter long femur of the holotype. These referred specimens, if legitimately assigned to Deltadromeus, would have indicated that members of the genus could grow up to 12.2 meters in length, approximately the size of a Tyrannosaurus rex. However, the referral of the coracoid, pubes, and hindlimb material catalogued under IPHG 1912 VIII to Deltadromeus has been questioned because the remains came from different horizons and localities in the Bahariya Formation, and actually exhibit notable differences from the holotype of Deltadromeus. The Deltadromeus skeleton has been found in the same formation as those of the giant theropods Carcharodontosaurus, Spinosaurus, and Bahariasaurus, which may be synonymous with Deltadromeus. No skull material has been found for either Deltadromeus or Bahariasaurus, and though carnivore teeth labelled as "Deltadromeus" are commonly sold in rock shops, there is no way of knowing if they actually come from this animal.
Many studies published since the original description of Deltadromeus have considered it to be a ceratosaur, although different studies disagree on what kind of ceratosaur. One 2003 study suggested it was a member of the Noasauridae, though others have found it to be more primitive, possibly related to the primitive ceratosaurs Elaphrosaurus and Limusaurus. A more comprehensive study of noasaurid relationships published in 2016 found that both of these interpretations were essentially correct, with Deltadromeus, Limusaurus, and Elaphrosaurus all found to be within the Noasauridae. A 2017 paper describing ontogenetic changes in Limusaurus and the affect of juvenile taxa on phylogenetic analyses placed Deltadromeus as a noasaurid in every analysis regardless of which Limusaurus specimen was used, although the analyses did not include Gualicho or Aoniraptor. According to the writers of the paper, resolving the phylogenetic positions of Gualicho, Aoniraptor, Deltadromeus, and megaraptorans is one of the most critical issues presently facing theropod systematics. The cladogram below follows a 2016 analysis by Oliver Rauhut, and Matthew Carrano.
''Deltadromeus'' as an avetheropod">Avetheropoda">avetheropod
The original description of Deltadromeus in 1996 found that it was a fairly basal coelurosaur, only slightly more advanced than the Late Jurassic genus Ornitholestes. In 2016, an analysis of Gualicho, a South American theropod considered to belong to the allosauroid family Neovenatoridae, finds Deltadromeus to be Gualicho's probable sister taxon. However, the analysis also noted that Deltadromeus shared many features with ceratosaurs and that if Gualicho was removed from the analysis, Deltadromeus would resolve to a member of Ceratosauria. In an analysis of Aoniraptor, which may be the same animal as Gualicho, Deltadromeus was found along with Aoniraptor and Bahariasaurus to probably form a still poorly known clade of megaraptoran tyrannosauroids different from the Megaraptoridae. A 2018 study by Porfiri et al. has supported the idea that Gualicho and megaraptorans were basal coelurosaurs, outside of both Neovenatoridae and Tyrannosauroidea. However, this study did not include Deltadromeus. The cladogram below follows the 2016 Gualicho analysis by Sebastián Apesteguía, Nathan D. Smith, Rubén Juarez Valieri, and Peter J. Makovicky.
