This turned out to be the least reliable indicator this year. The 2016 Best Novel Hugo winner was fourth out of five by ownership on both Goodreads and LibraryThing, and the 1941 Best Novel Retro Hugo winner was second on both, but a very long way behind. Having said that, both winners had the highest average rating from readers on both systems.
For the first time in the years I have been tracking these things, the consensus among bloggers called all four written fiction categories correctly. It's never been more than two out of four before.
Both Steve Davidson and I myself took the reputational risk of predicting who would win every single category. I got three wrong (Best Dramatic Presentation, Short Form; Best Editor, Short Form; Best Professional Artist) though in two of those three cases my second guess was right. Steve called only nine of the seventeen categories correctly. He and I were both caught out by Abigail Larson's win for Best Professional Artist, and we also both expected Doctor Who to beat Jessica Jones.
What this shows
is that the Hugos remain capable of surprising us, and even though this was a year where the written fiction winners were relatively easy to foresee, many of the categories were very difficult to call. I look forward to other peoples' commentaries next year.