NPP engine changelog | A House Divided

Non-player politicians (NPPs) fill offices, run in elections, govern, legislate, invest and run companies wherever players do not. This page tracks the NPP engine on its own version line, next to the game changelog.

Two settings decide what an NPP does in a world:

•Autonomy level (off, v0 to v6) is chosen per world and says which activities NPPs may perform there.
•Engine version is the shipped code and says how NPPs perform those activities. Every world on the same build runs the same engine version.

Engine majors follow the autonomy tiers. Engine 6 is the current overhaul, and its behavior runs only in worlds set to autonomy v6. A world below v6 keeps the engine 5 behavior.

Version rules: a minor version changes an NPP decision (what an NPP chooses, when, or how well). A patch changes no decision (a fix restoring intended behavior, or a cost-only change). A major adds an autonomy tier.

Engine 6 follows two fixed constraints:

•NPPs get smarter, not stronger. No engine 6 version raises NPP action points, income, starting money or election weight.
•The existing NPP election modifiers and NPP income rates stay as they are while the decision changes land. Once simulation shows how much stronger the changes make NPPs, the income NPPs receive without acting is stepped down to keep overall NPP strength level with engine 5.

Roadmap

Planned work in stages, in order. Each stage ships as one or more engine 6 minor versions behind autonomy v6, and each version is compared against v5 in seeded world simulations, with matched starts and more than one era, before it is listed as shipped. A shipped version appears below the roadmap with its own number.

Stage 1: running for office

•
Race choice by chance of winning. NPPs file where their party's strength gives them a real chance. (Strongest NPPs in senior races shipped in 6.1.0; weak parties putting them in winnable local races shipped in 6.12.0.)
•
Campaigning tied to a race. Campaign and advertising actions spend on the race the NPP is running in, weighted toward close races, instead of raising general influence and favorability. (Rallies in the NPP's own race shipped in 6.2.0.)
•
Consolidation. A clearly trailing NPP withdraws before the primary rather than splitting the vote. It does not endorse: NPP endorsements stay something players request. (Shipped in 6.4.0.)
•
Presidential challengers in non-player countries, inside the existing limit on how often a player's race can be challenged. (Shipped in 6.3.0.)
•
Career ladders. Ambitious, strong incumbents step up to open senior races (shipped in 6.9.0).

Stage 2: governing

•
Bills that can pass. Before sponsoring, an NPP counts likely votes and does not introduce a bill that is short. (Shipped in 6.5.0, counting party, district, donor and opposition pressure; whips and caucuses come later.)
•
Whipping its own program. A governing party whips co-partisans on the bills that serve its standing goals (shipped in 6.7.0). Opposition coordinates only against a government it can defeat (shipped in 6.11.0).
•
Ministers judged on their goal. Cabinet posts go to NPPs ranked on fit with the head of government, competence and the brief; ministers are judged on the standing goal their brief serves (shipped in 6.15.0).
•
Fewer writes. Ministerial orders and settings are written in batches per country (shipped in 6.14.3).
•
Currency defence. Central banks answer sharp currency moves with rates (6.6.0) and reserve-backed trading bands (6.13.0).

Stage 3: corporations

•Investing on value. NPP share and bond purchases compare price with fundamentals and sell on the same test, instead of fixed-probability rolls (shipped in 6.16.0).
•CEO choice by record. NPP-run companies choose CEOs from NPPs with a track record running companies in the same sector (shipped in 6.17.0).
•Consequences across politics and business. NPP officeholders weigh their holdings in votes (6.8.0), the conflict is shown on their profile (6.13.1), and voters mark down a vote their holdings decided (6.14.0).
•Batched purchases. The share-buying sweep was the largest single NPP cost in a turn (18.0 seconds and 1,465 database round trips on live turn 168). Markets now settle in parallel with one pool read per sweep (6.16.1). The bond sweep is still one purchase at a time.

Stage 4: cost

•NPP governing passes run countries in bounded parallel lanes instead of one after another (shipped in 6.14.2).
•Every engine 6 phase gets a per-turn round-trip budget, reported with the turn.

Stage 5: strength accounting

•Compare v5 and v6 worlds on NPP win rate against players, NPP wealth, and returns on NPP-run companies.
•Step down the income NPPs receive without acting until those measures match engine 5. The election modifiers are not changed.

Stage 6: default for new worlds

•Autonomy v6 becomes the default for new singleplayer and simulation worlds (shipped in 6.14.1). Running worlds never change level on their own. The live multiplayer world moved to v6 with game 1.16.

6.19.0 (game 1.16.0, 2026-10-15)

Stage 3 (corporations).

•NPP-run plants shrink to the market they serve (v6). A plant that sells everything it makes but is held far below its size by weak demand kept paying payroll and upkeep on the idle rest, because sold-out plants were treated as recovering. When an NPP-run plant's output stays under 60% of its size for 24 turns, it now retires part of the idle capacity each turn, keeping room to grow back by half. On live turn 202, loss-making NPP sectors ran at 38-46% of capacity while selling 92-100% of what they made. Player plants are unchanged.

6.18.0 (game 1.16.0, 2026-10-15)

Stage 3 (corporations).

•R&D is never cut to zero (v6). NPP-run companies with thin margins or short cash kept no R&D budget, so their technology score decayed every turn, their product quality fell with it, and lower sales pushed margins lower still. They now keep a small maintenance R&D budget, which is enough to hold technology near par. On live turn 199, NPP companies losing money averaged quality 43 against 70 for those above a 20% margin. Player companies at the same thin margins kept quality above 70.
•Losing companies change course sooner (v6). A company's strategy used to count as working whenever it did better than when it was adopted, so a company that started expanding while losing money kept expanding as long as it lost a little less, and was only re-judged every 48 turns. A strategy that still loses money no longer counts as working, and a loss-making company is re-judged after 8 turns. On live turn 199, 101 of the 120 loss-making NPP companies were still on the expand strategy.
•Fix: a caretaker seat left empty is filled again. A player company handed to a caretaker that was later marked as having no CEO stayed that way: the vacancy step skipped it because it was already caretaken. It now gets a caretaker again the next turn, and the owner's arrangement stays.

6.17.1 (game 1.16.0, 2026-10-15)

•Fix: a successor takes every seat a dead NPP held (v5). When an NPP dies, the NPP that replaces it now also takes its cabinet posts, its place as national leader, and any player company it was running as caretaker. The owner's caretaker arrangement stays in force. Before, those seats kept pointing at the dead NPP.
•Fix: consolidation respects a party chair's slate (v6). A trailing NPP is no longer withdrawn from a primary its party chair slated it to, or from any race the chair is managing for the party.

6.17.0 (game 1.16.0, 2026-10-14)

Stage 3 (corporations).

•CEOs chosen by track record (v6). NPPs now keep a record of how the companies they ran did, per industry: time in charge and the company's operating margin against other companies in the same country and industry. When an NPP-run company needs a CEO, it prefers a free NPP with a good record in its industry. A record counts only after half a game year and only if the NPP beat its peers. With no such candidate, the old rule applies. A CEO a player names is never overridden. The record is kept in every world, so a world switched to v6 already has history.
•Fix: a dead NPP no longer stays CEO. When an NPP dies (autonomy v5 and above), its NPP-run companies now get a new CEO the next turn instead of keeping the dead one.

6.16.1 (game 1.16.0, 2026-10-14)

•Faster share buying (v6). At v6 the NPP share-buying sweep runs each currency's market separately and in parallel, and reads each market's pool once per sweep instead of once per purchase. Tests confirm every purchase still balances to the cent and the result matches buying one at a time. Decisions are unchanged.

6.16.0 (game 1.16.0, 2026-10-14)

Stage 3 (corporations).

•Investing on opportunity (v6). NPPs used to decide whether to buy shares or bonds with a fixed roll before looking at the market, then picked what to buy. Now they look first: the better the best deal on offer (a share price below what the company is worth, or a bond yield above the minimum), the more likely they are to buy. Shares well above what the company is worth are more likely to be sold. Across the market the total amount of buying and selling stays about the same as before; it just goes where the value is.

6.15.0 (game 1.16.0, 2026-10-14)

Stage 2 (governing).

•
Ministers judged on their goals (v6). The head of an NPP government used to judge each minister only on how far its domains sat below target. When the government holds standing goals, a minister whose brief covers one is now judged on that goal instead.
- •A minister turning a failing domain around (at least 5 points of progress since the attempt began) is kept.
- •A minister in post when a goal on its brief failed, with no progress since, can be replaced even if the shortfall looks small.
- •A minister appointed after the failure is not blamed for it.
- •Minimum tenure, cooldowns and the limit on repeated reshuffles still apply.

6.14.3 (game 1.16.0, 2026-10-14)

•Ministers save their work in batches. NPP and caretaker ministers load their settings and orders once per country and save changes in single batched writes, instead of several database trips per minister. Applies at every autonomy level; decisions are unchanged.

6.14.2 (game 1.16.0, 2026-10-14)

•Governments plan in parallel (v6). At v6 the NPP governing phase runs up to four countries' domestic steps at once, then foreign policy one country at a time in the usual order, because one government's foreign moves shape the next. Below v6 it runs one country at a time as before. Decisions are unchanged.

6.14.1 (game 1.16.0, 2026-10-14)

•V6 is the default for new local worlds. New singleplayer and simulation worlds now start on autonomy v6, so they run every engine 6 change. Existing worlds keep their level, and the live multiplayer world stays on its own setting. The desktop client lists V6 and recommends it. No decision changes.

6.14.0 (game 1.16.0, 2026-10-14)

Stage 3 (corporations).

•Voters notice self-interested votes (v6). When an NPP's own companies decide its vote on a bill, it loses 1 favorability. "Decide" means it would have voted differently without them. This happens at most once per voting turn and never takes favorability below zero. Business interest is no longer free: an NPP that votes its holdings pays a small price with voters.

6.13.1 (game 1.16.0, 2026-10-14)

•Business interests are visible (v6). When an NPP's own companies pull its vote (6.8.0), the vote forecast on its profile now says so, naming whether the bill helps or hurts a company it runs or holds. Players and opponents can see the conflict. No decision changes.

6.13.0 (game 1.16.0, 2026-10-14)

Stage 2 (governing).

•
Central banks defend a falling currency (v6). Since 6.6.0 a central bank run by an NPP adjusts its interest rate when its currency moves, but in limited steps that a sharp fall can outrun. Its other tool, a trading band defended with the bank's reserves, was only ever used by player chairs.
- •Now, once a floating currency has fallen 15% or more over 24 turns, the NPP chair sets a band. Its top sits 5% above today's rate, which caps further weakening, and its bottom 10% below.
- •The bank defends the band with its own reserves under the usual rules, so it can never spend more than it holds.
- •The chair removes its own band once the currency's 24-turn move is back within 5%, after the normal wait between band changes.
- •A band set by a player is never touched, and government-held rates, pegs and shared currencies are left alone.

6.12.0 (game 1.16.0, 2026-10-14)

Stage 1 (running for office).

•Racing where the party can win (v6). Since 6.1.0 a party runs its strongest NPPs for its most senior races. Where the party is weak across the whole region, those regionwide races are the ones it cannot win, so that spent its best candidates on lost causes. Now a party whose organization in the region is below 75% of the leading party's runs its strongest NPPs in its most local races instead, where it can still win seats. The same races are contested either way.

6.11.0 (game 1.16.0, 2026-10-14)

Stage 2 (governing).

•The opposition picks its battles (v6). The opposition's NPPs used to vote as a bloc against every government bill, even against a government with a large majority, where the bloc could defeat nothing and only stopped opposition members backing bills they agreed with. Now the bloc holds only while the government side (governing party plus coalition partners) has less than 55% of the lower chamber. Above that, opposition NPPs vote on each bill's merits, home region and donors like any other member. The pre-filing vote count from 6.5.0 uses the same rule.

6.10.0 (game 1.16.0, 2026-10-14)

Stage 2 (governing).

•Vote counts include the government whip (v6). Since 6.5.0 a sponsor counts votes before filing, and since 6.7.0 an NPP government whips its own members on its goal bills. The count did not know about that whip, so it could set aside a goal bill the whip would have carried. The count now includes it, so the sponsor judges a goal bill the way it will actually be voted.

6.9.0 (game 1.16.0, 2026-10-14)

Stage 1 (running for office).

•Career ladders (v6). NPP incumbents used to re-file for the seat they held every cycle, so senior offices were filled only by NPPs without a seat. Now an ambitious incumbent with real standing can run for an open race for a more senior office in its home region instead. An open race is one with no candidate or defending incumbent from its party. This applies to an incumbent with ambition of at least 70 and political influence plus favorability of at least 100, holding a single seat. Its old seat is then filled the ordinary way. A party chair's slate assignment still takes precedence, and nationwide races are not part of the ladder.

6.8.0 (game 1.16.0, 2026-10-14)

The first stage 3 change (corporations).

•
Business interests shape votes (v6). An NPP used to vote on a bill that subsidises, protects or nationalises an industry as if it owned nothing. Now an NPP that runs a company, or holds at least 1% of one, weighs what the bill does to that company's sectors.
- •It leans for subsidies and sector tariffs in its own sectors.
- •It leans against ending those subsidies, tariffs aimed at its own company, and nationalisation of its company or sector.
- •Running the company counts twice as much as holding shares.
- •The pull is capped below a hard party whip, so party discipline and the NPP's own positions still matter.
- •Players can use this: a sector bill can win over the NPPs invested in that sector.

6.7.0 (game 1.16.0, 2026-10-14)

Stage 2 (governing).

•
Governments whip their own program (v6). An NPP government's members used to back its bills only through the general lean toward their own party's bills. Now, when an NPP heads the government, its members get a soft whip on the bills the government sponsored to serve one of its standing goals. Each member still weighs it against their own positions, home region and donors, and loyal members follow it more than stubborn ones, as with any whip.
- •An existing whip always takes precedence.
- •It is never applied in a party with a player as chair or vice-chair.
- •It is never applied to members of other parties.

6.6.0 (game 1.16.0, 2026-10-14)

Stage 2 (governing).

•
Central banks answer currency moves (v6). An autonomous central bank chair used to set its rate from inflation and growth alone, so a currency could fall sharply and the bank would not respond until the higher import prices showed up as inflation. Now a floating currency's move over the last 24 turns also counts.
- •A fall of more than 5% raises the bank's target rate by 0.1 points for each further 1%.
- •A rise of more than 5% lowers it at half that rate.
- •The shift is capped at 3 points.
- •The bank still moves toward its target in the same limited steps, with the same wait between changes, and the rate history records the currency move as part of the reason.
- •Pegged currencies and currencies under a shared central bank are not affected.

6.5.0 (game 1.16.0, 2026-10-14)

The first stage 2 change (governing).

•Counting votes before sponsoring (v6). An NPP sponsor used to file its highest-scoring bill whether or not the chamber would pass it. Now it first projects the vote in the chamber the bill starts in, using the same pressures NPPs vote with when no whip is set: their own positions, their home region, their donors, the pull toward their own party's bills, and the opposition bloc against government bills. Player-held seats are counted as voting with their party. A bill that does not project more votes for than against is set aside and the sponsor tries its next-best bill, up to three. If none has the votes, it files nothing that turn and keeps its slot. Tax-rate bills, whose rate is set when they are filed, are not counted.

6.4.0 (game 1.16.0, 2026-10-14)

Stage 1 (running for office).

•Trailing NPPs leave the primary (v6). In the last 12 turns before a primary closes, an NPP whose strength (political influence plus favorability) is below 60% of its party's strongest candidate in that primary withdraws. It does not endorse anyone; NPP endorsements stay something players request. Incumbents defending their seat never withdraw, and a primary with any player candidate is never touched. A withdrawn NPP cannot re-file in that primary.
•No late challengers (v6). NPP challengers no longer enter a primary in those last 12 turns, so a field that has consolidated stays consolidated.

6.3.0 (game 1.16.0, 2026-10-14)

Stage 1 (running for office).

•Presidential primary challengers (v6). NPPs could already challenge a same-party candidate in state and legislative primaries, but presidential primaries were left out. Now, in countries with no players, a presidential primary where a party has one candidate can draw a challenger from the same party. The challenger is picked from the party's five strongest available NPPs across the country, with ambitious NPPs more likely to run. The chance of a challenge and the limit of two challengers per country per turn are the same as in other primaries. Presidential primaries in countries with players are not challenged.

6.2.0 (game 1.16.0, 2026-10-14)

Stage 1 (running for office).

•
NPP candidates rally their own race (v6). An NPP's campaign action used to raise only its general political influence. NPPs have no campaign organisation, so their candidate Support, the race-level momentum that rallies move for players, stayed at the neutral 50. Now the first campaign action of a cycle by an NPP that is running in a race where players can rally becomes a rally for that race.
- •It costs the same action points and funds as the campaign action, and gives up that action's influence gain.
- •It moves Support by the same amount a player rally in the same kind of race does, with the same split between an immediate gain and a gain spread over the next four turns.
- •Players can rally once a turn. An NPP rallies at most once per action cycle, which is once every four turns.

6.1.0 (game 1.16.0, 2026-10-14)

The first stage 1 change (running for office).

•Strongest candidates in the senior races (v6). When a party fills a state's open primaries with NPPs, it now runs its strongest available NPPs for the most senior offices. Strength is the NPP's political influence plus favorability. The same primaries are filled as before; only who runs where changes. Before, NPPs were taken in load order and the lowest office was filled first, so the first NPP loaded ran for state senate and the last for governor.
•Nationwide executive slot (v6). A party's nationwide executive primary slot goes to its strongest available NPP instead of the NPP with the lowest id.

6.0.0 (game 1.16.0, 2026-10-14)

The foundation for the overhaul.

•Autonomy tier v6. A new tier above v5 that switches on engine 6 behavior. It adds no activity and no resources: a v6 world allows exactly what a v5 world allows.
•Engine version. The engine now carries its own version number, recorded on every turn log, and this changelog lists each version.
•Staggered action cycle (v6). NPPs take an action cycle once every four turns. Below v6 they all act on the same turn, which put the whole cost on one turn in four (27.6 seconds on live turn 168, mostly share and bond buying). At v6 each NPP has a fixed lane and a quarter of NPPs act on each turn. Each NPP still acts once every four turns with the action points it built up over that time. Selling shares, building party organisations and founding companies stay on the shared fourth turn.

5.0 (game 1.7.0, 2026-09-06)

•Persistent governing goals (autonomy v5). A government keeps up to five standing goals across agenda recomputes, grades each one as achieved, revised or failed, and feeds the result into its next agenda. Ministers hold a committed tier unless it stops advancing the goal, and bill sponsorship leans toward bills that serve an active goal.
•Ageing and succession (autonomy v5). NPPs age out and are replaced by successors.
•Difficulty that changes decisions. Local-world difficulty is split into resources and decisions, so a resource advantage is never presented as skill.

4.1 (game 1.0.0, 2026-07-20)

•Context-aware choices (autonomy v3 and v4). NPP actions respond to election timing, contested seats, government status and party finances, and stop spending on actions that can no longer change anything.
•Individual variation. Personality blends continuously, each NPP has stable quirks, and stubborn NPPs act more decisively than flexible ones.
•Limits on player impact. Caps on NPP bill sponsorship in player countries and on primary challengers against a player, a cooldown on NPP corporate attacks against the same player, and a 25% share of each market's float that NPP bond and share buying leaves untouched.
•Tier fixes. Two activities that switched off when a world moved from v3 to v4 now stay on at every higher tier.

0.0 to 4.0 (game 0.4.0, 2026-07-04)

The graduated autonomy ladder replaced the old on/off switch.

•v0: party organisation votes, and filling stalled executive seats in non-player countries.
•v1: the governing brain: agendas, cabinets, ministerial action, fiscal stance and opposition.
•v2: caretaker ministers and CEOs, corporate rivalry, and the first tier active in player countries.
•v3: campaigning, fundraising, saving, investing and founding companies, in non-player countries.
•v4: the v3 activities in every country.

Two settings decide what an NPP does in a world:

•Autonomy level (off, v0 to v6) is chosen per world and says which activities NPPs may perform there.
•Engine version is the shipped code and says how NPPs perform those activities. Every world on the same build runs the same engine version.

Engine majors follow the autonomy tiers. Engine 6 is the current overhaul, and its behavior runs only in worlds set to autonomy v6. A world below v6 keeps the engine 5 behavior.

Engine 6 follows two fixed constraints:

•NPPs get smarter, not stronger. No engine 6 version raises NPP action points, income, starting money or election weight.
•The existing NPP election modifiers and NPP income rates stay as they are while the decision changes land. Once simulation shows how much stronger the changes make NPPs, the income NPPs receive without acting is stepped down to keep overall NPP strength level with engine 5.

Roadmap

Stage 1: running for office

•
Race choice by chance of winning. NPPs file where their party's strength gives them a real chance. (Strongest NPPs in senior races shipped in 6.1.0; weak parties putting them in winnable local races shipped in 6.12.0.)
•
Campaigning tied to a race. Campaign and advertising actions spend on the race the NPP is running in, weighted toward close races, instead of raising general influence and favorability. (Rallies in the NPP's own race shipped in 6.2.0.)
•
Consolidation. A clearly trailing NPP withdraws before the primary rather than splitting the vote. It does not endorse: NPP endorsements stay something players request. (Shipped in 6.4.0.)
•
Presidential challengers in non-player countries, inside the existing limit on how often a player's race can be challenged. (Shipped in 6.3.0.)
•
Career ladders. Ambitious, strong incumbents step up to open senior races (shipped in 6.9.0).

Stage 2: governing

•
Bills that can pass. Before sponsoring, an NPP counts likely votes and does not introduce a bill that is short. (Shipped in 6.5.0, counting party, district, donor and opposition pressure; whips and caucuses come later.)
•
Whipping its own program. A governing party whips co-partisans on the bills that serve its standing goals (shipped in 6.7.0). Opposition coordinates only against a government it can defeat (shipped in 6.11.0).
•
Ministers judged on their goal. Cabinet posts go to NPPs ranked on fit with the head of government, competence and the brief; ministers are judged on the standing goal their brief serves (shipped in 6.15.0).
•
Fewer writes. Ministerial orders and settings are written in batches per country (shipped in 6.14.3).
•
Currency defence. Central banks answer sharp currency moves with rates (6.6.0) and reserve-backed trading bands (6.13.0).

Stage 3: corporations

•Investing on value. NPP share and bond purchases compare price with fundamentals and sell on the same test, instead of fixed-probability rolls (shipped in 6.16.0).
•CEO choice by record. NPP-run companies choose CEOs from NPPs with a track record running companies in the same sector (shipped in 6.17.0).
•Consequences across politics and business. NPP officeholders weigh their holdings in votes (6.8.0), the conflict is shown on their profile (6.13.1), and voters mark down a vote their holdings decided (6.14.0).
•Batched purchases. The share-buying sweep was the largest single NPP cost in a turn (18.0 seconds and 1,465 database round trips on live turn 168). Markets now settle in parallel with one pool read per sweep (6.16.1). The bond sweep is still one purchase at a time.

Stage 4: cost

•NPP governing passes run countries in bounded parallel lanes instead of one after another (shipped in 6.14.2).
•Every engine 6 phase gets a per-turn round-trip budget, reported with the turn.

Stage 5: strength accounting

•Compare v5 and v6 worlds on NPP win rate against players, NPP wealth, and returns on NPP-run companies.
•Step down the income NPPs receive without acting until those measures match engine 5. The election modifiers are not changed.

Stage 6: default for new worlds

•Autonomy v6 becomes the default for new singleplayer and simulation worlds (shipped in 6.14.1). Running worlds never change level on their own. The live multiplayer world moved to v6 with game 1.16.

6.19.0 (game 1.16.0, 2026-10-15)

Stage 3 (corporations).

•NPP-run plants shrink to the market they serve (v6). A plant that sells everything it makes but is held far below its size by weak demand kept paying payroll and upkeep on the idle rest, because sold-out plants were treated as recovering. When an NPP-run plant's output stays under 60% of its size for 24 turns, it now retires part of the idle capacity each turn, keeping room to grow back by half. On live turn 202, loss-making NPP sectors ran at 38-46% of capacity while selling 92-100% of what they made. Player plants are unchanged.

6.18.0 (game 1.16.0, 2026-10-15)

Stage 3 (corporations).

•R&D is never cut to zero (v6). NPP-run companies with thin margins or short cash kept no R&D budget, so their technology score decayed every turn, their product quality fell with it, and lower sales pushed margins lower still. They now keep a small maintenance R&D budget, which is enough to hold technology near par. On live turn 199, NPP companies losing money averaged quality 43 against 70 for those above a 20% margin. Player companies at the same thin margins kept quality above 70.
•Losing companies change course sooner (v6). A company's strategy used to count as working whenever it did better than when it was adopted, so a company that started expanding while losing money kept expanding as long as it lost a little less, and was only re-judged every 48 turns. A strategy that still loses money no longer counts as working, and a loss-making company is re-judged after 8 turns. On live turn 199, 101 of the 120 loss-making NPP companies were still on the expand strategy.
•Fix: a caretaker seat left empty is filled again. A player company handed to a caretaker that was later marked as having no CEO stayed that way: the vacancy step skipped it because it was already caretaken. It now gets a caretaker again the next turn, and the owner's arrangement stays.

6.17.1 (game 1.16.0, 2026-10-15)

•Fix: a successor takes every seat a dead NPP held (v5). When an NPP dies, the NPP that replaces it now also takes its cabinet posts, its place as national leader, and any player company it was running as caretaker. The owner's caretaker arrangement stays in force. Before, those seats kept pointing at the dead NPP.
•Fix: consolidation respects a party chair's slate (v6). A trailing NPP is no longer withdrawn from a primary its party chair slated it to, or from any race the chair is managing for the party.

6.17.0 (game 1.16.0, 2026-10-14)

Stage 3 (corporations).

•CEOs chosen by track record (v6). NPPs now keep a record of how the companies they ran did, per industry: time in charge and the company's operating margin against other companies in the same country and industry. When an NPP-run company needs a CEO, it prefers a free NPP with a good record in its industry. A record counts only after half a game year and only if the NPP beat its peers. With no such candidate, the old rule applies. A CEO a player names is never overridden. The record is kept in every world, so a world switched to v6 already has history.
•Fix: a dead NPP no longer stays CEO. When an NPP dies (autonomy v5 and above), its NPP-run companies now get a new CEO the next turn instead of keeping the dead one.

6.16.1 (game 1.16.0, 2026-10-14)

•Faster share buying (v6). At v6 the NPP share-buying sweep runs each currency's market separately and in parallel, and reads each market's pool once per sweep instead of once per purchase. Tests confirm every purchase still balances to the cent and the result matches buying one at a time. Decisions are unchanged.

6.16.0 (game 1.16.0, 2026-10-14)

Stage 3 (corporations).

•Investing on opportunity (v6). NPPs used to decide whether to buy shares or bonds with a fixed roll before looking at the market, then picked what to buy. Now they look first: the better the best deal on offer (a share price below what the company is worth, or a bond yield above the minimum), the more likely they are to buy. Shares well above what the company is worth are more likely to be sold. Across the market the total amount of buying and selling stays about the same as before; it just goes where the value is.

6.15.0 (game 1.16.0, 2026-10-14)

Stage 2 (governing).

•
Ministers judged on their goals (v6). The head of an NPP government used to judge each minister only on how far its domains sat below target. When the government holds standing goals, a minister whose brief covers one is now judged on that goal instead.
- •A minister turning a failing domain around (at least 5 points of progress since the attempt began) is kept.
- •A minister in post when a goal on its brief failed, with no progress since, can be replaced even if the shortfall looks small.
- •A minister appointed after the failure is not blamed for it.
- •Minimum tenure, cooldowns and the limit on repeated reshuffles still apply.

6.14.3 (game 1.16.0, 2026-10-14)

•Ministers save their work in batches. NPP and caretaker ministers load their settings and orders once per country and save changes in single batched writes, instead of several database trips per minister. Applies at every autonomy level; decisions are unchanged.

6.14.2 (game 1.16.0, 2026-10-14)

•Governments plan in parallel (v6). At v6 the NPP governing phase runs up to four countries' domestic steps at once, then foreign policy one country at a time in the usual order, because one government's foreign moves shape the next. Below v6 it runs one country at a time as before. Decisions are unchanged.

6.14.1 (game 1.16.0, 2026-10-14)

•V6 is the default for new local worlds. New singleplayer and simulation worlds now start on autonomy v6, so they run every engine 6 change. Existing worlds keep their level, and the live multiplayer world stays on its own setting. The desktop client lists V6 and recommends it. No decision changes.

6.14.0 (game 1.16.0, 2026-10-14)

Stage 3 (corporations).

•Voters notice self-interested votes (v6). When an NPP's own companies decide its vote on a bill, it loses 1 favorability. "Decide" means it would have voted differently without them. This happens at most once per voting turn and never takes favorability below zero. Business interest is no longer free: an NPP that votes its holdings pays a small price with voters.

6.13.1 (game 1.16.0, 2026-10-14)

•Business interests are visible (v6). When an NPP's own companies pull its vote (6.8.0), the vote forecast on its profile now says so, naming whether the bill helps or hurts a company it runs or holds. Players and opponents can see the conflict. No decision changes.

6.13.0 (game 1.16.0, 2026-10-14)

Stage 2 (governing).

•
Central banks defend a falling currency (v6). Since 6.6.0 a central bank run by an NPP adjusts its interest rate when its currency moves, but in limited steps that a sharp fall can outrun. Its other tool, a trading band defended with the bank's reserves, was only ever used by player chairs.
- •Now, once a floating currency has fallen 15% or more over 24 turns, the NPP chair sets a band. Its top sits 5% above today's rate, which caps further weakening, and its bottom 10% below.
- •The bank defends the band with its own reserves under the usual rules, so it can never spend more than it holds.
- •The chair removes its own band once the currency's 24-turn move is back within 5%, after the normal wait between band changes.
- •A band set by a player is never touched, and government-held rates, pegs and shared currencies are left alone.

6.12.0 (game 1.16.0, 2026-10-14)

Stage 1 (running for office).

•Racing where the party can win (v6). Since 6.1.0 a party runs its strongest NPPs for its most senior races. Where the party is weak across the whole region, those regionwide races are the ones it cannot win, so that spent its best candidates on lost causes. Now a party whose organization in the region is below 75% of the leading party's runs its strongest NPPs in its most local races instead, where it can still win seats. The same races are contested either way.

6.11.0 (game 1.16.0, 2026-10-14)

Stage 2 (governing).

•The opposition picks its battles (v6). The opposition's NPPs used to vote as a bloc against every government bill, even against a government with a large majority, where the bloc could defeat nothing and only stopped opposition members backing bills they agreed with. Now the bloc holds only while the government side (governing party plus coalition partners) has less than 55% of the lower chamber. Above that, opposition NPPs vote on each bill's merits, home region and donors like any other member. The pre-filing vote count from 6.5.0 uses the same rule.

6.10.0 (game 1.16.0, 2026-10-14)

Stage 2 (governing).

•Vote counts include the government whip (v6). Since 6.5.0 a sponsor counts votes before filing, and since 6.7.0 an NPP government whips its own members on its goal bills. The count did not know about that whip, so it could set aside a goal bill the whip would have carried. The count now includes it, so the sponsor judges a goal bill the way it will actually be voted.

6.9.0 (game 1.16.0, 2026-10-14)

Stage 1 (running for office).

•Career ladders (v6). NPP incumbents used to re-file for the seat they held every cycle, so senior offices were filled only by NPPs without a seat. Now an ambitious incumbent with real standing can run for an open race for a more senior office in its home region instead. An open race is one with no candidate or defending incumbent from its party. This applies to an incumbent with ambition of at least 70 and political influence plus favorability of at least 100, holding a single seat. Its old seat is then filled the ordinary way. A party chair's slate assignment still takes precedence, and nationwide races are not part of the ladder.

6.8.0 (game 1.16.0, 2026-10-14)

The first stage 3 change (corporations).

•
Business interests shape votes (v6). An NPP used to vote on a bill that subsidises, protects or nationalises an industry as if it owned nothing. Now an NPP that runs a company, or holds at least 1% of one, weighs what the bill does to that company's sectors.
- •It leans for subsidies and sector tariffs in its own sectors.
- •It leans against ending those subsidies, tariffs aimed at its own company, and nationalisation of its company or sector.
- •Running the company counts twice as much as holding shares.
- •The pull is capped below a hard party whip, so party discipline and the NPP's own positions still matter.
- •Players can use this: a sector bill can win over the NPPs invested in that sector.

6.7.0 (game 1.16.0, 2026-10-14)

Stage 2 (governing).

•
Governments whip their own program (v6). An NPP government's members used to back its bills only through the general lean toward their own party's bills. Now, when an NPP heads the government, its members get a soft whip on the bills the government sponsored to serve one of its standing goals. Each member still weighs it against their own positions, home region and donors, and loyal members follow it more than stubborn ones, as with any whip.
- •An existing whip always takes precedence.
- •It is never applied in a party with a player as chair or vice-chair.
- •It is never applied to members of other parties.

6.6.0 (game 1.16.0, 2026-10-14)

Stage 2 (governing).

•
Central banks answer currency moves (v6). An autonomous central bank chair used to set its rate from inflation and growth alone, so a currency could fall sharply and the bank would not respond until the higher import prices showed up as inflation. Now a floating currency's move over the last 24 turns also counts.
- •A fall of more than 5% raises the bank's target rate by 0.1 points for each further 1%.
- •A rise of more than 5% lowers it at half that rate.
- •The shift is capped at 3 points.
- •The bank still moves toward its target in the same limited steps, with the same wait between changes, and the rate history records the currency move as part of the reason.
- •Pegged currencies and currencies under a shared central bank are not affected.

6.5.0 (game 1.16.0, 2026-10-14)

The first stage 2 change (governing).

•Counting votes before sponsoring (v6). An NPP sponsor used to file its highest-scoring bill whether or not the chamber would pass it. Now it first projects the vote in the chamber the bill starts in, using the same pressures NPPs vote with when no whip is set: their own positions, their home region, their donors, the pull toward their own party's bills, and the opposition bloc against government bills. Player-held seats are counted as voting with their party. A bill that does not project more votes for than against is set aside and the sponsor tries its next-best bill, up to three. If none has the votes, it files nothing that turn and keeps its slot. Tax-rate bills, whose rate is set when they are filed, are not counted.

6.4.0 (game 1.16.0, 2026-10-14)

Stage 1 (running for office).

•Trailing NPPs leave the primary (v6). In the last 12 turns before a primary closes, an NPP whose strength (political influence plus favorability) is below 60% of its party's strongest candidate in that primary withdraws. It does not endorse anyone; NPP endorsements stay something players request. Incumbents defending their seat never withdraw, and a primary with any player candidate is never touched. A withdrawn NPP cannot re-file in that primary.
•No late challengers (v6). NPP challengers no longer enter a primary in those last 12 turns, so a field that has consolidated stays consolidated.

6.3.0 (game 1.16.0, 2026-10-14)

Stage 1 (running for office).

•Presidential primary challengers (v6). NPPs could already challenge a same-party candidate in state and legislative primaries, but presidential primaries were left out. Now, in countries with no players, a presidential primary where a party has one candidate can draw a challenger from the same party. The challenger is picked from the party's five strongest available NPPs across the country, with ambitious NPPs more likely to run. The chance of a challenge and the limit of two challengers per country per turn are the same as in other primaries. Presidential primaries in countries with players are not challenged.

6.2.0 (game 1.16.0, 2026-10-14)

Stage 1 (running for office).

•
NPP candidates rally their own race (v6). An NPP's campaign action used to raise only its general political influence. NPPs have no campaign organisation, so their candidate Support, the race-level momentum that rallies move for players, stayed at the neutral 50. Now the first campaign action of a cycle by an NPP that is running in a race where players can rally becomes a rally for that race.
- •It costs the same action points and funds as the campaign action, and gives up that action's influence gain.
- •It moves Support by the same amount a player rally in the same kind of race does, with the same split between an immediate gain and a gain spread over the next four turns.
- •Players can rally once a turn. An NPP rallies at most once per action cycle, which is once every four turns.

6.1.0 (game 1.16.0, 2026-10-14)

The first stage 1 change (running for office).

•Strongest candidates in the senior races (v6). When a party fills a state's open primaries with NPPs, it now runs its strongest available NPPs for the most senior offices. Strength is the NPP's political influence plus favorability. The same primaries are filled as before; only who runs where changes. Before, NPPs were taken in load order and the lowest office was filled first, so the first NPP loaded ran for state senate and the last for governor.
•Nationwide executive slot (v6). A party's nationwide executive primary slot goes to its strongest available NPP instead of the NPP with the lowest id.

6.0.0 (game 1.16.0, 2026-10-14)

The foundation for the overhaul.

•Autonomy tier v6. A new tier above v5 that switches on engine 6 behavior. It adds no activity and no resources: a v6 world allows exactly what a v5 world allows.
•Engine version. The engine now carries its own version number, recorded on every turn log, and this changelog lists each version.
•Staggered action cycle (v6). NPPs take an action cycle once every four turns. Below v6 they all act on the same turn, which put the whole cost on one turn in four (27.6 seconds on live turn 168, mostly share and bond buying). At v6 each NPP has a fixed lane and a quarter of NPPs act on each turn. Each NPP still acts once every four turns with the action points it built up over that time. Selling shares, building party organisations and founding companies stay on the shared fourth turn.

5.0 (game 1.7.0, 2026-09-06)

•Persistent governing goals (autonomy v5). A government keeps up to five standing goals across agenda recomputes, grades each one as achieved, revised or failed, and feeds the result into its next agenda. Ministers hold a committed tier unless it stops advancing the goal, and bill sponsorship leans toward bills that serve an active goal.
•Ageing and succession (autonomy v5). NPPs age out and are replaced by successors.
•Difficulty that changes decisions. Local-world difficulty is split into resources and decisions, so a resource advantage is never presented as skill.

4.1 (game 1.0.0, 2026-07-20)

•Context-aware choices (autonomy v3 and v4). NPP actions respond to election timing, contested seats, government status and party finances, and stop spending on actions that can no longer change anything.
•Individual variation. Personality blends continuously, each NPP has stable quirks, and stubborn NPPs act more decisively than flexible ones.
•Limits on player impact. Caps on NPP bill sponsorship in player countries and on primary challengers against a player, a cooldown on NPP corporate attacks against the same player, and a 25% share of each market's float that NPP bond and share buying leaves untouched.
•Tier fixes. Two activities that switched off when a world moved from v3 to v4 now stay on at every higher tier.

0.0 to 4.0 (game 0.4.0, 2026-07-04)

The graduated autonomy ladder replaced the old on/off switch.

•v0: party organisation votes, and filling stalled executive seats in non-player countries.
•v1: the governing brain: agendas, cabinets, ministerial action, fiscal stance and opposition.
•v2: caretaker ministers and CEOs, corporate rivalry, and the first tier active in player countries.
•v3: campaigning, fundraising, saving, investing and founding companies, in non-player countries.
•v4: the v3 activities in every country.