Hi all! I’ve been ramping up development over these last few weeks and lots of progress is being made. No exact dates on a launch schedule, but it’s getting close. Today, I’m going to be discussing the mechanics around campaign buildings, their impacts, and the finances around them.
“Ground game” is a common term you’ll hear in American elections, but it’s just another term for “campaign infrastructure.” A good ground game usually features a strong base of volunteers, a clear organizational structure in a county, and some sort of operation for turning out voters.
In The Electorate, you’ll be planting the seeds of a ground game through the use of campaign offices. In each county, you’ll have access to four distinct buildings that give passive effects each turn:
County Campaign Headquarters: Establishes a presence in the county.
Canvassing Center: Volunteers make contact with voters and drive turnout.
Volunteer Field Office: An improved space for volunteers to become more effective and lower the weekly costs of other buildings.
Merchandise Store: A storefront to drive donations through merch sales.
In order to build any office in a county, you first have to construct a County Campaign Headquarters. HQs add +0.01 to the state’s enthusiasm every turn, generate donations, but more importantly, unlock the other offices for construction. Canvassing Centers improves county turnout by 1% every turn, while Merchandise Stores generate revenue equal to 25% of the county’s registered voters every turn. Finally, Volunteer Field Offices amplify the impact of each by adding an additional +0.01 to and HQs generated enthusiasm, +0.10% to turnout from Canvassing Centers, and reduce the cost of operations for every building by 10%.
An example of office infrastructure in Los Angeles County. Still balancing the effects and costs of each office.
Each office has a fixed cost of construction, but their weekly costs are variable depending on the registered voter population of a county. For example, weekly costs for an HQ in Los Angeles can be upwards of $80,000 per week but will be significantly less in smaller counties. As a result, Volunteer Field Offices have an outsized impact in larger counties, but may not be worth the squeeze in smaller ones.
Meanwhile, Merchandise Stores are designed to bring in recurring revenue for your campaign. It’s important to realize that there are situations where these stores may not be profitable. Merch Stores require weekly inventory and payroll, so they’re expensive to run, plus a flat construction cost.
With these balances, it’s important to plan your ground game strategy correctly. You cannot spam buildings in every county, and expect to have enough funds to conduct the rest of your campaign. Additionally, your opponents will be making similar decisions.
In the last dev log, I explained what I had in mind for the Congressional system in the game. If you haven’t read it already, you can find it here.
I coded my theory for the system last week and found that it works pretty well. There’s a few things to balance, but it works well enough.
Basically, the game will generate a few thousand candidates using random first and last names, give them an archetype (more details on them in a later dev log since I’m still working out what these should be), a random name recognition score (from 0.01-1), a random ideology, and then a party that corresponds to their ideology, with an 8% chance of being an independent (that’s just how the math worked out, plus 10% felt too high).
Once the list of random candidates is generated, they’re filled into the districts at random. Districts will have a minimum of two candidates, but can go as high as eight. After they’re sorted, the script will loop through the counties inside the district and retrieve the voter population data for each one. With the data from each county, they’ll fill a voter profile for a total of 700,000-800,000 voters for the district (US Congressional Districts average about 750,000 people).
Then, the candidates inside that district will be bucketed by party, and their maximum vote totals as a share of the electorate in that district will be divided by the total number of same-party candidates. Then, to get their starting turnout, we multiply by 0.4, and then multiply that by their name recognition.
After all of that, you actually wind up with pretty believable voting numbers. A scenario that happened during testing showed that the Republican primary for a seat located in the San Diego suburbs was divided among three candidates. One had the archetype of “Local Hero” and had .80 name recognition and the Moderate ideology. Another was a Nationalist with the archetype of “Firebrand” with a name recognition of 0.43. Then, an incumbent Conservative known as a “Backroom Broker” had a name recognition of 0.96. The Nationalist had 12% in the polls, the Firebrand had 23%, and the incumbent Conservative had 55%. Scenarios like this happen across all 435 districts amongst all parties. It’s pretty cool and can create a lot of narrative-driven content.
Additionally, you’ll be able to endorse and campaign with these candidates.
In Los Angeles, I have the opportunity to campaign with Favian Manson, a Populist Republican.
It doesn’t make a huge impact on your presidential campaign, but if you’re interested in immersing yourself in the election, it’s a cool way to amp up the stakes.
Like I said earlier, progress is really churning along. The core UI panels are 90% programmed and ready to go, so that’s been a big boon to making the game feel alive. I’ll keep everyone up to date as I work through smaller items on our Discord server. If you haven’t joined already, you can here.
Thanks!