So last time, we were talking about the World Science Fiction Society Business Meeting, and all the proposals and such that were brought forth. The main event, of course, was E Pluribus Hugo – an amendment to modify the Hugo voting system to reduce the disproportionate impact of slate voters.
As a reminder, let’s go over what happened: about 13%-15% of voters participated in a mass slate vote – with pretty good but not perfect discipline – to support a slate which was specifically political in intent. As a result, they captured all the nominations in several important categories.
All those categories ended up going to NO AWARD as fandom decided to punish the slate activists for violating several decades of “we won’t do this” consensus. It wasn’t that the exploit was unknown; it was merely that using it had been considered socially unacceptable. Thou shalt not campaign, thou shalt not form parties, and so on. And the reason is simple: one party vs. unorganised parties always wins, so competing parties always arise in response, and the value of such an award – an award which has become purely a political football – is exactly zero.
E Pluribus Hugo changes the system in such a way that it reduces slates to their strongest candidates relative to their percentage of the total popular vote. It does not eliminate slates entirely, though it does reduce their presence on the final ballot to match the percentage of people voting for them; it does not single them out for special treatment; most of all, it does not need to be told, “this is a slate, discount it.” That judgement call never happens. It’s purely the fallout of the math.
The way it works is simple. Each category is treated separately, just like now. Each WSFS member gets to nominate up to five works in category, just like now – in fact, nothing the WSFS nominator does changes.
Each of these ballots gets assigned one point, which is split across all works nominated. In a full ballot of five nominees, each work would have 0.2 points, as well as one vote each, from a member.
The point total and vote total of all the nominated works from all ballots are added up. Then, the two works with the fewest points are pitted against each other, and the one with fewer votes – the one for whom fewer people voted – is eliminated.
This is important, because the point total never eliminates an nominee. Getting fewer votes eliminates a nominee. Total votes received remains the final call.
Once a work is eliminated, it is stricken from all ballots, and we start over again. If you nominated five works originally, and one was eliminated, your ballot now has four nominees, and each of those have a higher point value than before – a quarter point (0.25) instead of a fifth of a point (0.20). And the same steps are run through again, exactly as before.
Wash, rinse, repeat, removing the weakest each time, until five nominees remain; that is your final slate.
What this does in practice is start pitting slate entries against each other roughly midway through the tallying process. Assuming they have even reasonable support, the strongest – the one with the most people voting for it, which implies out-of-slate support – will emerge. If the slate is sufficiently weak, none of them will emerge, but for practical purposes, the number of survivors will be roughly proportional to the percentage of popular vote actually received.
“But Solarbird,” I hear you cry, “This lets some nominations from slates get on the ballot!” True! But only in proportion to their actual popular support. And in the event of political slates, it means we do not have to go to the NO AWARD option to block them.
Let’s say the events of this year were repeated under this system; with statistical approximations of real data, we’re pretty sure one of the Puppy candidates probably would’ve made it onto the ballot in most of their categories. It would’ve been the strongest; the one with the most outside support.
And that’s okay. If it’s crap, it’ll finish last, maybe behind NO AWARD, maybe not. But there will be four other nominees, because they’ll have the percentage of the ballot that aligns with their actual bulk support.
The rest of the ballot will provide a diversity of choices. We won’t have another year of five NO AWARD votes.
(And if it’s actually good – great! That’s kind of the point. Vote for it.)
This makes opposition slates completely unnecessary. Opposition slates arise when they are the only way to get non-slate works onto a ballot. Under the current system, that outcome is inevitable. Under E Pluribus Hugo, even if you do get an opposition slate, well, okay, maybe they get one nominee on the ballot too. That leaves three for traditional candidates.
Slates are a lot of work. Politically-minded slates are just as much work, even when the mighty power of spite drives the engines. So if you can’t stick it to the Whoevers without literally becoming the entire show, if you can’t lock them all out, then even all the ressentiment in the world probably won’t drive you to continue. There’s too much work and too little reward. There’s simply no point to it.
The system isn’t even political. It’ll reduce, say, an accidental Doctor Who episode slate down to its proportion of the vote just as effectively. Let’s say 60% of WSFS fandom puts down basically the same five episodes of Doctor Who for Dramatic Presentation – Short Form. Right now, they own the entire ballot. Under E Pluribus Hugo, they own 60% of the ballot, and other works can be considered too.
Because that’s the brilliance of it. I said this before, but it’s really important, so I’m going to say it again:
E Pluribus Hugo doesn’t know about intentional slates. It doesn’t need to be told, “this is a slate.” Nobody has to make that call, because it doesn’t matter. It’s kind of like a normalisation function applied to nominations. There are no arguments over whether a pattern or voting is intentional or a plot or intent or political – a lot of identical ballots will be normalised to a first-order approximation of their actual popular support, regardless.
That’s why it’s so elegant, and that’s why it’s so genius. It doesn’t lock anybody out; it just stops campaigns from locking everyone else out, dramatically reducing their value vs. their labour and monetary cost, and eliminating the incentive for opposition parties.
For me, that is fair. For me, that is enough.
I hope that, for the honest flank of the Sad Puppies, it will also be enough. One self-identified Sad came up and voiced active support for E Pluribus Hugo during the business meeting. Those who actually believe in the mythical SJW VOTER CABAL – which was emphatically demonstrated not to exist by the events of this year, but stick with me – will know that E Pluribus Hugo would normalise this supposed SJW CABAL slate just as effectively.
Is it sad that we’ve reached a point where this sort of engineering is necessary? Eh, maybe. Probably, even. But it has driven fandom to create what even some opponents at the business meeting called a more perfect nominating system.
Yes, it’s tedious as all hell to do by hand, but it can be done. Yes, it’s more complicated – but not much. It’s only a little different than what we do for final voting and for site selection already.
Yes, it’s more work for the Hugo administrators. That’s the downside. But from what I was hearing at the business meeting, there are a good number of inefficiencies in the current tallying system. Fix those, and the extra complexity of this system sounds to me like a wash. Develop the right tools – which there is now incentive to do – and you’re maybe looking at an improvement.
Do this right, and everybody wins. Everybody wins.
We have a chance here not just to “plug this one hole,” as the E Pluribus Hugo authors like to say their amendment does. We have a chance to make this whole system just a little bit better along the way.
Wouldn’t that be nice?
This part of a series of posts on the Sad/Rabid Puppy candidate slate-based capture of the Hugo Awards, and resulting fallout.