In the brief downtime between announcing the Hugo final ballot, and getting voting under way (which will be Real Soon Now), I reflected that the two disqualifications and two withdrawals from the ballot this year seemed rather low by recent standards. So I looked into the records, and found indeed that of the seven years that I have been involved with running the Hugos, only one had fewer such cases – two were disqualified, and one declined, in 2021, otherwise a really crazy year for Worldcon.

(For these purposes I’m counting a disqualification as any exclusion of an otherwise valid nominee by the administrators under their interpretation of the rules. This includes the various permutations under the Best Dramatic Presentation categories, and also the bad decisions made and published by the Chengdu Worldcon team in 2023.)

The proliferation of withdrawals and disqualifications is a recent phenomenon. I have access to the nomination statistics for 1980 and 1996, and for every year since 1998. From 1998 to 2002, and again in 2007, there were no disqualifications or withdrawals from the Hugo ballot at all, and in the four intervening years there was only one each time. (Ted Chiang, Terry Pratchett and Neil Gaiman declined fiction nominations in 2003, 2005 and 2006, and there was a disqualification in the Best Semiprozine category in 2004.)

One potential finalist was disqualified in 2008, and two potential finalists declined nomination in 2009, 2010 and 2011; and since then there have been at least three withdrawals and/or disqualifications each year. The high water mark was, infamously, 2023, where (according to the official statistics) twelve potential finalists were disqualified and another three declined nomination, though evidence suggests that votes for many more Chinese nominees were removed from the system at an earlier stage, effectively disqualifying over twenty of them without making it public.

The second highest total for withdrawals and disqualifications was the previous year, 2022, when I was Deputy Administrator. We disqualified seven potential finalists that year, four of them in the Best Editor, Long From category (where another potential finalist withdrew); a unique issue at that time was the blockage to global supply chains caused by the pandemic, as a result of which a lot of 2021 publication schedules slipped, though in my view it also shows the difficulties of voter awareness of the editing process.

The only other ballots that saw as many as seven disqualifications were the 1939 and 1941 Retro Hugos. 1939 (awarded in 2014) saw a lot of eligibility confusion, and in 1941 (awarded in 2016) three of the disqualified potential finalists had had sufficient support to qualify in both Dramatic Presentation categories, and of course they could only be on the ballot in one, and therefore were disqualified from the other.

Among the Hugos, the Best Dramatic Presentation categories have generally had the most disqualifications, largely thanks to the rule (or custom) preventing entire TV series and individual episodes of that series appearing on the same ballot. Thirteen BDP Short Form and five BDP Long Form nominees have been disqualified by administrators in the years that those categories have existed, though in many of these cases at least part of the material disqualified in one category appeared on the ballot in another.

The other category with a lot of disqualifications is the Astounding Award, previously the Campbell Award, where there have been seven disqualifications over the years where I have data (including one each in 1980 and in 1996). Sometimes voters (and indeed writers themselves) are uncertain as to when a writer’s career actually started.

The only disqualification for Best Fan Writer on record was the incomprehensible decision to exclude Paul Weimer in 2023; it’s rather difficult to see how anyone who has published anything fannish in the year of eligibility could be ruled out in that category. Apart from Retro Hugos, nobody has ever been disqualified in Best Fanzine or Best Novella, at least in the years for which I have data; nor for the Lodestar, which is also a recent innovation and whose criteria again are broad. Best Game or Interactive Work is the only category where there has not yet been either a withdrawal or a disqualification, but since it has only been going for two years, there is plenty of time…

The largest number of voluntary withdrawals of finalists who would otherwise have qualified numerically is six, in 2016. There were five withdrawals in 2015 and also last year, 2024. As noted above, the last year in which there were no withdrawals from the regular Hugos was 2013.

Seven finalists for the Best Novel and the Best Editor, Long Form categories have withdrawn from the ballot. Pro Artists have declined nomination five times, and Fan Writers and authors of both Novellas and Novelettes four times each, in the years where I have full data.

We have yet to see a voluntary withdrawal in the Best Graphic Story or Comic, BDP Short Form, Game / Interactive Work, and Editor Short Form Hugo categories, or for the Astounding Award or its predecessor, as far as I know. The first and so far only withdrawal from the Lodestar was this year, the first and only withdrawal from Best Related Work that I know of was last year, and the the first only withdrawal from BDP Long was in 2023.

There are two very striking shifts in the numbers. Up to 2011, there were an average of 0.375 disqualifications each year. Since 2012, counting the regular Hugos only, there has been an average of 4.00 disqualifications each year. It’s an abrupt change.

The shift in the number of withdrawals is a little later. Up to 2014, the average was 0.73 per year. Since 2015, the average number of withdrawals from that year’s Hugos is 3.4.

The five rounds of Retro Hugos run between 2014 and 2020 saw no withdrawals at all, hardly surprising in that few of the nominees were in a position to accept or decline nomination, but there were an average of 4.4 disqualifications each year.

(Not that it is a significant difference, but the average number of withdrawals in 2017, 2019-22 and 2024-25, the years where I was personally involved with administering the nominations, is lower – 2.43 rather than 3.4 – and so is the average number of disqualifications – 3.57 rather than 4.00 – but I think this simply shows that the two big years for withdrawals were just before my time, and also I fortunately was not involved with the massive number of disqualifications in 2023.)

I think we are seeing a couple of different effects here over time. Taking withdrawals first: this had never been a huge factor in the Hugos until the Puppy years, when (as noted earlier) a record number of potential finalists declined nomination in both 2015 and 2016. Perhaps one of the lasting effects has been that nominees now feel more comfortable about saying no in general. Also, the aftermath of Chengdu drove the number of withdrawals up again – two of the five in 2024 were directly related to the previous year’s events.

(Kathryn Duval has pointed out to me in conversation that it’s also possible that Hugo administrators in the olden days did not need to be as diligent in chasing nominees for consent as we have been since she and I first administered the awards in 2017. That perhaps is another effect of the traumas of 2015/2016.)

The massive increase in disqualifications since roughly 2012 has several causes. The biggest chunk of disqualifications has been in the Best Dramatic Presentation categories, starting from the year that the entire first series of Game of Thrones was on the ballot, and usually because of a conflict of nominations between the two categories; I have written before about this. And I noted earlier that the special circumstances of the pandemic hit Best Editor, Long Form in 2022.

The constitutional criteria, which are complex in some cases, must also be a factor. The Astounding/Campbell rules are somewhat arcane. The rules in the Artist categories are frankly obsolete. And have you ever had to explain the concept of a Semiprozine? (In Korean?) It all causes a lot of head-scratching for us administrators – it’s not surprising or blameworthy that voters can get it wrong. And the more categories that are added, the greater the opportunity for everyone to make mistakes.

But the other big change, one that almost exactly matches the explosion in the number of disqualifications, is the impressive and welcome surge in the number of voters. I don’t think it is as widely appreciated as it should be that the numbers participating in Hugo voting shifted abruptly upwards in 2009-2011, and now show no sign of declining to their previous level. Before 2009 there had only once been more than 1500 votes on the final ballot, and never been more than 800 voters at the nominations phase. Since 2011, only one year (2023 / Chengdu) has seen less than 1800 final ballot votes (peaking at 5950 in 2015, the first Puppy year), and the lowest number of nomination votes cast was 1249 in 2021 (peaking at 4032 in 2016, the second Puppy year).

Probably the biggest single factor here is the Hugo Voter Packet, which gives hundreds of dollars / pounds / euros worth of books to voters who buy a WSFS membership. It started in 2009 and was really integrated into Worldcon marketing from 2011, almost exactly matching the expansion in participation.

But I think that there was also an effort – perhaps it is too much to call it a campaign – by many people, perhaps in reaction to the 2007 ballot which included only one work of fiction by a woman, to broaden the appeal of the Hugos and make them more diverse. This is a Good Thing. The Puppy argument that the Hugos were locked in a vicious circle of declining participation and political correctness was precisely backwards: by the early 2010s, Hugo participation was rising, not falling, and this was adding some very welcome and needed diversity to the ecosystem.

The effect has been to bring in a cohort of voters who are less invested in some of the older (indeed, oldest) categories, as fan culture itself is de-emphasising the traditional channels. Twenty years ago, in 2005, 546 nominating votes were cast in the Hugos, and a nominee needed 20 votes to get onto the Best Professional Artist ballot, 36 for Best Semiprozine, 24 for Best Fanzine, 30 for Best Fan Writer and 26 for Best Fan Artist. This year there were 1338 nominating votes, almost 2.5 times more than in 2005, but the effective thresholds to qualify for those five categories are the same or lower: 14 for Best Professional Artist, 38 for Best Semiprozine, 25 for Best Fanzine, 27 for Best Fan Writer and 16 for Best Fan Artist. (Though of course there are now six finalists per category rather than five.)

Analyzing the historical levels of participation in each category in depth is for another blogpost (and maybe someone else will do it before I do, which is fine by me). But I think it’s clear that in a number of categories, the Hugo electorate of today is broadly less invested than the Hugo electorate of twenty years ago, and it is therefore more likely that well-known but unwilling or ineligible nominees will be chosen.

As an administrator, I always feel a bit sad and uncomfortable when removing any nominee from the ballot. Most people’s votes are cast in good faith, and they should in general be respected. At the same time, the nominees themselves have the absolute right to choose whether or not to participate in the Hugos; and the rules are there for many reasons (mostly good reasons) and need to be implemented to maintain the integrity of the process. So when it has to be done, it has to be done.

I think I’ll leave it at that. If you want to play with the data yourself, I’ve put it in a Google sheet here.