Late game snowballing/power disparities

[Blue Ghost]

I’ve noticed a consistent trend in the games I’ve played, that from about the Modern Era onward, power disparities between civs get absurdly large. In the early and mid game, some civs may fall behind due to bad luck, but for the most part, things are relatively even; there’s rarely more than a 2-policy difference between first and second place. But later on, the snowball effect becomes so strong that the lead of the top civs becomes insurmountable. If I have a slight lead at the Industrial Era, it’s common for me to have my tier 3 tenet before the next civ even gets their ideology, or have a 20-tech lead at the end of the game. If a civ falls slightly behind, they seem to lag and be unable to catch up. I don’t know what causes this effect—I suspect it has to do with the strength of late game buildings and policies, and faith purchasing great people—but I just wanted to throw this out there and see if others have a similar experience.

 

[CrazyG]

Maybe its time to increase your difficulty?

I used to have this happen a lot before, but much less often now.

 

[Arbi]

I second that experience. For me it's not mainly between the AI and the player, but even more so among the AIs.. I usually compete with 1 or 2 AIs (on large maps), and then mostly concerning social policies (while techs are a little bit closer together overall). It's not unusual for the first 2-3 Civs (AI and Player) to be MANY policies ahead, e.g. deep (6+) into ideologies, while only 3/10 Civs in the game even have an ideology.

It does not make sense to me to ascribe that to the difficulty if the disparity among the AIs is so pronounced as well.

 

[Omen of Peace]

There are quite a few rubber-banding mechanisms in the game (tech costs, spies, trade routes).
But policies and tech are inherently "win-more" (the more you have, the better the yields/military power/...). And AI have an inherently win-more mechanic in the form of their instant boosts: they get them e.g. when founding a city or building a wonder.

So snowballing does make sense, and I'd argue that it's a good thing as long as it's not too much, too early. Like CrazyG, I'd say things are better in that respect than they used to be.

 

[MarvelOfRain]

I’ve noticed a consistent trend in the games I’ve played, that from about the Modern Era onward, power disparities between civs get absurdly large. In the early and mid game, some civs may fall behind due to bad luck, but for the most part, things are relatively even; there’s rarely more than a 2-policy difference between first and second place. But later on, the snowball effect becomes so strong that the lead of the top civs becomes insurmountable. If I have a slight lead at the Industrial Era, it’s common for me to have my tier 3 tenet before the next civ even gets their ideology, or have a 20-tech lead at the end of the game. If a civ falls slightly behind, they seem to lag and be unable to catch up. I don’t know what causes this effect—I suspect it has to do with the strength of late game buildings and policies, and faith purchasing great people—but I just wanted to throw this out there and see if others have a similar experience.

Wars, wonders and using the fact that you have a good start to your advantage (somebody that spawns on Tundra will just fall further and further behind). Also, different civs go for different strategies and some such as Denmark, Rome etc. might not do very well if they do not manage to win a war and if they do they start running ahead. It just part of the game IMO. There is the question of some civs often being really really far behind. But at least in my games, it is not that bad. For example, in my last game, Genghis was quite behind and managed to at least catch up to everyone even without major conquests.

 

[SuperNoobCamper]

Maybe its time to increase your difficulty?

I used to have this happen a lot before, but much less often now.

That's what i say to my self before starting another immortal game that i'm sure it will end in a rage quit after t150.

 

[Randomized]

Yeah as others have said that's just the nature of the game. And it's fine like this imo.

 

[Blue Ghost]

There are quite a few rubber-banding mechanisms in the game (tech costs, spies, trade routes).
But policies and tech are inherently "win-more" (the more you have, the better the yields/military power/...). And AI have an inherently win-more mechanic in the form of their instant boosts: they get them e.g. when founding a city or building a wonder.

So snowballing does make sense, and I'd argue that it's a good thing as long as it's not too much, too early. Like CrazyG, I'd say things are better in that respect than they used to be.

I would argue that the insane disparities I’ve been seeing are too much, too early. Modern era is late game, but there are still three eras left, and this is the only time ideologies come into play. No one should be out of the running at this point unless they did extremely poorly.

Wars, wonders and using the fact that you have a good start to your advantage (somebody that spawns on Tundra will just fall further and further behind). Also, different civs go for different strategies and some such as Denmark, Rome etc. might not do very well if they do not manage to win a war and if they do they start running ahead. It just part of the game IMO. There is the question of some civs often being really really far behind. But at least in my games, it is not that bad. For example, in my last game, Genghis was quite behind and managed to at least catch up to everyone even without major conquests.

At what point did he catch up? In my experience, the early and mid game are well balanced in the ability for players and AI to make up for major deficits. But I’ve never seen an AI that was behind come anywhere close to closing the distance after the Industrial era.

 

[SuperNoobCamper]

In my current Emperor game as Assyria the closest AI to me, Songhai with 2 vassals was still in the industrial era without ideology while i was at the last row in the tech tree fighting his lancers and GGs with XCOMs and Modern armor.
I have seen this enough times to get the feeling that this is not snowballing as much as it is the very nature of late game turns and things just happen too fast, like how expanding a single scientist generated over 30K science giving me what's effectively 4 techs, a great writer is equal to a social policy and a great artist is a couple of golden ages which is something the human player is much better at planning and executing.

 

[AndreyK]

May be make that GP instant yields give 8 (or whatever) turns of appropriate yields not counting other instant yields, so not giving those crazy amounts?

 

[CrazyG]

I think I hit over 60,000 science in my last game as Babylon.

 

[JamesNinelives]

There are quite a few rubber-banding mechanisms in the game (tech costs, spies, trade routes).
But policies and tech are inherently "win-more" (the more you have, the better the yields/military power/...). And AI have an inherently win-more mechanic in the form of their instant boosts: they get them e.g. when founding a city or building a wonder.

So snowballing does make sense, and I'd argue that it's a good thing as long as it's not too much, too early. Like CrazyG, I'd say things are better in that respect than they used to be.

The primary balances for technology are (at least in theory) happiness and military capacity. I agree with policies though - for me the other significant no-penalty factor is production (it's pretty much always a good thing to increase when possible).

I've also found runaways less of an issue than they used to be. I can't speak for the most recent patch though, don't have enough experience with it.

 

[andersw]

I've noticed quite a bit difference between civs I play and options.
The ones I'm able to snowball most with are probably Japan, Egypt, Austria and France.
With them I can scale with artistry or statecraft which then (for me) easily outperforms fealty in my wide warmonger games.
And even allows me to pop over to industry (right path first) without tanking happiness too bad.
Looking at AI on emperor/immortal it's usually statecraft AIs that are the most snowbally ones.

 

[JamesNinelives]

I've noticed quite a bit difference between civs I play and options.
The ones I'm able to snowball most with are probably Japan, Egypt, Austria and France.
With them I can scale with artistry or statecraft which then (for me) easily outperforms fealty in my wide warmonger games.
And even allows me to pop over to industry (right path first) without tanking happiness too bad.
Looking at AI on emperor/immortal it's usually statecraft AIs that are the most snowbally ones.

Japan and France certainly have mechanisms that can lead to snowballing. I don't know if this is something we want to change, personally I haven't found them too strong in my games (king/emperor, huge map, epic speed).

Austria I think is specifically a civ whose strengths all turn up in the late game - they can be very strong if they have a good start, but can also get crushed if they have a poor one. Again, I don't know if that is something we want to change. It can be very annoying when they control every other city state. Maybe moving some of their benefits to earlier on would make then more balanced? It seems some civs are intentionally designed to have one major play though (e.g. the Songhai and early era warfare). Doesn't appeal to me but I have to assume someone enjoys playing them?

Egypt I have seen go the classic snowball. If they grab key wonders early they can just keep smashing them out. I haven't seen it happen a lot though, at other times they can become victims of neighbouring warmongers.

The worst snowball I've seen was China, although that was some patches ago. It might be useful to check what the leader balance thread have to say on the subject, see if any of them stand out.

 

[andersw]

Japan and France certainly have mechanisms that can lead to snowballing. I don't know if this is something we want to change, personally I haven't found them too strong in my games (king/emperor, huge map, epic speed).

Austria I think is specifically a civ whose strengths all turn up in the late game - they can be very strong if they have a good start, but can also get crushed if they have a poor one. Again, I don't know if that is something we want to change. It can be very annoying when they control every other city state. Maybe moving some of their benefits to earlier on would make then more balanced? It seems some civs are intentionally designed to have one major play though (e.g. the Songhai and early era warfare). Doesn't appeal to me but I have to assume someone enjoys playing them?

Egypt I have seen go the classic snowball. If they grab key wonders early they can just keep smashing them out. I haven't seen it happen a lot though, at other times they can become victims of neighbouring warmongers.

The worst snowball I've seen was China, although that was some patches ago. It might be useful to check what the leader balance thread have to say on the subject, see if any of them stand out.

I assume some tall AI civs snowball harder late, especially in high difficulties, china hit such a heavy nerf that it went from one of the strongest to rarely relevant in my games.
There is ofc a big difference on maps also, its a lot easier for a tall tradition civ on island maps and if you fiddle with game speed, number of city states and size that also affects some civs more than others.

 

[MarvelOfRain]

I would argue that the insane disparities I’ve been seeing are too much, too early. Modern era is late game, but there are still three eras left, and this is the only time ideologies come into play. No one should be out of the running at this point unless they did extremely poorly.

At what point did he catch up? In my experience, the early and mid game are well balanced in the ability for players and AI to make up for major deficits. But I’ve never seen an AI that was behind come anywhere close to closing the distance after the Industrial era.

Okay then probably still in the mid-game (start of Industrial era).

I would strongly disagree though that civs should be able to catch-up in Modern Era. Most games I play are decided at that point or at least have the clear frontrunners. Nerfing the instant yields or late-game buildings will only slow down the process of actually getting to a win.

 

[kenneth1221]

I would strongly disagree though that civs should be able to catch-up in Modern Era. Most games I play are decided at that point or at least have the clear frontrunners. Nerfing the instant yields or late-game buildings will only slow down the process of actually getting to a win.

My two cents on this point specifically: Modern era is what, 1920 in real-world times? And I know a lot of people here don't care so much for mapping the game to real-world events as compelling gameplay, but historically a good number of countries have shifted in and out of superpower status since the modern era. Maybe the late game would be more interesting and there would feel like there was a reason to keep playing after assured victory if there was actually the chance of a shift in the status quo, like the late game crises in Stellaris, but that doesn't really translate to Civ.

This was probably less of an issue in BNW because parts of the game were slightly more decoupled from each other, whereas in VP culture and production are king.

 

[MarvelOfRain]

My two cents on this point specifically: Modern era is what, 1920 in real-world times? And I know a lot of people here don't care so much for mapping the game to real-world events as compelling gameplay, but historically a good number of countries have shifted in and out of superpower status since the modern era. Maybe the late game would be more interesting and there would feel like there was a reason to keep playing after assured victory if there was actually the chance of a shift in the status quo, like the late game crises in Stellaris, but that doesn't really translate to Civ.

This was probably less of an issue in BNW because parts of the game were slightly more decoupled from each other, whereas in VP culture and production are king.

I once heard someone in some let's play discussing this exact point. It, however, falls apart once you realize that there are countries like Germany or the USA that are global superpowers but only exist for a couple of hundred years. To emulate that you would need a way to split your empire or for city-states to form their own new civs.

Also, I don't know about everyone else, but I personally don't play civ for the late-game and therefore quite like the extra snowball aspects that Vox Populi introduced. You still get some games where the endgame is exciting (I had one racing over the entire continent to get to the Maya that suddenly jumped ahead in Science and started going for space), but at the same time, you don't need to sit through dozens of pointless turns.

 

[Stalker0]

Maybe the late game would be more interesting and there would feel like there was a reason to keep playing after assured victory if there was actually the chance of a shift in the status quo, like the late game crises in Stellaris, but that doesn't really translate to Civ.
.

The most interesting idea I've seen on this was that once Ideologies are picked, the X civs with lowest score actually become vassals of the greater powers. Effectively it mirrors the superpower concept, and at the same time gets the "riff raff" out of the way so you have the runaways go at it for the final part of the game.

 

[JamesNinelives]

I think that it's worth mentioning that while there are games with runaways, I've played a good number of games in which competition was still pretty fierce in the late game with no clear unbeatable odds. Certainly there are smaller nations who aren't going to make it, but at least there were multiple contenders for the top spot. That seems ideal to me.