Chaos Horizon

2015 Hugo Stats: Initial Analysis


Some preliminary numbers from the stats:

There were 5,950 total voters.

Rabid Puppies looks to be a little less than 10% of that. Comparing Vox’s recommendations in swept categories to the first round results (i.e. those who probably followed the suggestions):
Best Novella (Wright, “One Bright Star”): 556 (took fourth with 1050 eventual votes)
Best Short Story (Rzasa, “Turncoat”): 525 (took fourth with 1064 eventual votes)
Best Related Work (“The Hot Equations”): 595 (took second with 973 votes)
Best Editor, Short Form (Vox Day): 586 votes (took fifth with 900 votes)
Best Editor, Long Form (Toni Weiskopf): 1216 (obviously more people voted for her)
For reference:
Campbell (Eric Raymond): 489 votes (eventually took 4 with 748 votes)

I think those numbers are pretty clear: 556, 526, 595, 586. That’s the Rabid Puppy range, or at least those who closely followed VD’s suggestions. Should we call it 550-525? (Raymond was close to that range, but fewer people make it to the bottom of the ballot to vote the Campbell). I think the number who voted VD for Best Editor is probably closest to the actual number.

Initial Rabid Puppy Estimate: 550-525
That makes around 10% of the total vote, which is in line with what I expected.

I think we can use these same numbers to grab a “Sad Puppy” initial estimate, or at least the most hardcore Sad Puppy supporters (who voted all the Rabid/Sad Puppy picks above No Award). If look at the second set of numbers I gave, that’s 1050, 1064, 973, 900. Even Vox Day picked up 320 more voters from No Award. Is it fair to say those are the Sad Puppies? We’d get 1064-900 for the total Puppy vote. It looks to me like 500-400 Sad Puppies. I want to be looser here, because maybe some Rabid Puppies didn’t follow the VD suggestions, and maybe some other voters drifted in and voted for these texts.

Initial Sad Puppy Estimate: 500-400
We’ll have to refine that number over the next few days.

We can use the same swept categories to estimate the “No Awarders”: the people who voted for No Award over every Puppy pick. Here’s those numbers:

Best Novella: 3459 No Awarders
Best Short Story: 3053 No Awarders
Best Related Work: 3259 No Awarders
Best Editor, Short Form: 2672 No Awarders
Best Editor, Long Form: 2496 No Awarders

One more interesting number:
Best Novelette: 1732 No Awarders (remember, Heuvelt was a non-Puppy pick! In the final pass, he beat No Award 2618 to 2078).

That 3459, 3053, and 3259 number are pretty close. That seems the max No Award number: people who couldn’t stand any Puppy Pick. When there were more valid choices, such as in the Editor awards, No Award was still picking up 2600-2500 votes. In a case where a category was almost swept, the number was close to 2000. So I’m calling the No Awarders at 3450-2500. That’s a huge number, over 50% of the total pool.

I’m stunned at the 2500 No Awarders in the Editor categories; there were some mainstream, decent editors on that list. If 2500 people were voting No Award on that, that’s out of principle. So here’s how I’m estimating:

Initial Estimate of No Awarders Who Voted No Award out of Principle: 2500.
Initial Estimate of No Awarders Who At Least Considered Voting for a Puppy Pick But Eventually Didn’t: 1000.

Those numbers will clearly need some work. So that leaves us:
Core Rabid Puppies: 550-525
Core Sad Puppies: 500-400
Absolute No Awarders: 2500
Primarily No Awarders But Considered a Puppy Pick: 1000
That sums up to 4600 voters. We had 5950, so I think the remaining 1400 or so were the true “Neutrals” or the “voted some Puppies but not all.”

UPDATE, 8/25/15: By looking closely at the Best Novel category, I’ve updated my estimate, breaking the Neutrals into two categories:

Core Rabid Puppies: 550-525
Core Sad Puppies: 500-400
Sad Puppy leaning Neutrals: 800-400 (capable of voting a Puppy pick #1)
True Neutrals: 1000-600 (may have voted one or two Puppies; didn’t vote in all categories; No Awarded all picks, Puppy and Non-Alike)
Primarily No Awarders But Considered a Puppy Pick above No Award: 1000
Absolute No Awarders: 2500

END UPDATE, 8/25/15

Some percentages (estimates, not precise):
No Awarders: 3500 / 5950 = 59%
Neutrals: 1400 / 5950 = 22%
Rabid Puppies = 10%
Sad Puppies = 9%

What the Best Novel category would have looked like with No Puppy votes:
Ancillary Sword, Ann Leckie
The Goblin Emperor, Katherine Addison
The Three Body Problem, Cixin Liu
Lock In, John Scalzi
City of Stairs, Robert Jackson Bennett

Other initial Best Novel analysis: Goblin Emperor lost the Best Novel to Three-Body Problem by 200 votes. Since there seem to have been at least 500 Rabid Puppy voters who followed VD’s suggestion to vote Liu first, this means Liu won because of the Rabid Puppies. Take that as you will.

I’m going to get some sleep. I’m tired, so I’m sure I slipped at least once on one of these numbers. Too much data! Happy analyzing!