…that last post got a bit derailed by a sudden and immediate loss of personal objectivity.
OK the category to look at for EPH fun is Best Fan Artist. This category has a wide field, an intrinsically hard to judge topic, and low participation by voters. This means a much higher chance of a tie. One ‘feature-not-a-bug’ qualities of EPH is that it should reduce actual ties but that does mean the results start to look complex and arbitrary.
Virtual ties (in my head at least) are also important – where the number of votes between two people is very close. If it is down to one vote then it is good that EPH adds more elements to the contest.
At the low end of the nominations were these outcomes: (v = votes, p = EPH points)
- v 39 p 28.25 Spring Schoenhuth
- v 33 p 29.00 Steve Stiles
- v 34 p 24.00 Elizabeth Leggett
- v 33 p 24.75 Ariela Housman
- v 33 p 20.17 Galen Dara
- v 27 p 19.57 Richard Man
- v 28 p 14.83 Kathryn M. Weaver
Lots of close vote totals and low points. The stats PDF has a nice explanation:
“Pay attention, this is complicated.
If Galen Dara had had 0.33 more points, Elizabeth Leggett would have been eliminated on an earlier count and Galen Dara would probably have been a finalist.
Ariela Housman needed 1 more vote to displace Elizabeth Leggett from the final ballot.
To do the same, Richard Man needed at most another 7 votes worth 4.43 more points, Megan Lara needed at most another 9 votes worth 8.00 points, and Kathryn Weaver needed at most another 9.17 points.
Steve Stiles would have been displaced by Ariela Housman with 5.25 fewer points, and Spring Schoenhuth would have been displaced by Ariela Housman with 7 fewer votes, or 6 fewer votes worth more than 3.5 points.
If Spring Schoenhuth had had precisely 6 fewer votes worth precisely 3.5 points, she and Ariela Housman would have tied for the last ballot spot and both would have qualified when Mansik Yang was disqualified.
Put another way. If Galen Dara had been *less* popular with people who had voted for higher ranked nominees THEN she might have made the ballot. Likewise, if Steve Stiles’s support was more common among the other top rated nominees then he wouldn’t have made the ballot.
Is that fair? YES! Look at what it is doing. In borderline cases, close races are being decided by results that MAXIMISE the number of voters the resulting set of finalists represent. Put another way, more voters get somebody they voted for in the list of finalists. OK, in this case, the loaf is getting sliced microscopically thin but the only super-fair way of resolving this would have been to have 9 or 10 finalists.
Onto the Rabid nomination vote (not counting hostages).
- Novel 85 votes
- Novelette 77 votes
- Short Story 87 votes
- Best Editor Long 83 votes
- Semipro 80 votes
- Fanzine 85 votes
- Fancast 76 votes
- Fan writer 80 votes
- Campbell 91 votes
J.Mulrooney, the Castalia House nominee for Best Novel and the Campbell got 6 more votes for the Campbell than for Best Novel – which is a bit odd I suppose. John C Wright’s Short Story would have also picked up some votes from random Puppies still involved and I’d have thought it would have landed further from the Rabid mean…but it was effin awful even by JCW’s standards so maybe not.
So a mean in the low 70s and a median of 83 votes. Which looks to be irrelevant because those votes are probably all from 2016 members. The Rabid vote drops precipitously in the final numbers.
- Novelette 45 votes
- Short story 58 votes
- Best Editor Long 32 votes
- Semipro 17 votes (ouch)
- Fanzine 22 votes
- Fancast 15 votes
- Fanwriter 20 votes
- Campbell 26 votes
That’s it for now. Have to eat ice-cream for breakfast it seems 🙂