I went to my first WorldCon this weekend. Nominally, I was there to do some Maze of Games signings and play games with fans. But really, I was there to see the Hugo Awards. With all the hubbub, I wasn’t going to miss out on being there when it all went down.

It was a staggeringly cool show. As the haze outside reeked of wildfire smoke, hosts David Gerrold and Tananarive Due got into the apocalyptic spirit of the night, battling the Grim Reaper and handing out asterisks. Robert Silverberg led the crowd in singing the Hare Krishna chant. A Dalek threatened to exterminate us. It was great.

I sat in the audience watching people cheering when no one got any awards, and was struck by a singular thought: No one wants this. Literally no one present was getting what they wanted: the Sad/Rabid Puppies’ favored creators weren’t getting any awards, the deserving creators weren’t getting any awards, the fans weren’t giving any awards.

As the show went on, I thought about who I was: a game designer, an award show host, a creator of systems, a solver of complex puzzles. And a second singular thought struck me: I can fix this.

Now, no one has asked me to fix the Hugo Awards. I’m not even sure anyone wants it fixed. But the important thing to understand is that it can be fixed, by applying some principles of game design and behavior analysis. This is what I do every day, and if you want it, World Science Fiction Society, I’m happy to do it for you.

Here’s the deal: Awards aren’t real. They’re just the results of a game. The Puppies gamed the system. If the system is gameable, then it can be made ungameable. So here are some game design principles for fixing the Hugos.

1. Don’t solve the problem people say you have. Solve the one you have.

One of my rules I tell all my playtesters is “Don’t tell me what you think will happen. Tell me what happened.” People will complain about all sorts of things, but it’s important to figure out whether there is a real problem, and if so, what the problem is.

So let’s listen to folks. When George R.R. Martin—maybe the most popular author who cares about these things—says, “I think the Sad Puppies have broken the Hugo Awards, and I am not sure they can ever be repaired,” he is telling you what he thinks will happen; we can’t learn much from that. When he later describes the outcome of many No Awards winning as “the best result we could have hoped for, and a victory for WorldCon, fandom, and the Hugos themselves,” he is describing what happened. He is doing so emotionally, but we can learn something from it. We can learn that GRRM’s instinct was right: an award show that doesn’t give out its awards is broken. We can then seek the root problem.

To do so, let me give you a parallel example.

Let’s say you’re the president of a bank you’ve called The People’s Bank. You leave the front door unlocked so that one of your offshore accounts can drop off a million dollars late at night. A lower-class street gang sees the door is open, walks in, and steals the money under the cover of darkness. What is the problem you have?

People will tell you it’s those awful gangs. I suppose that’s not entirely inaccurate. No one asked them to steal your money. They’re bad people, and should be punished for it.

But hey, you might just take a look inward and ask yourself whether it was a good idea to leave the front door unlocked with lots of money inside. It’s your security system, and you didn’t think out the consequences of imposing that system. You should fix the system.

Which is a great idea, but it’s still not the problem you have.

The root problem you have is that the street gang wanted the money in the first place. You had something of value and other people wanted it enough to break the rules. What you are dealing with here is the problem of inequity. Some people believe they are not getting what they want from your “People’s Bank,” and maybe if you can solve that, you can solve your problems in a way everyone wants.

So, the Hugos: The Puppies are a symptom. The fact that they broke your rules first doesn’t mean anyone can’t break them. And the fact that they wanted to break them enough to cause you to set your awards on fire means you should solve the problem of inequity before someone else decides to do this again.

2. Don’t pay people to do things you don’t want them to do.

The above rule was coined by legendary game designer Jonathan Tweet, and I teach it to all my collaborators. At heart it requires you to look at what your game’s economy does and whether you want it to do that.

You might say, “The Hugo Awards doesn’t pay anyone!” And, Mr. Strawman, you’d be wrong. The Hugo Award has a value. Heck, even a Hugo nomination has a value. The ability to say “My Hugo-nominated novel…” is a real thing of worth. Ask anyone who has one.

So what’s the value of a Hugo nomination? I can’t say. But let me ask it like this: If I gave you a onetime chance to pay $2000 so that you could say you were a Hugo nominee, would you do it? And what if you could crowdsource the $2000? What if a mere donation to your GoFundMe of $40 could allow a friend of yours to say she helped you achieve your dreams?

And then, after all that, what if I told you that doing this can get a whole lot more of your friends nominations for no additional cost? Would you do it then?

And even if you wouldn’t do that, do you know someone who would?

Well, this analysis on the Amazing Stories site says it would work.

If it’s accurate (and I am not inside-baseball enough to know), then the problem of inequity is simply defined: the Hugo Awards are worth more to people who don’t have them than the price of acquiring them. For the price of $1600, you can suborn enough votes—in this case, 40 of them—to get a short story nomination; for a mere $6000, you can suborn the 150 votes to get a novel nomination. And once you have enough to get a novel nomination, you can get all the nominations you want.

So to solve your inequity problem, you should do one of three things:

1. Lower the value of a Hugo nomination so that no one wants it that much
   
2. Raise the price of buying a Hugo nomination so that no one can afford to game the system
   
3. Be more equitable in your nominations
   

Course 1 is inadvisable because you want to keep the value of the awards high; if no one wants them, you suffer in other ways. 

Course 2 is smart but tricky; if you make a membership $500, say, then those 150 votes cost 75 grand—but maybe you lose the good-faith members who can’t afford it. 

Course 3 is the most cost-effective, but somehow you have to get the marginalized interest groups their nominations while maintaining fairness. Sure, you might enjoy mocking conservative, straight white men when they say they’re marginalized, but they have some power here, and simply laughing that power out of the room won’t make it go away. And when it’s not conservative, straight white men, you won’t be laughing anyway. You’ll be trying to find a way to get the marginalized people what they want. So you should have a way to do so.

3. When one mechanic works and another doesn’t, use the one that works.

The Hugo Awards’ description of the voting system uses the word “complicated” three times. It is clearly not so complicated that it cannot be gamed. But it is complicated enough that parts of it work better than other parts. In an environment like that, a game designer removes the parts that don’t work in favor of the parts that do.

The awards system is a two-part mechanical process:

1. People can buy memberships and nominate up to five works per category, which are culled to a list of fifteen potential nominees, of which five are selected.
   
2. Members of the current WorldCon can rank the nominees plus “No Award” in desired order, and after a series of devil-take-the-hindmost cullings, the highest vote-getter wins.
   

Depending on who you are, you might say one of those two parts is broken. I’m going to presume the Hugo administrators think Mechanic 1 didn’t work and Mechanic 2 did work in 2015. Mechanic 1 got slates of putatively less deserving creators; Mechanic 2 got No Award to win out over those slated creators.

So Mechanic 1 is likely flawed. One solution would be to replace Mechanic 1 with Mechanic 2 wholesale. There are a couple of problems with that. The first is that the Hugos value the input of members who cannot attend WorldCon, which moves from place to place and thus changes who can come each year. The second is that the timing of the nominations occurs before those members decide whether they will attend.

One approach that might work is to acknowledge the hundreds of annual SF/Fantasy conventions that occur in the winter and spring of each year; for example, NorWesCon in Seattle. If the strategy was that you had to either attend WorldCon or one of these other cons to vote, it would be much harder to do the point-and-click vote-stuffing that happened with the nominations.

A more subtle approach is to break the nomination process in half. You could continue to have members nominate some number (I’d recommend “one”) of their favorite works, but then have the culling process happen at the satellite cons. So buying a membership gets you the right to log an opinion, but not to cull the list to five nominees.

Any number of other versions of this approach would work. The point is that the entire system doesn’t need to be reanalyzed, just the part that broke. And the method of fixing might be gleaned from the part that didn’t.

4. Punish behavior that doesn’t fit the game you want to play.

Most games don’t have a rule that says “Don’t cheat.” That’s because cheating is universally punished by game players. But they do have all sorts of rules that say “Don’t play in a manner that doesn’t look like cheating but still sort of is.” Texas Hold ‘Em doesn’t permit string-betting. Bridge doesn’t let you tip off your partner with coded signals. Hockey doesn’t let you blast the puck back down the ice instead of facing the offense. When a player is caught doing these things, the penalty can be quite severe.

The people who ran the Puppies slates did not break the “Don’t cheat” rule, but they did act against the spirit of the game. The spirit of the Hugos is “Pick the best writers, artists, and editors.” It is not “Pick a slate of people who show you can wreck the awards.” (This is where an analysis of the differences between the groups of Puppies could be useful, but seriously, who has the time.)

There is no world in the multiverse where Vox Day is the best editor, or Carter Reid is the best artist, or John C. Wright penned three of the year’s five best novellas. The people who backed these nominations are trying to wreck the awards, so it’s worthwhile to purge them from the system. This is equivalent to Xbox and Twitter punishing harassing behavior by suspending accounts.

The simplest way to do this is to suspend all memberships of people who did not attend WorldCon but did vote for the Puppies slate. Or you could reset the membership to before the slating happened. Refund the dues of anyone who supported the con but didn’t attend, and then start over with a different system. That money you got was from people who wanted to trash your stuff, so maybe you don’t want that money anyway.

Regardless, you need to ban the behavior that got you into this mess. Slating is bad for everyone, because it encourages people to vote for works that they haven’t read. So suspend the membership of anyone who publishes a slate. Doesn’t matter if it’s my friend Mr. Scalzi doing the slating. If Vox Day can’t do it, no one can.

This stuff may seem uncomfortably draconian to some members. Draconian measures succeed for a reason. After chasing out the trolls, you might have a group people want to belong to.

5. Issue a patch.

We game designers screw up a lot. In tabletop games, we issue FAQs and lists of errata; I am something of an errata-generating machine, it turns out. In videogames, we issue patches that wipe away the effects of mistakes we make.

2015’s Hugo Awards didn’t have the effects that the fans wanted. For example, my friend Mr. Rothfuss flat-out earned the Best Novella award for The Slow Regard of Silent Things. It should not be allowed to pass that he doesn’t get it just because the trolls stole it from him.

So let’s fix that. For the categories that got No Award, give out the awards to the people who deserved them. GRRM has a fine list drawn from the actual balloting, giving awards to Rothfuss, Ursula Vernon, Jo Walton, John Joseph Adams and Liz Gorinsky; that can serve as a one-time patch to the 2015 No Awards. (For awards that were given out, such as to artist Julie Dillon and the novel The Three-Body Problem, those get to stick. You don’t re-award prizes that people walked home with.)

For criminy’s sake, you made these badass trophies. So give them to the people who merit them.

Doing so has two positive effects: it gets the deserving creators what they deserve and it tells the trolls that they don’t get to win. This step is crucial to the process; if you don’t do it, the trolls have a strong reason to try to break the system again.

In fact, this is an easy change to the bylaws: If No Award wins a category, the Hugo committee may decide to award that Hugo to a deserving winner in that category. That’s a threat. And it will work.

So that’s it. A five-step program to fixing the awards. That’s the result of a few hours of research and pontificating; it’s possible that a longer process would produce different recommendations. But it’s a place to start. If you’re worried that your system is gameable, you might gain some value from asking us game designers to fix it.

Thanks for listening. I hope to see you in Kansas City.

Mike