Ideas how to improve the Espionage System

[axatin]

The new patch includes some changes to the Espionage UI as well as some bugfixes, and I feel this is a good opportunity to discuss the Espionage system in more detail, because in my opinion it needs some reworking. There was a previous thread about espionage (thanks to azum4roll for pointing it out) but I'd prefer to open a new one as there have been some changes since then. Before discussing possible improvements, I want to describe how espionage works in detail:

Spoiler :

Spy Mission Duration:
Each spy mission has a base duration, to which two percentage modifiers are applied. The Spy Power is a negative modifier, so it reduces mission length, and the City Resistance is a positive modifier and increases mission length. City Resistance is based on the Security Level, the value that is shown as a star rating for each city in the espionage screen. The total mission duration is always between 50% and 150% of the base mission duration.

Spy Power:
The base spy power for each spy is 25. To this value, percentage modifiers are applied:

• +50% if playing on Quick Game Speed
• +25% for each level the spy is above the first rank.
• +25% for open borders
• +25% for England

Security Level:
The Security Level is a value between 5 and 50 (shown as a star rating) that changes during the game. Each city is initialized at level 50 when the first spy is recruited. Each turn, city security is modified based on the Spy Resistance, which in turn is based on city population, trade routes and other factors. Spy Resistance is shown as a percentage value, and it modifies City Security each turn by (Spy Resistance value)/20. So, if the shown value is -20%, City Security will decrease by 1 each turn.

Spy Resistance is the sum of the following:

• -10% per trade route going to or coming from the city
• -2% per population
• a negative value for city unhappiness which is calculated as (City Population - Local Happiness) / (City Population) (*)
• +50% if a counterspy is present in the city
• +50% for Constabularies, Police Stations and Military Bases, +15% from NIA
• +10% for each Public Works
• +34% for the "Empiricism" Policy (Rationalism Branch)

After each completed foreign Spy Mission, City Security is reset to the maximum value of 50.

(*) This seems to be a formula that has been forgotten to update, as it is exactly the formula for local unhappiness from the base game (where 1 Pop = 1 Unhappiness), and all of the VP unhappiness sources are completely ignored in it.

(In this city, City Security decreases by 8/20 = 0.4 per turn)

City Resistance:
City Resistance, the negative modifier for spy mission duration, is calculated based on the Security Level explained above. To this value, percentage modifiers are applied. These modifiers are:

• The Spy Resistance percentage modifier that has already been used to calculate the change per turn of city security
• The difference in the number of policies between spy owner and targeted civ, multiplied by -10%, capped at +-50% (so spy missions are quicker if the targeted civ has more policies than the spy owner)
• The difference in the number of technologies between spy owner and targeted civ, multiplied by -2%, capped at +-50%
• A modifier based on the number of previous spy missions done by the civilization in that city (the square of the number of missions: +1% for one previous mission, +4% for two missions, +9% for three, etc.)

ID and Kill Chance:
These chances are calculated as a product of two factors. The first factor is (City Security * 2)/100. The second one is based on the mission type and possible counter-spies in the city. Each mission has a base ID and Kill Chance (see below). To the mission ID chance the value of +15 is always added. If there is a counter-spy in the city, both ID and kill chance are increased by +25 for each level the spy is above the first rank. Counter-espionage missions increase ID and kill chances as specified in the mission descriptions.

The total ID chance can't be smaller than 10%, and the total kill chance can't be higher than 75%. Spies are killed after completing their missions, so the mission result is always applied.

Spy Experience:

• For completing an offensive mission, a spy gets 50XP plus a value based on the mission
• Successfully rigging an election gives 20XP, a coup gives 50XP.
• Killing an enemy spy gives 50XP
• As a diplomat, uncovering intrigue gives 20XP (Until this version, there was a bug in the code that if a wonder is constructed in the target city, it gives 20XP to the diplomat for every turn in which the wonder is under construction, causing diplomats to rapidly level up. This has been fixed.)

Issues related to the espionage system and mission duration calculation in general:

• Spy resistance usually has a high negative value (at least until late in the game, when it can suddenly become a very high positive value), so most of the time, city security is constantly at the minimum value, unless it has recently been reset by a completed enemy spy mission
• The first point, together with the fact that counter-spies don't actually prevent enemies from completing their missions, makes counter-spies basically useless
• There is a strange double-dipping wrt spy resistance, which is used in two separate calculations, and which is also a value that's not controllable by the player at all
• If the modifier to city security (spy resistance plus modifiers from having more/less techs or policies) is close to -100%, city security doesn't matter at all, because the final value will always be 0
• Conversely, the value of that modifier doesn't matter at all if the city security is close to 0

Issues related to specific missions:

• The reward from the siphon missions is calculated based on the duration of the mission, meaning that mission duration effectively doesn't matter at all. There is no difference in the reward for completing two missions of 30 turns each and the reward for three missions of 20 turns each
• Yields are stolen as instant yields, which (a) encourages the gamey behavior of always spending all the gold one has to prevent it from being stolen, (b) can be frustrating and (c) contributes to overall yield inflation as the spy owner always gets the full amount even if that's more than can currently be stolen
• Compared to the other missions, the "copy great work" mission is very weak

Ideas how to improve this:

(1) I would like to keep the system with the variable city security in general, but make it more controllable for the player, and more relevant. The idea is to make the values by which city security changes per turn much lower, so that we have a more gradual change, and to remove the re-setting of city security after a completed mission. City security should be more of a long-time investment.

Spoiler :

The negative modifiers need to be drastically reduced and we need some inherent positive modifier to make sure city security is not always falling. Initial idea (all values are the amount by which city-security changes per turn):
Negative:

• -0.04 per population
• -0.1 per excess unhappiness
• maybe a negative modifier for trade routes, if it is necessary to punish tall players more

Positive:

• +0.01*(City Strength)
• +0.08 per Public Work
• +1 for a counterspy
• +0.5 for Defense Process
• +0.2 for constabularies, police stations

Remove the spy effects from the military base, and add instead two more buildings in the Constabulary / Police Station line (two or three techs apart each)

(2) The modifiers for more or less policies/techs guarantee that espionage is a catch-up mechanic (which is in my opinion sorely needed given that most games are decided long before they're finished). They should be applied to the mission length directly and not to city security.
New Modifiers: +-4% for each policy, +-4% for each tech, both capped at +-20%.

(3) If a counterspy kills an enemy spy, their mission is not completed, and no mission results come into effect. To make up for this, base mission durations are reduced and an additional modifier to kill chance is added: -10% kill chance for every level the operating spy is above the first level.

(4) The siphon missions don't result in an instant yield, but in a yield change per turn. When a mission ends, a certain amount of gold will be transferred from one civ to the other each turn (like a trade deal). The transferred amount is based on the average of the city yields while the spy was in it to conduct the mission. The effect duration is always the base mission duration (so operating quicker than the base mission length will be rewarded).

(5) The "Copy a Great Work" mission is changed to "Steal a Great Work".

(6) If a spy has not been ID'd after a mission, the next mission can be chosen and is no longer automatically the same as before.

I'm thinking about making some proposals next session based on these ideas, and I would very much like to hear your thoughts about it. Or also more generally your thoughts about the espionage system and how to improve it.


One last thing: In case someone doesn't like the espionage system at all, with the new patch it should be possible to revert it back to the original system (only tech stealing, no events) by setting the option BALANCE_CORE_SPIES_ADVANCED to 0 in Vox Populi/Core Files/Core Changes/CommunityPatchDLLChanges.xml:

Spoiler :

I haven't tested it that thoroughly, but it should work (except for some incorrect tooltips). If you do so, I'd recommend also to set BALANCE_CORE_SPIES to 0 in Community Patch/Core Files/Core Changes/CommunityPatchDLLChanges.xml (this controls spies scaling with map size, having that many spies to steal technologies would be too strong).

 

[usadefcon1]

I think that an ideal system would look like this:

Actions should be percentage based on what the spy finds when they get there (if the city has scientist-citizens then tech stealing should have a higher chance of being successful, etc...)

Show the player what the odds are and then have the player make the choice of what action to persue.

An example would be:

Our adversity has discovered the secret of Rifling, what action should I take?

A) Attempt to Eavesdrop on the secret (50% to steal tech, 25% of failure, 25% of being killed)

B) Bribe the scientists who discovered it (Costs 100 gold, 80% steal tech, 20% of being killed)

C) Remain Undercover (No risk, attempt to find other technologies they have researched)

An example on the defensive side would be:

I have found an enemy spy in our city, what should I do?

A) Kill the enemy spy immediately (70% of eliminating the spy, 10% chance of finding their origin, 20% they escape)

B) Track the enemy spy (50% of finding their origin, 50% they escape)

C) Bribe them, costs gold (80% of finding their origin, 20% they escape)

 

[Revolutionist_8]

The new patch includes some changes to the Espionage UI as well as some bugfixes, and I feel this is a good opportunity to discuss the Espionage system in more detail, because in my opinion it needs some reworking. There was a previous thread about espionage (thanks to azum4roll for pointing it out) but I'd prefer to open a new one as there have been some changes since then. Before discussing possible improvements, I want to describe how espionage works in detail:
...

I think your proposals and arguments make a lot of sense. It'd be a significant quality improvement, without some enormous changes or completely new mechanics instead of the current one (which I personally really like, just needs some adjustments). I'm absolutely in favor of it!

Moreover, I've the feeling that you can actually implement these things, so that's also a plus (= that we don't have to wait for too long )!

One thing that maybe missing, is that Great Works can not be stolen from Capitals as far as I know, right? I think it should remain (if it's really stealing).

I think what also should be taken into consideration, is to have the ability to really have an impact against enemy spies. What I mean by that is having the building for that era, combined with having a high level spy on a counter-spy mission should prevent enemy spies with a very high chance. I think your proposals would ensure something like that (hard to tell just from written numbers), I just thought I might as well emphasize it, as it has been a real concern for many players for quite a while now.

 

[Stalker0]

I will say the new system in this version is a big improvement. Now that the UI is working properly and coups have been reigned in, I do find spying has a more solid foundation now. The immedate change to me, the infiltrate city option (the very first option) has way too long a duration, 20 turn based on standard speed. Sorry I'm not wasting 20 turns of spy time to get some vision.

Second, I have theorized that tourism stealing could ultimately be too strong, and if humans/AI got smart about it, constantly taking the tourism of a tradition capital could effectively shut down CV options.

Third, I do wish spy xp gains were a bit higher. When I do a 30 turn spy mission, by gum my spy should level!

 

[azum4roll]

If you can code it to start an event of siphoning yields BEFORE the mission actually completes, I have the following requests too:

1. There should be a notification of "your yields are being stolen for X turns" to the target, so you don't have to keep track of yields of every city every turn to see if you're being spied on.
2. However, there's a possibility of instantly assigning a counter-spy to try to make the mission fail. Are the yields returned if it fails? Should we fix the identify/kill chance at the start of mission?
3. This can be applied to tactical missions like Infiltration and Incite Rebellion. Complete the mission right away, but the spy is stuck for X turns after.
4. You can likely code the counter-spy missions I proposed in the old thread too. I've modified it slightly to account for the changes you proposed.

On establishment in owned cities, a spy would immediately identify all foreign spies of same or lower level acting in the city. Then you can pick from two actions:
1. Immediately attempt to capture/kill the foreign spies. All foreign spies immediately end their mission, and have a chance to be killed based on the original kill chance and spy level difference. Already started spy events immediately end (lose vision, end resistance, end yield siphoning, end specialist kidnapping).
2. Tail the foreign spies. They'll continue their mission with 100% identification chance and higher kill chance. You gain science and culture based on the number of spies acting in the city (like Police Station).

If there's no foreign spy, the city gains X happiness for 10 turns. Both actions last for 10 turns, but the tailing event still continues until all foreign spies leave the city.

 

[alchx]

Mission durations should definitely be revisited. Minor missions should have very short deadlines

Provide visibility around the city - I propose to remove it or make a maximum of 5 moves. I don't understand why it became a mission. This is originally a free bonus of finding a spy in an enemy city in vanilla. By the time the spy arrives in the city, he spends at least 5 moves - travel plus penetration and deployment of an intelligence network. We use spies to survey the territory around the city only during the war to remove the fog of war - in these cities we need sabotage, uprisings, theft of the treasury, the destruction of buildings, farms. Any missions that will reduce the city's defenses and production capacity.

Destruction of farms around the city - moves 8-10. 3 farms will regenerate by workers 6-9 turns plus tile movement time (which can be =0 if there are roads). Not the most useful mission to use a spy. Extremely situational. In addition, the spy can be killed, and this is a very ridiculous waste of a spy.

Theft of treasury, faith, science - moves 8-10. I see no reason to steal the treasury for 20-30 moves. These are basic missions for 'familiarization' with the city and development of spy skills. After completing them, you can move on to more difficult missions, if the rank of the spy allows. In addition, the short duration of the mission leaves no time to deploy a counter-intelligence officer if the enemy notices a drop in the city's income or increased religious pressure coming from a spy.

I will not comment on the missions of destroying buildings, damaging the garrison, stealing specialists. I didn't use them very often, but causing damage to the garrison and the impossibility of repairing the city for 5 turns is a very situational mission. Usually, by the time it is completed, either my army is very far from the city, or there are so many armies near the city that the mission loses all meaning.

I agree with the replacement of the 'create a duplicate' mission with the 'art theft' mission. We take risks and waste time. If we are successful, then the enemy must suffer and lose the original. I'll leave the duration without comment.

The longest mission should obviously be stealing the technology. The mission offers a very significant reward. But here, too, a balance is needed. For starters, you can try to limit the duration of the mission to 30 moves. Stealing technology for 40-60 moves in later eras is completely useless. After turn 400, I do not do this and, if necessary, simply switch cities to direct science generation. On average, I managed to steal 3 technologies for the entire game. It seems to me that this is very small. In addition, in one patch, the price of late technologies was increased. To begin with, it is worth trying to balance the duration of the mission so that it is possible to steal 5-7 technologies per game. This is taking into account the risks of spy kills, counter-spy buildings, policies

 

[Stalker0]

One new mission that is 100% needed, delay work by 1 turn.

There are so many situational reason this is needed. Case in point, I sent a spy to Beijing. It just so happens that on the turn I set up my mission, beijing used an inquisitor and is now in rebellion. So that means I can't do any of the steal missions, because there is "not enough of that yield to siphon". Yet if I waited just one turn, I'd be back in business. Considering it can take 5-6 turns to setup a spy, that is infuriating. I want to be able to delay.

 

[Stalker0]

The longest mission should obviously be stealing the technology. The mission offers a very significant reward. But here, too, a balance is needed. For starters, you can try to limit the duration of the mission to 30 moves. Stealing technology for 40-60 moves in later eras is completely useless. After turn 400, I do not do this and, if necessary, simply switch cities to direct science generation. On average, I managed to steal 3 technologies for the entire game. It seems to me that this is very small. In addition, in one patch, the price of late technologies was increased. To begin with, it is worth trying to balance the duration of the mission so that it is possible to steal 5-7 technologies per game. This is taking into account the risks of spy kills, counter-spy buildings, policies

The number 1 problem with Tech Steal right now is....its city agnostic. Doesn't matter if your capital is fort knox, you can just hop over to little dinky city and steal a tech. So your only as strong as your weakest link. further, this means you can send multiple spies to a civ all at once and steal multiple techs.

 

[alchx]

The number 1 problem with Tech Steal right now is....its city agnostic. Doesn't matter if your capital is fort knox, you can just hop over to little dinky city and steal a tech. So your only as strong as your weakest link. further, this means you can send multiple spies to a civ all at once and steal multiple techs.

I remember that in vanilla, technology theft speed was affected by the science output of the city and the fastest theft was in capitals and / or science cities with several Academies. Trying to steal technology in a small undeveloped city was pretty useless, mission duration was about 3 times longer.

 

[Tekamthi]

I like the new spy system better than the old, though it still feels slightly disconnected to the rest of the in-game world, and I agree with OP sentiment that it could be developed further. Changes discussed so far seem on the right track for smaller, incremental changes than come to mind personally. I think potential exists for even greater integration with the game-world, while tackling some of the issues raised in this thread:

• spies should have to move to their target city, not just teleport (at least in the pre-flight era and/or to cities without airports). If you're moving to capital on other side of the world, part of the fun should be getting there. We can't do on-map units like civ 2, that would be complete, impossible rework, but the existing spy system could have pre-flight movement rules, ie spies can relocate a max of x tiles (expanding over time with tech developments) in one "jump", or move along any trade routes connecting their current city to a more distant one. Same applies to fleeing spy, should have to get away without being caught -- this would add depth to the existing spy system just by piggy-backing existing mechanisms: checks for plot distances and trade routes are trivial. With some AI adaptations to this spy-movement paradigm, the whole spy v spy, cat and mouse aspect might really come alive.
• tech steal should depend on whats in the city when the mission starts -- eg any buildings in city allow that building's required tech to be stolen, same with garrison unit, units in production etc. If there are no "tech a" buildings or units or production in the city etc., then "tech a" cannot be stolen from that city. This would sort out the small cities somewhat without relying on arbitrary limitations or opaque math formulas

 

[azum4roll]

The number 1 problem with Tech Steal right now is....its city agnostic. Doesn't matter if your capital is fort knox, you can just hop over to little dinky city and steal a tech. So your only as strong as your weakest link. further, this means you can send multiple spies to a civ all at once and steal multiple techs.

Which is why there should be an additional constant value for mission duration that's not affected by security. Should be set to a high value for tech steals.

 

[Stalker0]

Which is why there should be an additional constant value for mission duration that's not affected by security. Should be set to a high value for tech steals.

That's not enough, because I still can send 4 spies to your civ and steal 4 techs simultaneously, creating a massive slingshot. there needs to be some restriction that you can only have 1 spy on a tech steal mission per opponent or something like that.

 

[axatin]

I think that an ideal system would look like this:

Actions should be percentage based on what the spy finds when they get there (if the city has scientist-citizens then tech stealing should have a higher chance of being successful, etc...)

I would say that for the offensive part we already have a quite similar system. It's outcome is not random, but other than that we already move spies to cities to check what's there and then we decide which missions to do based on what the city's strengths are. And I'm unsure if more randomness would make the system more attractive... I agree about the defensive part, will come back to it later.

think your proposals and arguments make a lot of sense. It'd be a significant quality improvement

Thank you! I'm not sure about the steal GW restriction to non-capital cities, will check it later, but it can of course be implemented if needed.

spies should have to move to their target city, not just teleport (at least in the pre-flight era and/or to cities without airports). If you're moving to capital on other side of the world, part of the fun should be getting there. We can't do on-map units like civ 2, that would be complete, impossible rework, but the existing spy system could have pre-flight movement rules, ie spies can relocate a max of x tiles (expanding over time with tech developments) in one "jump", or move along any trade routes connecting their current city to a more distant one. Same applies to fleeing spy, should have to get away without being caught -- this would add depth to the existing spy system just by piggy-backing existing mechanisms: checks for plot distances and trade routes are trivial. With some AI adaptations to this spy-movement paradigm, the whole spy v spy, cat and mouse aspect might really come alive.

If we implement those rules, then we can actually also make spies on-map units, invisible to others and ignoring borders, and once they're in a city, you click on a button and they disappear from the map and are available in the espionage screen. I like the idea because teleporting really is a bit lame, but as you said, it would be a major rework, and I'm not sure if the effect would be worth the time needed for the implementation. At least for now, I'd say there are more important issues.

tech steal should depend on whats in the city when the mission starts -- eg any buildings in city allow that building's required tech to be stolen, same with garrison unit, units in production etc

Hmm, I think that would encourage gamey behavior. Going for a science victory? Better not build any medical labs and station the helicopter gunships outside the cities...

Mission durations should definitely be revisited.

I agree, it would be part of the proposal to make the missions shorter. I think your proposed mission lengths are maybe a bit too short, espionage shouldn't be too much micro-management and also not too annoying for the players on the receiving end, but maybe 20 turns for siphoning (10 through 30 after modifiers) and 15 or even 10 for destroying improvements would be reasonable.

 

[axatin]

the infiltrate city option (the very first option) has way too long a duration, 20 turn based on standard speed

Provide visibility around the city - I propose to remove it or make a maximum of 5 moves. I don't understand why it became a mission

One new mission that is 100% needed, delay work by 1 turn.

Maybe we make the increased visibility the standard effect of a spy having established surveillance in a city. Increased visibility is always active as long as a mission in being conducted, and if a spy arrives in a city and no missions are available, the spy is set to surveillance mode (as in the base game, when there's no tech to steal) , also has the increased visibility and an event popup shows up once a mission becomes available.

The number 1 problem with Tech Steal right now is....its city agnostic

I'm a bit confused now, hasn't it been changed some patches ago that the tech steal mission is only available in the capital? Or was this planned and never implemented? (in this case my strategy of using level 3 spies only in capital cities was not optimal ^^)

Anyway, limiting tech stealing to capitals would be the best solution to the problem in my opinion.

Third, I do wish spy xp gains were a bit higher. When I do a 30 turn spy mission, by gum my spy should level!

The easiest solution for that would be removing the limitation that xp is always set to 0 when a new level is reached. Currently, when a spy has 75 xp, and completes a mission worth another 75 xp, it levels up and has 0 xp on the new level then.

Re: the tourism missions, maybe it would really be good to merge them, as proposed by @Stalker0 on discord. Or at least swap them, so that copying a gw is the level 2 mission (stealing a gw would probably be too strong in this case) and siphoning tourism the level 3 one.

 

[axatin]

If you can code it to start an event of siphoning yields BEFORE the mission actually completes, I have the following requests too

This was my first idea when I thought about the siphon missions: Every mission, a certain amount of gold/science/etc disappears, and when the mission is completed, all the collected gold is transferred to the spy owner or back to the target player (if the spy was killed). The missions are somehow written to follow this design. The big disadvantage of this would of course be that the targeted player would immediately know that a spy mission is being conducted against him. Even if we hide all notifications and make it an obscure instant yield, a skilled player could still find it out and in most cases could also identify the city in which the enemy spy is. Shouldn't spies be undetectable except for counter-spies?

Re: counter-spies: The proposed changes here would already increase their value (at least that's the intention). We must be careful not to make them so strong that espionage is killed off completely. A mission to immediately detect all enemy spies and force them to abort their missions would do exactly that, and it would also encourage using counter-spies to hop from city to city to successively hunt spies. One thing I liked about the coup chance proposal that recently passed is that is makes long-term stays in a city more valuable and faces the player with the decision to take immediate, high-risk action or wait longer for a more secure approach.

We currently have one mission with 50% ID and 50% kill chance. What about adding a new mission with 100% ID chance, 25% kill chance, and it is immediately shown if any spy missions of a lower level are currently being conducted? So we have the choice between a more aggressive stance (with the risk of wasting our time because no enemy spies are present) and a mission that guarantees us that enemy espionage is slowed down (all spies have to return home after their missions, enemy players can be confronted and asked to stop spying on us) and that also shows us if it is worth keeping the spy in the city in the first place.

 

[alchx]

This was my first idea when I thought about the siphon missions: Every mission, a certain amount of gold/science/etc disappears, and when the mission is completed, all the collected gold is transferred to the spy owner or back to the target player (if the spy was killed). The missions are somehow written to follow this design. The big disadvantage of this would of course be that the targeted player would immediately know that a spy mission is being conducted against him. Even if we hide all notifications and make it an obscure instant yield, a skilled player could still find it out and in most cases could also identify the city in which the enemy spy is. Shouldn't spies be undetectable except for counter-spies?

If profitability begins to fall, then this will definitely be a good reason to place your counter-spies. Instead of constant coup attempts and election fraud in CSs.

You can simply reduce the chances of identifying spies, since due to identification, spies are obliged to return home (some chances of failure should remain, since even in reality there are various incidents and accidents). And increase the chances of identification if there is a counter-spy in the city with his own network of informants (but do not report this modifier at the start of missions).

 

[Tekamthi]

If we implement those rules, then we can actually also make spies on-map units, invisible to others and ignoring borders, and once they're in a city, you click on a button and they disappear from the map and are available in the espionage screen. I like the idea because teleporting really is a bit lame, but as you said, it would be a major rework, and I'm not sure if the effect would be worth the time needed for the implementation. At least for now, I'd say there are more important issues.

Agreed, it would take some careful thought to make this spy-transit game enjoyable and not just tedious busy-work. I am imagining some existing features will be tightened up to exclude or greatly restrict the smaller cities (ie tech steals); in such a scenario, the need to transit through the smaller, lower security cities could keep them relevant to the spy game without them being the main targets.... best to nail down the existing features first though.

Hmm, I think that would encourage gamey behavior. Going for a science victory? Better not build any medical labs and station the helicopter gunships outside the cities...

You'd still have to build the helis though, and the suggested steal rules would leave a window open for at least 1 turn (at best player buys helicopter and moves it out next turn -- sharp-eyed spy starts mission during this turn to steal tech). Or maybe it can include units in city-radius. As far as the buildings go, the competing priorities of deciding whether to exploit new tech immediately, and/or to protect these investments with counter-spy and anti-espionage buildings, or focus on other projects first, suits the spirit of civ -- we need more game-related decisions tying this system to the game world. The thematic realism of connecting tech steals to the actual tech present in city is highly-desireable to me, though i tend to lean this way more heavily than most.


Another thought, unrelated to above: have you considered somehow tying diplomats and envoys etc into the spy system. Civ 2's system, arguably the best of the series, had diplomats stationed in city act as defensive units against espionage. While i enjoy their presence in VP, I've always found these diplomacy units very one-dimensional...

 

[azum4roll]

That's not enough, because I still can send 4 spies to your civ and steal 4 techs simultaneously, creating a massive slingshot. there needs to be some restriction that you can only have 1 spy on a tech steal mission per opponent or something like that.

It could be done in the old version.

 

[azum4roll]

it would also encourage using counter-spies to hop from city to city to successively hunt spies

Except you can't, since the spy's still stuck in the city after the mission for X (adjustable) turns. The spy also can't go on the offensive missions during this.
The foreign spies also won't know about the counter-spy until the kill mission is used, or their own mission ends (which shows the "identified" message and triggers diplomatic dialogues).

 

[azum4roll]

Draft of proposal:

All these are assuming (1), (2), and (6) in the OP are already in effect.

Note that currently spies never die in a mission if there's no counterspy in the city. This proposal completely removes any kill "chance" from espionage.

Spy resistance isn't factored in city resistance anymore. Only tech/policy differences are! To recap, city resistance only modifies mission duration.

Steal technology
Only usable in a capital. But there is no restriction to multiple spies performing this mission in the same city.

Broaden surveillance in city
Now automatically in effect after establishment.

Stay put
1-turn mission that does nothing and always has 0% identify and kill chance. Alternatively if possible, add a "Select Mission" button to the spy UI and this mission lasts forever.

Arm the local populace/Target local city defenses/Radicalize the rural citizenry
Event fires on the next turn instead of on mission completion, but mission duration stays the same.

Siphon missions
On selection, the siphon event triggers immediately, and the target is notified. The event lasts a fixed amount of turns like kidnap specialist, with siphon% based on target security level (higher security level = lower siphon%).
The spy mission still lasts for a variable number of turns depending on target city resistance, and the identify roll/kill check is done on the last turn of the mission, using the starting security level. If not killed nor identified, the spy stays in the target city and can pick a new mission. If identified/killed, the spy returns to the hideout/dies, and it triggers the diplomatic dialogue + penalty.

Sabotage local tourism
Percentage lowered to 50%. Now also forge a GW on mission completion, if any. Need a new name.

Create a forgery of a Great Work in the city
Removed.

Pilfer religious relics
Percentage lowered to 50%.

Counter-spy
On establishment (this now takes 5 turns), immediately identify all foreign spies on or below the counterspy's level operating in the city (regardless whether the spy event has ended). Then you get to pick from the following missions:

Capture foreign spies - 20 turns
10 turns later, mark all foreign spies on or below the counterspy's level for execution. They die after their missions are complete. Gain 25 / per executed spy.

Sabotage mission - 20 turns
End all negative spy events performed by foreign spies on or below the counterspy's level immediately. The city is immune to all negative spy events for 10 turns.