Hugo Voting: Let’s Look at the Record Yet Again

Posted on by
by Shieldforyoureyes Dave Fischer – Own work, CC BY-SA 3.0

By Jo Van Ekeren:

I’ve spent the last couple of years exhuming statistics and ephemera about the Hugo Awards from various sources, including old Usenet posts on Google Groups, old fanzines, archived con websites, and various historical documents which have been scanned and made available online (and I give my thanks to those of you who have been making those archiving efforts, especially Joe Siclari, Edie Stern, Mark Olson, and Bill Burns).

I’ve managed to resurrect full or partial statistics for around 23 additional years beyond what was posted at TheHugoAwards.org. A few years have already been posted there, and I will be gradually rolling out the rest of them over the next few months as I get them formatted into readable documents.

This post is an expanded and updated version of earlier statistical analyses by George Flynn:
Hugo Voting: Let’s Look at the Record by George Flynn [1988]
Hugo Voting: Let’s Look at the Record (Again) by George Flynn [1999]
and of an update by Jed Hartman which pulled in some additional years and electronic vs. paper voting numbers:
Hugo stats: numbers of nominating ballots by Jed Hartman [2018]

I’ve updated it with Site Selection ballot numbers, Advance Membership numbers, and Hugo participation percentages for 2000-2019, plus Retro Hugo data, as well as showing the difference between the number of categories which were on the nominating ballot versus the number of categories which had sufficient participation to be on the final voting ballot.

I’ve got source citations for all of the numbers included here. A lot of the information came from documents on The Hugo Awards, Fancyclopedia 3, FANAC, eFanzines, the SMOFS Long List, old Usenet posts on Google Groups, and the Wayback Machine. If you have questions about where one of the numbers came from, you can message me here.

You are welcome to link to the full Google document — and certainly can make a backup of it if you wish — but please be aware that I expect it to continue to change as more bits of information become available.

Please do report to me any errors or omissions you might notice, either in the comments on this post, or by submitting a message here.

What does the most recent data about Hugo nominators and voters tell us?

  1. Tracking of the electronic vs. paper nominations and votes, at the turn of the century, was helpful in evaluating the amount of electronic uptake by Hugo voters. That hit 99% in 2011, and has remained there ever since. Now this comparison tracking is chiefly of interest in noting how many remaining members are either unable or unwilling to nominate and vote electronically.
  2. From 1989 through 2007, participation in the final ballot was consistently under 20% of the Advance membership (those eligible to participate in voting). In 2008, both overall membership numbers and Hugo participation began to rise steadily. It is likely that common acceptance and the ease of the ability to nominate and vote electronically contributed significantly to this. In addition, 2008 was the first year of the annual Hugo Voter Packet – containing finalist works which were not otherwise available for free – and this has also likely contributed to the rise of member numbers and of Hugo participation among members.
  3. The ratio of Supporting to Attending members has also steadily risen in the last 10 years, and while some of this can be attributed to the Puppy campaigns of 2015-2016 as well as to fans from the U.S. being unable to attend overseas Worldcons in London and Helsinki, it seems clear that access to a large number of free works in the Hugo Voter Packet is also a contributing factor. Percentage of eligible advance member participation in the Hugo Awards is now at an all-time high, at 40% to 50% of the eligible membership.
  4. Site Selection, which has remained a mail-in or on-site endeavour, has seen somewhat of a decline in participation in the last 10 years. This is likely due to having only one bid site in many of those years, but possibly also somewhat due to people who previously voted for both Hugos and Site Selection by mail in the past now only voting for the Hugos online. This is not likely to change unless and until it becomes common for bidcoms to be willing to have electronic voting for Site Selection.

Hugo Voting: Let's Look At The Record Again (1971-1999), by George Flynn



Over
seas


Year


Worldcon             


Location		[1]
# of
Cate
gories
Valid
Nominating
Ballots
Valid
Final
Ballots		[1]
# of
Cate
gories
Site-
Selection
Ballots
Advance
Member
Count		Final
Hugo
Vote
Ratio
197129 - Noreascon IBoston93437329—1,60045.8%
197230 - L.A.Con ILos Angeles927055092561,50036.7%
197331 - Torcon IIToronto, Canada11350708113752,20032.2%
197432 - Discon IIWashington DC12?930126452,60035.8%
O197533 - Aussiecon OneMelbourne, Australia12267600125281,88031.9%
197634 - MidAmeriConKansas City124861,595129933,60044.3%
197735 - SunConMiami Beach12500800128842,80028.6%
197836 - IguanaCon IIPhoenix135401,246131,1544,20029.7%
O197937 - Seacon '79Brighton, UK134671,160139204,12628.1%
198038 - Noreascon TwoBoston135631,788131,5495,44732.8%
198139 - Denvention TwoDenver124541,247121,6804,52927.5%
198240 - Chicon IVChicago126481,071121,1195,00021.4%
198341 - ConStellationBaltimore126601,322127295,50024.0%
198442 - L.A.con IILos Angeles135131,467131,3686,74021.8%
O198543 - Aussiecon TwoMelbourne, Australia13222443135272,19920.1%
198644 - ConFederationAtlanta135681,267131,863 (’88)
1,276 (’89)		5,400
[6]		23.5%
O198745 - Conspiracy '87Brighton, UK13567990131,3734,95320.0%
198846 - Nolacon IINew Orleans144181,178141,4554,72125.0%
198947 - Noreascon 3Boston13539980131,6366,10016.1%
O199048 - ConFictionThe Hague, Netherlands14291486141,0883,41814.2%
199149 - Chicon VChicago133521,048132,0865,12620.4%
199250 - MagiConOrlando14498902142,5095,29717.0%
199351 - ConFranciscoSan Francisco15397841141,2825,83414.4%
199452 - ConAdianWinnipeg, Canada14649491141,4394,38811.2%
O199553 - IntersectionGlasgow, Scotland14477744141,5544,90015.2%
199654 - L.A.con IIILos Angeles14442939141,0646,00015.7%
199755 - LoneStarCon 2San Antonio13429687131,4674,40015.6%
199856 - BucConeerBaltimore13471769132,1685,13115.0%
O199957 - Aussiecon ThreeMelbourne, Australia13425438138202,42518.1%

Hugo Voting: Let's Look At The Record Yet Again (2000-2020),
by Jed Hartman and Jo Van Ekeren
Over
seas		YearWorldcon             Location[1]
# of
Cate
gories
Valid Nominating Ballots
Valid Final Ballots
[1]
# of
Cate
gories		Site-
Selection
Ballots		Advance
Member
Count		Final
Hugo
Vote
Ratio
TotalElec%ElecTotalElec%Elec
2000
[2]		58 - Chicon 2000Chicago1340713031.9%1,07147544.4%131,6985,26220.4%
200159 - Millennium PhilconPhiladelphia1349517836.0%1,05028226.9%132,0945,01320.9%
2002
[3]		60 - ConJoséSan José1462637159.3%92469775.4%141,0344,42220.9%
2003
[4]		61 - Torcon 3Toronto, Canada1473877647861.6%141,4814,20418.5%
200462 - Noreascon 4Boston1456736664.6%1,093141,6865,61319.5%
O200563 - InteractionGlasgow, Scotland1554643679.9%68455280.7%15[7]4,16916.4%
200664 - L.A.con IVLos Angeles1553343481.4%71160084.4%141,5614,12817.2%
O2007
[5]		65 - Nippon2007Yokohama, Japan1540934083.1%589159024,69112.6%
200866 - Denvention 3Denver15483895158264,06222.0%
200967 - AnticipationMontréal, Canada167991,0741,04096.8%167633,81228.2%
O201068 - Aussiecon 4Melbourne, Australia168641,094165262,89837.8%
201169 - RenovationReno161,00699298.6%2,1002,08699.3%167604,68844.8%
201270 - Chicon 7Chicago171,1011,922179325,21836.8%
201371 - LoneStarCon 3San Antonio171,343132999.0%1,848171,3484,46841.4%
O201472 - Loncon 3London, UK171,923188998.2%3,5873,57199.6%177788,58041.8%
201573 - SasquanSpokane172,122211999.9%5,9505,91499.4%172,62510,32157.6%
201674 - MidAmeriCon IIKansas City174,032401599.6%3,130171,3216,17450.7%
O201775 - Worldcon 75Helsinki, Finland182,464245899.8%3,3193,31599.9%181,2277,67243.3%
201876 - Worldcon 76San José191,813179599.0%2,8282,81099.4%197266,39344.2%
O201977 - Dublin 2019Dublin, Ireland201,800179799.8%3,0973,08999.7%208806,000?51.6%
O202078 - CoNZealandWellington, New Zealand191,584158299.9%2,221221699.85195874,48649.5%

Hugo Voting: Let's Look At The Record for the Retro Hugos

Year
Held		Retro
Year		Worldcon             Location[1]
# of
Cate
gories
Valid Nominating Ballots
Valid Final Ballots
[1]
# of
Cate
gories
TotalElec%ElecTotalElec%Elec
1996194654 - L.A.con IIILos Angeles13111---605---10
2001195159 - Millennium PhilconPhiladelphia121304836.9%86215718.2%10
2004195462 - Noreascon 4Boston131319673.3%84110
2014193972 - Loncon 3London, UK1623322697.0%1,3071,29599.1%10
2016194174 - MidAmeriCon IIKansas City1648147598.8%86911
2018194376 - Worldcon 76San José1720419294.1%70368897.9%9
2019194477 - Dublin 2019Dublin, Ireland1821721498.6%83482699.0%11
2020194578 - CoNZealandWellington, New Zealand1812011797.5%52151699.0%12

No.Footnote Explanation
[1]Number of categories includes the Hugo Awards, the Astounding Award, the Lodestar/YA Award, and any other special categories or awards announced that year. Discrepancies between total nominating categories and total voting categories are the result of categories with insufficient nominations being dropped from the final ballot.
[2]Chicon 2000 received 1,101 Hugo ballots, of which 475 were electronic ballots and 626 were paper ballots. 30 ballots were invalid, which left 1,071 valid ballots. It is unclear how many of the 30 invalid ballots were paper vs. electronic.
[3]ConJosé received 940 Hugo ballots. There were 697 were electronic ballots, 226 paper ballots, and 17 fax ballots. 16 ballots were invalid, which left 924 valid ballots. It is unclear how many of the 16 invalid ballots were paper vs. electronic vs. fax.
[4]Torcon 3 received 805 Hugo ballots, of which 478 were electronic ballots and 327 were paper ballots. 29 ballots were invalid, which left 776 valid ballots. It is unclear how many of the 29 invalid ballots were paper vs. electronic.
[5]The number of final Hugo ballots for Nippon 2007 is unknown. The quoted figure is the number of Novel ballots / 80%, which is the average percentage of final ballots cast for Novel during that stretch of years.
[6]Site Selection went from 2 years to 3 years in advance
[7]Site Selection went from 3 years to 2 years in advance

Discover more from File 770

Subscribe to get the latest posts to your email.

20 thoughts on “Hugo Voting: Let’s Look at the Record Yet Again”

  1. Wow, Jo! That’s alot of work. Thank you to you and the other data gatherers.

  2. Seconding @Ultragotha! Amazing work–and tracking of history of amazing work in the stats of voting.

  3. Hunh, thought I had posted a comment about the hard work you’ve done, Jo. Sorry about that. Thank you!

  4. I am very grateful for Jo’s work on this, and delighted that she agreed to join the Hugo Awards Marketing Committee as part of this work.

  5. Nice work.

    On problem I think I see is that there are two columns, both labeled # of Categories. For the Retros, they have different data.

  6. There’s one for number of categories in nominations, and one for number of categories on the final ballot. (See footnote 1.)

  8. Thank you so much for all the hard work of data crunching and research.

  12. I appreciate the thanks, and I have done quite a bit of work putting together Hugo statistics, but I would like to make it very clear that the first table above is one created by George Flynn back around 2000, and the second table is one created by Jed Hartman in 2018, which was then greatly expanded by me. I’ve asked for titles to be added to those tables to make that more clear.

  13. I see that it’s been claimed on Twitter that my post fails to acknowledge efforts which have made the Hugo Awards more accessible.

    Apart from the enhanced promotion and encouragement to participate in the Hugo Awards in which hundreds of Worldcon fans have engaged in recent years — which is visibility, rather than accessibility — if there are other efforts to make the awards accessible which I have indeed failed to acknowledge, I hope that someone will let me know what they are, so that I can rectify that omission.

  14. Greg Hullender: One problem I think I see is that there are two columns, both labeled # of Categories. For the Retros, they have different data.

    P J Evans: There’s one for number of categories in nominations, and one for number of categories on the final ballot. (See footnote 1.)

    It’s not too surprising that there have been a lot of categories cancelled in the Retro Hugos. Long Form Editor would be very difficult to do in retrospect, and I’m not sure that there was anything which fit the current definition of Semiprozine back then. And I don’t know whether there would have been such a thing as Fancast — I suspect the radio programs would all have been considered Procasts.

    In the regular Hugos, it looks like a category was cancelled in 1993 (Best Translator) and in 2006 (Best Videogame).

  15. One thing that would be interesting to see, if statistics are available, is change in how many categories people voted/nominated in as average. I’d guess there would be a big change in voting after the Hugo packet started rolling out, but it would be interesting to see if the numbers would support that.

  18. JJ, as far as I can tell, any numbers for Retro Hugos are either missing or included as part of the others.

  19. Talking of Retro Hugos, does anybody know if CoNZealand is planning to do them? Cause I’ve been considering setting up an open recommendation spreadsheet for the Retro Hugos along the lines of Renay’s Hugo Spreadsheet of Doom.

  20. One thing that would be interesting to see, if statistics are available, is change in how many categories people voted/nominated in as average. I’d guess there would be a big change in voting after the Hugo packet started rolling out, but it would be interesting to see if the numbers would support that. — Hampus Eckerman

    There is not, of course, any way to correlate which people voted in which categories in one year versus another year — in other words, how many people had not voted in a given category in the past, but started doing so after receiving the voter packet became common practice.

    However, we are able to track participation in each category as a percentage of total participants (how many people submitted a ballot which included at least one nomination/vote in a particular category), and I’ve done a spreadsheet which shows that.

    Hugo Nominating and Voting Stats 1964-2019

    If you scroll to the far right and down on this spreadsheet, you can see how percentage participation has changed by decade. (This sheet is set up such that the 2019 results will be the last year in this decade, so the nominating and voting stats are only for 9 years right now in the current decade.)

    Voting participation in Novel, Novella, Novelette, Short Story, Related Work, Fanzine, and the Campbell have gone up from 2 to 6 percent in the last 10 years (since the packet was insitituted). Voting participation in Dramatic Presentation Long Form, Editor Long Form, Professional Artist, and Fan Writer have gone down a few percent, and voting participation in Editor Short Form and Semiprozine have gone down significantly in the last decade.

    Categories which have received the most nominating participation in recent years are, in order:
    1. Novel
    2. Dramatic Presentation Long Form
    3. Series
    4. Short Story
    5. Novella
    6. Novelette
    7. Dramatic Presentation Short Form
    8. Related Work

    Categories which have received the most voting participation in recent years are, in order:
    1. Novel
    2. Dramatic Presentation Long Form
    3. Novella
    4. Short Story
    5. Novelette
    6. Dramatic Presentation Short Form
    7. Series
    8. Related Work

Comments are closed.