My Almost Certainly Ill-Advised Proposed Award Voting Process
Posted on September 13, 2015 Posted by John Scalzi 57 Comments
In light of recent events and posts, I’ve been asked, if it fell to me to create a literary award, how I might work the voting process.
My response is, first, I think I would rather pull out my own teeth with pliers than to take on the work and aggravation of helming an award, and this is from someone who was (only very nominally, and insulated by a couple of layers of extraordinarily competent people) previously in charge of the Nebulas. I’m super-impressed with anyone who can handle an award on the front lines. It’s not a gig for me.
Second, if you put a gun to my head and made me do one, or, alternately, put the gun to my head but then promised me that someone else would have to actually run the things so that all I had to do was think up the process, then here’s what I would do, for the process of a popularly-voted award.
1. Categories: Doesn’t matter, think up any category or categories you want, as the process would be the same no matter how many categories there are. I would suggest that every category would have to have a minimum number of initial voters to be considered; say, 500.
2. Who votes: Anyone can vote. Each voter gets their own ID, which can be used only by them. Stupidly obvious attempts to game the system can be disallowed by the poor bastards who actually have to run the system at any step in the process, but for reasons that will become obvious in a second, stupidly obvious attempts to game the system here doesn’t offer much long-term benefit.
3. How the vote works: There are three voting rounds: Nomination, long list, and finalist.
Nomination: Everyone votes for one and only one work (or person, if it’s that sort of category) in the category. The top ten or twelve vote-getters are sent to the long list stage (ties, etc are fine but the goal would be to get number of long list nominees as close to the ideal long list number as possible).
Long List: Everyone votes for up to three works on the long list, none of which can be the single work they originally nominated. That’s right! You have to choose something else in this stage, and hope enough other people like the work you originally nominated to include it among their own selections!
But what if people choose not to make selections in the stage in the hope that their lack of selection of other work will bump up the chances of their preferred work? Well, I would consider making a rule that says failure to participate in this round counts as a point against your original choice’s score in this round — which is to say if you don’t vote in this round, a point is deducted for your original choice’s score in this round (presuming it made the long list at all). You’re better off voting if you want your original selection to make it to the final round.
In this round, the top five or six vote-getters graduate to the final round. Hope your original choice made it!
Finalist: This vote is done “Australian Rules” style, where each voter ranks the works from first to last choice. “No Award” is an option in this round, so if you hated everything in the long list round, this is where you may register your disapproval. The winner is the one which collects the majority of votes, in either the first or subsequent balloting rounds.
Why would I do the voting this way? Because it emphasizes both individual choice and community.
- Picking a single work in the first round makes you really think about what you loved that year and forces hard choices early; knowing that you will have to rely on other people to carry your choice into the final round also makes you think about what you believe others will find worthy.
- Picking an initial single work also avoids obvious slating, while a long list allows for the possibility of a wider diversity of choices for the finalist round.
- Forcing people to make a selection other than their original choice in the long list round makes them consider what else out there might be worthy of consideration, and also again punishes attempts at obvious slating.
- Three choices for a finalist slate of five or six also again cuts down on obvious slating and allows for diversity in the finalist round of voting.
- “Australian Rules” in the final round allows for a consensus vote for the best work in any particular category.
If you want to further reduce any chance of slating you could employ EPH to the long list round, but you get the idea.
Would this work? Got me. And as I noted I’m not going to go out of my way to implement them, because: Ugh, effort. But if anyone wants to try it and see how it works for them, knock yourself out. Could be fun. As long as someone else but me does the work.
Now: Pick it apart in the comments!
If there is an extremely popular work in a given year, it will receive far-and-away the majority of the nominations, and then *nobody will be able to vote for it from the long-list*.
Suggested amendment: only two nominations, and you can’t vote for *both* in the long list (but you can include one or the other in your long-list votes).
Evan H:
“If there is an extremely popular work in a given year, it will receive far-and-away the majority of the nominations, and then *nobody will be able to vote for it from the long-list*.”
Is this a bug or a feature, though?
But how are you going to make sure only the right people vote? You need trust gating or non-fans might get in.
Evan H. Even in close affinity groups (excluding active slate attempts) there will be enough diversity of nomination that the big popular thing will be a fair number of people’s second choice, assuming even a moderate number of voters. It won’t do as well in round two, but something big and popular ought to make it through even with pretty solid land slides otherwise.
Scalzi:
“Is this a bug or a feature, though?”
If it’s a feature, then why let everybody vote? The intent seems no longer to be a reflection of majority popular opinion… If the point is too be snooty, just have a panel of judges (e.g. the Phillip K. Dick award) and be done with it.
Evan H:
“If it’s a feature, then why let everybody vote?”
I think the point would be more, if X was amazingly popular, then some folks who might initially vote X will think, “Maybe I’ll vote for something other than X because I’m pretty sure it’ll get nominated by others and then I can vote for it in the second round,” thus offering a larger diversity of initial choices.
Also, then, what roninkakuhito noted.
My point being that I wonder the nature of the nomination process itself wouldn’t do some correcting for a rush to nominate all the same thing.
You could *add* numbers of nomination choices to the totals of long list votes, while keeping the stipulation that a participant’s long list votes must be different than their nomination choice.
This way, a *hugely* popular nomination won’t be penalized. In most cases, however, the nominated works which make it to the next stage would do so with a relatively small percentage of the total number of nominations: so, in most cases, adding in the number of nominations a work gets wouldn’t make much too much difference.
I’m not sure the “you can’t vote on the long list for one you nominated” is needed or desirable. First, by definition, you’re picking two items you didn’t nominate, so you’re still having to think about what else deserves consideration. Second, if you absolutely believe, “Xzzzy, a life” is the best work released the past year, you’re saying “And you cannot vote for the one you think is the best.”
On the implementation side, the tricky part is making sure that there’s one vote per person. Which is why I wouldn’t want to run this either ;)
I don’t know how others feel, but I would find having to play the “I really want to vote for X, but maybe for strategic reasons I better vote for Y”-game kind of irritating. Also, it’s leading folks to choose their nomination based on cunning-devious-strategy-game-theory, rather than on a basis of “this one is my most favorite thing”.
An additional thought – how will people feel about “This is the best thing done this year, a lot of people agreed so it got on the long list – but they wouldn’t let me vote for it and it didn’t make the finals? WTF kind of screwed up award is this?”
Fascinating. I would not want to be in charge of implementing. Oy the complaining.
Tasha:
Yes. YES.
I agree with Charles R.’s suggestion above. If you add in the number of initial nomination votes that a work received to get onto the Long List into its vote total for the final list, then in essence everyone who nominated it has also voted for it to go on the final list. So, for the Long List, everyone must then consider that list and ask which *other* works are worthy. An insanely popular work will still get on the final list, but there will be a reasonable assortment of other works to compete against it on the final ballot. (This modification will also eliminate a significant source of complaining, a benefit in and of itself)
“You have to choose something else in this stage”
Reminds me of how I look at presentations from biotech vendors (I’m in Pharma R&D): naturally every vendor will say their technology is the best in whatever the domain is, so of course I take their claim to be the best with a grain of salt. But I pay special attention whenever they talk about their competitors: anybody whom more than one vendor mentions as being their strongest competitor is likely to be quite good at what they do.
Any system that forces the voters to think past “this is my favorite” when they make their vote is unnecessarily cruel.
I would think the requirement to vote for some other works or take a deduction on your own nomination would be both horrendously complicated to make work in practice, and also a bad idea since it would encourage people to vote for things they had not read. I mean, it would be nice if they had read everything else and were not blindly voting for their own nom, but that isn’t going to happen. The rule on not voting for your own nomination would be better on its own, would probably be enough 9 times out of 10, otherwise you are just going to get a lot of random votes.
The “can’t vote for your nominee” part reminds me of Half Past Human by T.J. Bass, where you only get one vote, you can’t vote for yourself, and you need two votes to stay alive.
@Donald Brown: I’m pretty sure that would only happen to works that were championed by a relatively small cohort of extremely enthusiastic fans, but are otherwise not well-liked. If they reflect something that a broad selection of the community are likely to think is worth voting for, they’ll be likely to be nominated on the votes of all the people who pick it because they can’t vote for their own fave.
I really like this mechanic, because it means the finalist candidates need to have both (a.) people who are passionate about them enough to nominate in the first place and (b.) an otherwise large community consensus that they’re honestly worth the award. It means of the five or six finalists, none of them will be books that draw a vocal minority but otherwise are generally not liked. This mechanism makes the voting *more* populist, not less.
I’m not just thinking of this year’s focus group, either. A minority of people who are interested in nominating nothing but the most elaborately literary SF would also be diminished in their capacity to advance their preference, if the work is complex to the point of being genuinely inaccessible to casual readers. (Which, like, I’m not saying that kind of work shouldn’t win awards. Just that it wouldn’t likely win *this* award, and that’s something I can imagine people liking about this award as a recommendation list or source of pride.)
That element is also mitigated by the Australian-rules finalist voting, but if we want an award to genuinely reflect the popular sentiment, why not bake it in at every possible layer? (Layer 1: people have to think it should get the award. Layer 2: plenty of other people need to agree that, if not their fave, this is also a reasonable option. Layer 3: at least 50 percent of people need to prefer giving the award to that work over not giving the award out at all.)
If your goal is just to make sure a certain work doesn’t win, can’t you effectively vote against it by nominating it and then not participating in the second round?
Everyone who has every complained about our political system should remember this discussion (i.e., this stuff is hard!)
I think we have reasonably large coverage on the awards front already.
http://www.sfadb.com/Awards_Directory
107 awards, 57 of which appear active (my definition: awarded in either 2014 or 2015).
It may just be easier to let things roll as they are.
I really like the outline of this process — and it has some interesting resonances with the one by which the Yuletide fanfiction exchange is run. (A summary for the gallery: one may nominate up to 3 fandoms whether or not one signs up to participate in the story exchange. Participants must make 3 to 6 unique requests, and must offer to write in at least 4 different fandoms — thus, the rules require you to offer to write a story in at least one fandom you didn’t nominate. Yuletide consistently attracts 1500-2500 participants nowadays, and the annual lists of nominated fandoms regularly run to 3500-4000 or so.)
I am not convinced that a correction for high-scoring Round 1 nominees is necessary. If the system is working as designed, most Round 1 voters will be unable to vote for their Round 1 choice in Round 2, because with only one nomination apiece, most voters’ choices will be eliminated entirely rather than making it into Round 2. The number of Round 2 voters whose Round 1 choice is in fact on the longlist (call this Z) should therefore be relatively small compared to the overall voting pool, and requiring those voters to choose other works should not greatly reduce the support for any given work — especially since the Z value is likely to be split among several of the Round 2 works. (And if it isn’t, then I tend to agree with the idea that the very-high-scoring Round 1 works should be handicapped a little, in order to build a more consensus-friendly Round 3 ballot.)
I like CharlesR’s suggestion as clarified by DavidK44.
Also, I would suggest if you don’t vote in a category at the long list stage, you don’t get to vote in that category in the final stage. So anyone can nominate, but only those who vote at long list stage vote for the finalists.
How to verify unique voters: well, the most likely thing to work is a security key system, and as long as the system isn’t changed year to year, the key wouldn’t have to be either. There’s still the possibility that key holders would try to provide their keys to other parties, but a fairly simple login tracker would catch that (if I can think of it, surely some actual coding type person has invented it).
OR you can just add the votes in the Long List stage to the votes in the Nominating stage to determine what goes into the Final stage, maybe with the Nominating votes worth 1/2 a Long List vote, just to make the process complicated enough to discourage any kind of gaming of the system.
chachai: The problem isn’t about verifying a security key is only used once. The problem is tying a security key to a specific person. With elections (in the US), this is a solvable problem: if you can vote, you have a social security number. You were either assigned one at birth, or you got one with your green card.
There’s no such system for book awards, and any such “invented” system would be woefully easy to bypass (and no one’s handing out their SSN to vote on an award). The Hugos actually have a pretty decent system: if you’ve paid to get in, you can vote; one ticket, one vote. But, as we saw with the “voting scholarships”, it’s easy to bypass if you’ve got the cash to burn, even if you’re there for a *cause* and not as a *fan*.
And even with the SSN verification in place, there have been a few *successful* cases of dead people and pets voting in the US.
So, really, the only way to do it and be *sure* no one’s voting duplicate is if everyone knows everyone who’s voting. At which point, what you have is a panel, not a populist award.
How about a game John?
Create a ‘Shiny Award’ and an ‘Only Fan Boys Award’ (pick your preferred titles) here on Whatever where your readers vote for their favorite works under the rules above as modified by contributor comments and see how it goes? On the Fan Boys Award, add in a number of slate votes using a number of slate voters representative of the total participant count.
And perhaps bump up participation by adding a door prize or donation to a deserving charity if you’re so inclined.
You’re forgetting the part where I said I’d rather pull my teeth out with pliers than administer an award.
@Jeff Wyonch: 107 awards, 57 of which appear active (my definition: awarded in either 2014 or 2015).
Wow. I’m reminded of a joking remark a friend once made; he was part of a new business school that was trying to make a name for itself by doing well in at least one of the dozen or so published rankings: “We want to be one of the 40 schools that are in the top 20.”
More seriously, assuming that there’s some good reason to give books popular awards, it’s worth asking how you want voters to behave and what you’d like them to come away thinking about the result. Wikipedia’s article on tactical voting gives a good idea of the scope of the issues.
Well, it would only be a test, you see….
Um, as an Australian living in an AFL state, can I just say that every time someone says “Australian Rules” I start thinking football, not voting systems? (“Australian Rules” is the Australian shorthand for the football code played by the Australian Football League). Why not just call it preferential counting, and have done with it?
That little terminology note aside, I don’t see any obvious problems with the system as described (aside from the joy of having to explain preferential counting to ‘Murricans[1] once again, because it definitely suffers from the “Not Made Here” problem). However, I’m with our Gracious Host on not wanting to try and administer it. There are easier (and often more pleasant) ways of dealing with masochistic impulses, after all.
[1] ‘Murricans (or “Merkins”, as per afp usage, or “USAliens” as per my favourite term) are the sorts of US citizens who think the world stops at the borders of the continental United States of America, and have to have the existence of the outside world (and its technological advancement beyond stone knives and bearskins) explained to them in short sentences using small words. Every. Single. Time.
Ten second thoughts by somebody who hardly cares about book awards but spends his days doing research on voting technology:
The problem with most forms of “rank-choice voting” or “instant runoff voting” is that people get all twisted up trying to game the rankings. If you’ve heard of Arrow’s Impossibility Theorem (https://en.wikipedia.org/wiki/Arrow%27s_impossibility_theorem), that’s what it’s all about. Once you’ve got more than two choices, there’s no “optimal” voting system, based on a long list of criteria that you might want to optimize.
Salzi’s new idea here (which I do like) is that your choices in one round are precluded from the subsequent round. The devil’s in the details, but this essential idea is very much worth keeping around. You could take that and merge it together with an “approval voting” scheme (https://en.wikipedia.org/wiki/Approval_voting) and something really unusual and interesting comes out.
Round 1: vote for as many or as few candidates as you like. Each gets one or zero votes from every voter. Add them up. The top N of them advance to round 2.
Round 2: vote for as many or as few candidates as you like, except you can’t vote for anything you voted for in round 1. Each gets one or zero votes from every voter. Add them up. The top vote-getter is your winner.
The game theory on this is non-obvious and worth some deeper thought. If you vote *all* your preferences in round 1, you can’t select *any* of them if they make it to round 2. The optimal strategy might well be to vote for all the “unlikely-to-win” candidates and toss a coin for each of the “likely-to-win” candidates. Heads you vote them, tails you don’t. This means your “unlikely-to-win” candidates are most likely to make it to round 2, where they gain the publicity benefit of making it to round 2 (“a finalist for the XXX award!”), while your “likely-to-win” candidates have a 50% chance of gaining your support in the second round, so the broadly popular ones are still quite likely to win.
The oddball downside is that the computer managing the voting system now needs to track every voter and how they voted, and it could well be subject to hacking attacks, malicious vote flipping, etc. Amazingly enough, modern “end-to-end” cryptographic techniques can largely solve this problem. Details are beyond the scope of a blog posting comment, but poking around the literature for “additively homomorphic encryption” will lead the reader into a maze of twisty passages, all alike.
Erh. I like the overall idea, but I do think the “you can’t vote for what you really LURVED in round 2” actually is a bug. And requires way too much thinky. At least you only have to worry about ranked voting and strategery on the final ballot for the Hugos. Having to do it right up front is, as @Jack Lint said, cruel. (And would probably reduce participation, either b/c “But but Book A is the bestest work of ALL TIME!” or simple “Meh, too confusing”.) @DavidK44 has it sussed, I think.
It’s a nasty scary Australian bug, not a cute one like the katydid who waved at you last week.
“I don’t know how others feel, but I would find having to play the “I really want to vote for X, but maybe for strategic reasons I better vote for Y”-game kind of irritating.”
This is a particularly weird problem in light of using “Australian rules” voting in the final round, which among other things is designed to minimise strategic voting as much as possible.
Australia does have a voting system designed to whittle down hundreds of candidates into a palatable handful – if a candidate reaches the quota to secure a place, all the second preference votes are counted, and they all count as partial votes for those second preferences (the value depending on how much surplus they have). However, in this system slates are a feature, not a bug, and Scalzi is trying to design a voting system that both frustrates slates and eliminates strategic voting.
If it truly was Australian rules voting, your best choice would count for six, and you’d get one point for any votes alongside it.
This is a sports joke.
If someone has already pointed this out, my apologies (I didn’t see it). To me, it looks like this type of thing could favour well organised slates.
Let’s assume I am an Infected Canine. I have 4 works I want to fill the slots in a given category. I get all my cultists to send me an email registering their intent to participate; I assign them each a number from 1 to 4.
Each of my cultists is instructed to nominate the work that corresponds to their number, and then vote for the other three. Thus, all the works that I want are all guaranteed to get votes equal to 75% of my cultists numbers. This is not going to be enough to shut out Guardians of the Galaxy, but it will certainly serve to swamp less popular categories. Of course it will still fail to the No Award vote, but so did the unaltered system.
So, what am I missing here? Why wouldn’t this work?
@RSA What I’m getting at is this isn’t a problem that needs to be solved. If anything, it gets worse once you consider all writing awards (the sheer number of awards). One award had a loophole exploited. That award is in the process of closing the loophole. There is enough diversity of awards as it is. I’m thinking case closed.
@RSA I’m not being testy. Thinking about awards administration may be a fun intellectual exercise, but most of the active awards I linked to have had their own controversies, and most have solved them. And these awards run the gamut of juried to open vote.
It’s an interesting thread.
I was thinking it might be interesting to have the Hugos, at least for novels, be a two-year voting process.
First, add some novel categories for various sub-genres. The first year votes would be for those sub-genres.
The second year, the nominees for *the* Best Novel Hugo would be the various sub-genre novel winners from the first year.
@meru in other parts of Australia, we think of it as giving you a point for missing. But I do have to say, it is one of the easiest, most democratic games in the world to play.
* Players may position themselves anywhere on the field,
* Players use any part of their bodies to propel the ball in any direction (except the easiest, throwing it, so Colin Kaepernick is partly bought down to my level),
* If you decide to run with the ball, you have to bounce it (or at least touch it to the ground) every 15 metres / yards,
* If you don’t get rid of the ball when you’re caught running with it, it’s a free kick to the opposition,
* The ball is not round (so Renaldo is partly bought down to my level),
* You get big points for kicking the ball between the big posts, but you also get a point for effort if you miss by a bit..
Which, somehow, gives me an idea for an Award system.
1) Open for nominations
2) Close for nominations
3) All nominations receive an Award.
The only work is 10 minutes coding and someone decent at designing (the coding is a piece of poo, but apparently my design chops are non existent) to get a website up that takes the name of the work, the name of the author, and generates a certificate that that the user prints or saves.
glad to help :)
I also don’t like the bit where you can’t vote for your favorite in the second round.
But it’s a sneakily likeable proposal, all the same.
Next, John should name his award and design a trophy (I’m thinking we can maybe trick him into doing all the work, after all?)
The reason to do any kind of fannish activity is simply to meet an unmet need. The question any new award has to answer is, what does your award do that isn’t already being done (and/or done better) by existing awards? IN other words why does anyone care about a new award?
Mathematically, any round of selection that isn’t based on some sort of preference can be gamed by a sufficiently large number of people. Using your suggestion as an example: the Vile Faceless Minions (VFM for short, it’s how some of them sign their emails) pick six works and divide their nominations evenly across those; if their block nomination number is greater than whoever is the seventh choice on the first round, they have six titles coming to the long list. They then divide into two groups, numberically (they have assigned numbers), and anything making the final ballot has to beat half their number. It’s steep obstacle. The VFMs don’t vote on any consideration of what they’ve read; the point is to swamp the ballot by voting and nominating in lockstep, so they have to follow orders from their leaders in order to impact the result at all. The voting population has to be large enough and/or have some effective mechanism to prevent an organized clique from swamping the ballot.
@Jeff Wyonch: What I’m getting at is this isn’t a problem that needs to be solved.
Yes, I gathered that. I think you’re right.
My problem with having a three-stage nomination is the timing involved. How many months do you want to give people to have a reasonable chance of reading all the works in the long list that they think they want to read? How long time should you give people from the appearance of the shortlist package (if any) to the deadline for voting?
Given that it takes a while to negotiate that package, there will be a gap between deadline for the voting of the longlist and the appearance of the shortlist package.
Even if nomination ends on December 31st, I can’t make the timeline add up. Unless we either discard the Hugo package or make it into a two year eligibility award. (That is not automatically a bad idea.)
Jeff Wyonch: I think we have reasonably large coverage on the awards front already.
It’s interesting that that list does not include “Worldcon Guest of Honor” under the Career Awards, since it’s a once-in-a-lifetime thing that is considered as prestigious as winning a Hugo.
How long is the long list and is the presumption that you read everything on it?
The part about not picking your own 1st round choice just doesn’t strike me as all that workable, and encourages weird tactical voting behavior.
1 nomination means that voters have to decide between the obvious popular pretty good choice and the last known but favorite choice.
The most obvious strategy for the first round is a straight up vote swap; find someone who will nominate the work you want to vote for and yourself nominate the work they favour. Anyone determined to vote tactically in this way won’t have to think too hard. They will have to talk to someone who has a different favourite to them.
(Next I’ll set up the nomination exchange website, followed by a vote futures market)
H’mmm…I’d rather have a top-5 list, with 5 “points” awarded for your top book and 4 for the second, et cetera.
Much simpler than the three-step proposal, and less likely to cause confusion, which is bad because it can discourage people from participating. Also, it allows point thresholds for nomination; if a specific work gets more than, say, 5% of the total available points, it gets on the ballot!
Just vote on your top 5, then rank your top 5 from the ballot. Winner is the one that gets the most points from the ballot vote.
Categories, eh, I’d add best sub-genre for a couple of sizes, winners to be decided by yours truly for the sake of humor, for example Best Mil-SF (obviously The End of All Things would win), Best Fantasy (Stormlight 3 will win when Sanderson finishes it), and Best Blows Floored’s Mind Away (Lock In or whatever Sanderson writes next). And of course there would have to be a humor category, Best Parody of a Bigot or Troll. That would be hilarious.
(not completely serious, but a category for Best Parody would be fun).
Almost 24 hours, and while clearly there is agreement on “Administration: do not want” and “hard stuff is hard,” not once, NOT ONCE, has anyone harped on a name for the thing.
Slackers.
I’m not sure it helps against slates… might even hinder, in fact. Given that a slate is, by definition, organized, you just have to pick four nominees and divide up your minions accordingly.
Let’s say your chosen works are Abbasids and Anacondas, Big Barney’s Betelgeuse Bandits, Chronicles of a Catalonian Cataphract and Death in Delta Delphinis. You split your minions into four parts, each one voting for A, B, C or D in round 1.
Come round 2, your minions each vote for the three they didn’t pick in round 1. (Assuming you have enough minions to get all four onto the longlist, or you have a couple of backup candidates to go for if they don’t.) You then have four different groups of voters going for (B, C, D), (A, C, D), (A, B, D) or (A, B, C). Instead of just one slate, you’ve got four closely related ones. Some of the bright sparks who crunched the numbers for EPH can probably work out what happens next (not me, I only did “O” level maths), but I can’t imagine it’s that good, from the slate-neutralizing point of view. The expert psephologists can fill us in. (I just love using the word “psephologist”, it gives me a warm feeling inside.)
It would be interesting to see this modelled, as some have done with EPH.
Whatever happens with this, I hope the award itself is a bronzed pair of pliers surrounded by a pile of teeth.
“The Toothpullers”
Sadly for Scalzi, if I only had the two choices, I’d pick him administrating awards over the tooth pulling.
I would have the top two vote-getters (maybe only the top one) in the nomination stage automatically go to the final round. They would essentially get a bye for the long list. This should avoid the game-theory problems set out above. Alternatively, you could have any nominee getting more than a certain percentage of the nomination vote automatically move to the final round – if no nominee gets that much early love, then everything goes as you planned.
This way, the overwhelming favorite is going to make the final list and everyone can spend their voting energy on looking all other candidates without worrying about whether superpopularfantsticwonderstory will be on the final list.
Dear John,
I’m with the consensus on this. Not letting someone vote for their nominee in the second round is overthinking the situation and encourages voters to overthink the situation. It inclines to distraction from the task of trying to pick the work one genuinely likes best, and except in some very unusual corner cases won’t make a difference in the final outcome. Also, a Conspiracy can game it just as easily as a straight up vote. So, KISS.
I do particularly like the idea of being able to make only one nomination. Other than ’cause it’s a game, really, I’ve never understood why the Hugos let one nominate a whole bunch of works. You should nominate the work you like the very best, that you think is the very most deserving of winning. Nothing else. In the unlikely event that you really think two or three works are genuinely equally meritworthy, well… if the general voting populace agrees with you on that, those 2-3 works are all going to get a whole lot of nominating votes and they’ll make the ballot.
And.. if the voting populace doesn’t, your alternate candidates didn’t stand a chance in the final vote, anyway. Better luck next time.
Yeah, there are weird corner cases where it could matter. Voters’ Paradox. There always will be. But it ameliorates one known bug, which is the ease with which a Conspiracy can dominate a multiple-nominee ballot.
pax / Ctein
These are all tongue firmly in cheek…
Fine… The “Do Not Want Award”? Harpy Award? “You Must Game This Award To Win It”?
“Second Choice Award” “ATGNWT Award”
@JJ The folk who run that site (it was previously run by Locus) are probably only considering awards that issue a trophy, certificate, or cash award. You can always bring this up with them.
Here’s a brand new award from the fine folks in Nashville that just set a list of categories, a huge list of nominees, and anyone with a web browser can vote.
OGH is up for one! (Against Stephen King; good luck there, John!)
http://www.killernashville.com/silver-falchion-readers-choice-awards-best-novel-speculative-horror-science-fiction-fantasy/
A perusal of the other categories reveals names like Vandermeer, Rusch/Smith, KJ Anderson, Andy Weir, etc. There’s even kids’ categories!
I’m finding I read kind of a lot of mystery last year.
If the people running The Hypothetical Award are trying to prevent people from scamming the award by voting multiple times, the simplest and lowest-budget solution I can think of is to have everyone cast their votes through a Facebook app.
Facebook tries to prevent the same person from maintaining multiple accounts—having each account correspond to one and only one real-life human being is essential to their business model—and while they don’t do a perfect job of it, I don’t expect a group of volunteers is going to do any better.
(It’s true that there are still two or three people on the planet who don’t have Facebook accounts, but if you’re trying to create The Hypothetical Award From All Of Science-Fiction Fandom, I think fans who do have Facebook accounts constitute a representative subset of fans as a whole.)