As is so often the case, we do not know the extent to which the HDR processing improved the original shots, unless we are lucky enough to get some additional insight into the process.
In my opinion, HDR work should include a median shot - i.e., a straight from the camera with no pp that also happens to be just on the verge of clipping the highlights.
Also recommended is that the mode (JPG or RAW) be mentioned, as well as the specific number of different images that were used in the merge; and whether or not those images were actually different shots or just differently processed versions of a single RAW file.
Lastly, the processing software should be mentioned - e.g., PSCS3 Merge to HDR, Photomatix, etc.
One of the benefits of all this for the critics is the issue of whether or not the photo actually needed HDR treatment. or was it treated as HDR just to see if it would have any significant effect. Another benefit to all would be the inherent tutorial aspect of the process - i.e., giving the viewers a taste of what a great tool is HDR.
One thing that really struck me about the choice of images used is the fact that the exposure times were in the ratio of 64:1 which is 6 f-stops. That is an extreme range, from my experience in that 3 to 4 f-stops is about as large a range as I have used.