This haplogroup has in some degree lifted itself out of a confusing history, to a point at which we can at least begin to sort its phylogenetic relationships with several SNPs (both above and below it, on a tree diagram) that were identified before the discovery (in the fall of 2011) of DF27, in 1000 Genomes project samples.
As far as we know at this point, DF27(S250) is immediately below P312(S116), and is one of three very large haplogroups on that same level; the others, both better known in the literature, are U152(S28) and L21(S145). Smaller groups on this level appear to include L238, DF19, and the very recent discovery DF99.
Because of the sequence of their discovery, several subclades of DF27 already had their own haplogroup projects under the FTDNA tent before DF27 had one. These include L176.2, SRY2627, and M153. There is also a considerable volume of posts on the broader WorldFamilies R1b forum pertaining to Z196, Z209, Z220, L165, DF17, and other SNPs found under this DF27 haplogroup.
The main haplogroup project website is at http://www.familytreedna.com/public/R1b-DF27/
That site has several links, including one to a DF27 Yahoo research group that one may join to keep track of the latest SNP discoveries and refinement of the phylogeny. More generally, the latest version of the ISOGG tree for the R1b haplogroup is authoritative:http://www.isogg.org/tree/ISOGG_HapgrpR.html
However, ISOGG is rarely if ever up to date with the most recent discoveries; the acceptance process and vetting take time. An excellent source for newer discoveries in the phylogeny is the experimental haplotree maintained by Christopher Morley:http://ytree.morleydna.com/experimental-phylogeny
At that site, click on the url for the most recent update; then scroll down (perhaps a bit more than 40 pages) to the part of the R1b tree that includes DF27 and its subclades.
A graphic representation of the DF27 haplotree that is quite current at this writing may be seen at this url:https://dl.dropboxusercontent.com/u/17907527/R1b-DF27_Descendency_Tree.jpg
New discoveries are arriving at a rapid pace from next-generation (chip) testing, such as Geno2 and Chromo2; from full genome and full Y-chromosome sequencing of samples from DF27 males; and from the screening or retesting of samples in older projects (such as 1000 Genomes), once the locations and mutations involved in new SNP discoveries have become available to academics and to citizen-scientists. The field is changing rapidly, and the DF27 phylogeny will change as well. These remarks are pretty current in late October, 2013. If you are reading them much later than that, they are probably outdated. But perhaps the posted links can still lead interested parties to updates.