What EPH would have meant in practice

[nwhyte]

MidAmeriCon II, this year's WorldCon now in full swing in Kansas City, has published a retrospective count of the Hugo nominations of the last two years as they would have been if the proposed new EPH system had been implemented. This is to an extent a counterfactual exercise - particularly for last year, when a number of finalists withdrew

Comments

[anonymous]

I'm pretty sure that the diversity effect you're seeing is due to a small sample. I see no plausible mechanism by which EPH could systematically hurt gender diversity in normal years.

[nwhyte]

Just because you don't see a plausible mechanism, whoever you are, doesn't mean that there isn't one.

When EPH was first proposed, I expressed the concern that port that the quirky candidate that got a lot of people's fourth or fifth nominating votes will lose out. I think that is what we are seeing here.

[despotliz]

The bad effects are partially because EPH makes ties much less likely, I think. So while I don't think it's great that the fancast nominees drop off, I don't think it's bad to end up with 5 nominees instead of 7.

Anyway, at this point we ratified it. I think you are now instructed by the business meeting to produce the nominees for next year's Hugos as though they had been run under EPH and the old system, and since the Hugos will be earlier in the con next year there should be ample time to compare the two options before the final business meeting and if necessary vote to suspend EPH for 2018.

[nwhyte]

For the record, I was very much reassured by the data released on Sunday with regard to the 2016 and 1941 ballots. The 1941 ballot would have been barely changed (Hannes Bok would have dropped off the Best Professional Artist category, but I put him fifth anyway), and the 2016 ballot would have been vastly improved. So I'm more ready to accept that the changes that concerned me on the 2014 ballot are essentially edge effects. I shall reflect this in a full post once the full 1941 and corrected full 2016 statistics have been published.

And, as you rightly point out, I am now obliged to count next year's nominations both ways, though the EPH results are the ones that will count.