Heading off on a tangent from my last tangent. In my previous post I talked about ‘windows’ for authors and the Hugo Awards. Having done some number crunching for that post I thought I’d explain a bit more about the numbers etc.
The data I’m looking at was taken from the Wikipedia pages on the Hugo Awards for Best Novel, Novelette, Novella and Short Story. While there is overlap between finalists for these categories and related categories such as Fan Writer, Editor, or the various kinds of Best Newcomer awards, I restricted it just to the specific story awards. In cleaning up the names I did end up removing various notes etc, so for example Ken Liu appears five times: three times for works he wrote and twice for works he translated. I did check for variations in name spellings but I did not simplify aliases (so Mira Grant and Seanan McGuire appear as separate people). I ignored the retro-Hugos.
For each writer I tracked for each year they they were a finalist, the number of years since they were first a finalist. So the data for George R. R. Martin would look like this if isolated:
Using pivot tables in Excel, that data gave every author a maximum “change” which is the width of their Hugo window so far.
As I said in the earlier post, the results are hard to analyse because the spread of window sizes looks like it has something like a Pareto distribution. Lots of people have a 0 year window (they were only finalists in one year) but a small number of people (Asimov and Martin) have a very big window. The mean number of years for a window is about 5.5 years but that mean is not a terribly useful number because most finalist have a smaller window than that.
The other issue is that the potential size of windows increases over time. The Hugo Awards have been running since 1953, so the biggest possible window is 67 years but last year it was only 66 years and so on. Of course, practically we really don’t expect anybody to have a 60+ year interval of being a Hugo finalist. It’s not impossible but it’s unlikely.
So a key question is what’s the effective upper limit of a Hugo window for writers? [Note: other categories this could be quite different] Well, 40 years seems to be a round number for the ceiling to window if I can mix building metaphors. This graph tells the story:
The graph tracks the maximum value for all authors for the window in that year. The orange line is simply the number of years since 1953, which is the hard arithmetic limit on the window size.
What happens is the maximum window size grows each year, lagging a bit behind the orange line. This continues until the mid-1980s and the window size grows much more slowly. It is, coincidentally or not, around the point in time where people have claimed a change in character to the Hugo Awards. The qualitative change is that it around the point where the last people who were finalist in the mid 1950s were finalists for the last time. Of course there had already been massive generational shifts in the awards by then.
What I find fascinating is that to make sense of this number really takes thirty years of data. To see if things really are different now with these numbers, I’ll need to wait until 2050!
In terms of windows from then, is actually a lack of change. The future doesn’t look that different from the past. Here are all of the growing windows for every author (i.e. the size of an every author’s window in each year they were a finalist). The orange line is again the arithmetic maximum (e.g. if somebody find an unpublished story by Alfred Bester and publishes it and it is a Hugo 2021 finalist etc…)
Are there patterns in there? Maybe. I fancy I can see waves of diagonal lines with shorter lines between them and maybe that zone pre-2010 is a little thinner…could be but I don’t think that’s a scientific conclusion 🙂