Ngram viewer, so appearances of the string in the google books corpus. I didn't add 'red nails' as I expected there would be more noise than signal with such a common string.
No idea why Queen of the Black Coast is so much higher than the others. Bêlit is a well-loved figure, though, and some do consider it one of the stronger stories, plus it is likely to be a major touchstone if someone is discussing Howard's racism.
1865 is noise - some other appearance of the string
Sorry...I still didn't understand exactly what this graph is supposed to be measuring. Sales? Mentions of a particular title in internet threads? Mentions in scholarly articles? I had no idea what this data is showing me - it seems to show a massive growth in the popularity of a particular Conan story in recent years, but that seems very implausible.
So I did some digging. The data presented as evidence of the rising popularity of REH stories is very misleading.
Here is a specific academic criticism to using Ngrams in the way attempted:
We therefore observe that the Google Books corpus encodes only a small-scale kind of popularity: how often
n-grams appear in a library with all books given (in principle) equal importance and tied to their year of publication (new editions and reprints allow some books to appear more than once). The corpus is thus more akin to a lexicon for a collection of texts, rather than the collection itself. But
problematically, because Google Books n-grams do have frequency of usage associated with them based on this small-scale popularity, the data set readily conveys an illusion of large-scale cultural popularity. An
n-gram which declines in usage frequency over time may in fact become more often read by a particular demographic focused on a specific genre of books. For example, “Frodo” first appears in the second Google Books English Fiction corpus in the mid 1950s and declines thereafter in popularity with a few resurgent spikes [
4].
and
Even if we are able to restrict our focus to popular works by appropriately filtering scientific terms,
the library-like nature of the Google Books corpus will mean the resultant normalized frequencies of words cannot be a direct measure of the “true” cultural popularity of those words as they are read (again, Frodo). Secondarily, not only will there be a delay between changes in the public popularity of words and their appearance in print, normalized frequencies will also be affected by the prolificacy of the authors. In the case of Upton Sinclair’s Lanny Budd, a fictional character was vaulted to the upper echelons of words affecting divergence (even surpassing Hitler) by virtue of appearing as the protagonist in 11 novels between 1940 and 1953. Google Books is at best a limited proxy for social information after the fact. [emphasis added]
It is tempting to treat frequency trends from the Google Books data sets as indicators of the “true” popularity of various words and phrases. Doing so allows us to draw quantitatively strong conclusions about the evolution of cultural perception of ...
pmc.ncbi.nlm.nih.gov
Edit: to illustrate the issue, try typing in Gandalf, Frodo, and Gollum separately and compare your results. Or try putting in "Conan the Barbarian" and compare to Queen of the Black Coast. Apparently the former peaked in 1960 and the latter in 2020. That make sense to anyone?
I also note that Ngrams were not designed for, nor purport to measure the popularity of entire works of fiction in the way implied by the presented graphs. They are nonsense.