All prudent role-players know that Dungeons & Dragons (which I usex to encompass all pretend-elf games derived from the 1974 original) in no way reflects a medieval fantasy setting, pseudo-European or otherwise. If you haven’t yet come to this conclusion, do yourself a favor and read the seminal “d&d is anti-medieval” post from my colleague at the Blog of Holding. If you are one of those for whom a 2016 blogpost isn’t a sufficient appeal to authority, note also that Gary Gygax said in Dragon Magazine #25 that the important factor in making D&D a “fantastic medieval game” is “medieval technology, not necessarily feudalism with primogeniture, entail, and a Salic Law.” (However, I am not sure even this distinction holds up given that firearms apparently have been widely available through the Forgotten Realms for hundreds of fictional years and were introduced to AD&D in Dragon Magazine #60 just a few years after Gygax made this claim). 

Any discussion of what systems of government should be present in a setting is typically framed by what real world time and place in our world is most closely analogous to the time and place of the setting. If your setting is more Ancient Rome than Ancient England, you are more likely to have a republic of elites rather than a monarchy by divine right. And because most elf-game settings tend to hew toward mythical feudal Europe (although my colleague at Against the Wicked City makes a good point that at least the P/OSR version of D&D is better suited to sword-and-sorcery fantasy-horror or gonzo science fantasy settings), we get variations on the political systems of feudal Europe as an assumed part of play. If your setting has kings, queens, princesses, barons von such and such, and dukes so and so, it goes largely without much fanfare. It is only when you deviate from this norm that it is given particular focus. Our vaguely medieval fantasy setting isn’t like other girls. 

There are, of course, more modern settings that would almost require a democracy (although maybe a shambolic one). Take, for instance, the near-modern high-weird setting of Electric Bastionland. While Chris McDowall’s latest project, Mythic Bastionland, seems to demand an old fashioned pseudo-King Arthur or the equivalent, Electric Bastionland is a better fit for politics that more closely resemble our own. So much so that, in 2020 when Electric Bastionland was still shiny and new (it still remains ever-golden), I considered writing a hack called “Electoral Ballot Land”. The pitch: “You have a Flailing Candidate. They have a polling deficit. Talking to voters is your only hope.” Maybe Chris will stoop to this level after he tackles the stars with Intergalactic Bastionland. 

What is not often a consideration when adding this type of setting detail is what system of government is most gameable. To clarify this question, I mean which system of government (1) gives the players readily obtainable information about who is in charge in the setting, (2) allows the players leeway in impacting who is in charge in the setting, and (3) makes it important to the setting who is in charge. If who is pulling the strings is a complete mystery (e.g., if the city is ruled by a council of anonymous masked lords whose identities are actually secret [so not like Waterdeep where IYKYK]), it doesn’t really matter who is in charge. If players have no ability to change who is in power (e.g., the invincible overlord is actually invincible no matter how much power the players accumulate), it doesn’t really matter who is in charge. If whoever is in charge isn’t really in charge (e.g., there is a monarch, but their role is largely ceremonial and everything is actually controlled at an extreme local level by random minor nobles of little importance who don’t give a rat’s arse what the monarch has to say), it doesn’t really matter who is in charge. This is just applying the Information Choice Impact doctrine to forms of government to see which promotes player agency, which in turn makes for the most engaging gameplay. 

You should include democracy in your games for the sole reason of gameability. While any form of government can possibly meet this criteria (a monarch with open rivals that the players could ally with or against if they choose would be a classic), a government that is a democracy is almost certain to hit the mark. A democracy, as defined by noted government-knower Gary Gygax, is “Government by the people, i.e. the established body of citizens, whether direct or through elected representatives.” This definition elides a lot of the trickier questions about what is and is not a democracy that I’ll leave to the political mad scientists. 

Under this definition, even a monarchy can be a democracy if every time a monarch dies all of the minor nobility (in this case, the established body of citizens) get together and vote on which of them will be the new monarch. If your setting has a democracy, the players know who is in charge (the relevant electorate), the players can more easily influence the outcome of who is in charge (more on this in the next sentence), and who is in charge presumably matters. If the electorate is massive, say in the hundreds of millions, it is difficult for the players to impact the government, so smaller scale democracies are ideal from a gaming perspective. 

The player characters can then go about influencing the outcome of the elections either through legitimate means (campaigning [not the typical kind adventurers do] and attempting to persuade the electorate) or the more fun underhanding methods (bribes, blackmail, threats, misinformation). There is so much on the table when there is an election happening in your fantasy setting that isn’t typically the case with a mostly stable monarchy. A democracy is essentially a perpetually unstable monarchy and from the perspective of the players, that is a good thing. The monarch probably isn’t going to die every couple of years, try as they might.

In Barkeep on the Borderlands, the monarch is dying but that wasn’t enough. Because a dying monarch is an impending power vacuum, I included all sorts of factions that were hankering to fill it. One such faction was itself going through turmoil: the quasi-parliament, which was holding an election for prime minister at the same time. If the dying monarch was a hole in a donut, the potential ouster of the incumbent prime minister was a further hole in the donut hole. The goal was to give players as many bites at the apple to influence the world as possible. Save the monarch’s life or not, influence the election or not, all of these choices have consequences.

But how do you mechanically simulate an election? If you ask this question on reddit, you’ll get something like “Performance checks on campaign speeches would be the core of how I would do it, with advantage if they pass a persuasion check.” This is, frankly, really bad advice. This method means that it isn’t important what the persuasion is or what is said in the campaign speech, it is just a total random chance influenced by two charisma-based skills. Not to mention how disconnected this is with how elections work. Have you ever changed your vote because of a candidate’s campaign speech? This is basically how you should handle an election if you don’t care about it and wish your players would stop with this nonsense so let’s just get this over with. So what if you actually care about the results and you want player actions to matter (but not be totally determinative, because that subverts realism)?

First, set a baseline for how the election plays out if the player characters do nothing. Doing nothing is a choice the players can make and there should be some consequence if they don’t intervene. In Barkeep, the Incumbent was favored but not guaranteed to win even absent player action. The referee was instructed to model the Incumbent’s percentage share of the vote by rolling 6d20 and totalling the highest 4 dice. This creates a very round bell curve hovering around 56% of the vote–a very close margin. In more binary terms, it means the Incumbent would lose 34% of the time. Ye Olde Nate Silver would have a field day with this, I’m sure.

The uncertainty of the election (with a tilt due to one candidate being favored) is important because it means if the players are invested in the outcome, they will need to roll up their sleeves either way. If the favored candidate was guaranteed to win, then the players could just sit back and watch if that was also the candidate they wanted to win. But with this method of generating the default results, it isn’t a sure thing that the Incumbent will stay the Incumbent. The players aligning themselves with the Incumbent may be savvy, but they will also want to help their new ally or else they’ll find themselves with an enemy in power if the Incumbent loses.

After setting the election’s baseline, you need to consider the ways that the players can influence the election. Because only parliamentarians (a fairly common faction in Barkeep) could vote in the election, it meant a small electorate where changing the mind of a single member has a meaningful impact on the result. This is where a democracy gives players the most agency. To reflect this, I outlined that if any specific Reform-aligned parliamentarian is rendered unable to vote (sinister phrasing), it reduces the Incumbent’s odds by 1% and vice versa for Royalist-aligned parliamentarians. If the jolly crew casts ballots (either illicitly or by becoming parliamentarians themselves [the adventure includes a few avenues for this to happen]), each vote influences the election by 1%. And rousing speeches do have an impact, although they aren’t determinative like reddit seems to suspect: rousing speeches sway the result by 1d6%. I leave it up to the referee to determine whether a speech is rousing.

Because the election is set to be swayed by as few as 5% on average, changing even a few votes can be determinative for who wins the election. This means that players can absolutely be the deciding factor for the election if they make that one of their goals over the course of the adventure. Even a single very rousing speech could swing a close election. 

After you provide avenues for player influence over the election, you need to set the consequences for the various outcomes. If the result is space aliens take over the world no matter who wins, you really needn’t have gone through all that trouble. So the candidates should be materially different in how they would govern. 

Taking Barkeep again as an example, a victory for the Incumbent maintains the status quo, while their opponent will act to empower the monarchy and punish anti-royalist sentiment with draconian methods. I also added a third possible outcome: a blowout victory. While the status quo is maintained if the Incumbent squeaks out a victory, if they win 70% or more of the vote (an unlikely outcome at slightly more than a 1-in-20 chance; literally a critical hit for the Incumbent), they are empowered to enact their agenda. To quote directly from the adventure: 

“In the statistically unlikely event that the Incumbent wins over 70% of the vote, they grow a spine for the first time in their life and will instigate the overthrow of the monarchy, working with Academy students to form the world's first democratic magocracy. In one year’s time, the Keep is thoroughly transformed.” 

This is basically a backdoor way to end the monarchy, whether the Heir succeeds at saving their royal parent or if they take the throne themselves. And, because this is an outcome that would delight the revolutionary students of the Academy, it gives the players an opportunity to form an alliance between two factions to stomp out a third. How will the Monarchy-aligned Church of Chaos respond to such an alliance though? Or the anti-Monarch but also anti-Keep faction of goblins? It is the possibility of players getting caught in this web of factions that makes a social adventure like Barkeep on the Borderlands really come alive at the gaming table. And nothing riles up competing factions quite like an election. 

If you are in America, remember to go vote today if you haven’t already. Don’t vote for the fascist.