Worst/Easiest AI leaders in the game

Among the AI leaders, who would you consider to be the "worst" at playing the game optimally?

Behaviors like picking Honor and Exploration, slow expanders, ignores city states etc.

kickthecommie said:

Among the AI leaders, who would you consider to be the "worst" at playing the game optimally?

Behaviors like picking Honor and Exploration, slow expanders, ignores city states etc.

Click to expand...

EDIT: Just use the link Balkoth posted.

- A high expansion/low growth flavour means the AI will be more inclined to expand.
- low expansion/high growth (i.e. gandhi) means the AI will be less inclined to expand
- a high bias towards ignoring minor civs will...... make the AI ignore city states.

As for picking Honor first, leaders who have high offense and military_training flavors will be more likely to go for honor.

You could also use this chart: http://civdata.com/

Many leaders have unique units and attributes that make you be good with different starting social policies. For example, warmongering civilizations can let you build up to social policies that work well with warmongering social policies such as honor and other later era social policies that are also for warmongering. The same thing goes for cultural civilizations that let you get more social policies that let you get more social policies from the civilization's social policies and your social policy's social policies. There are also civilizations that are good with religion that make you match up well with social policies that are good with religion and faith such as piety.

Songhai is one of the worst AIs out there. I haven't seen it excel at anything good. It's even worse in early game because they just build a ton of warriors and go honor.

Polynesia is also pretty bad. They don't know what to do on Pangea type maps.

Babylon as well believe it or not. They bulb their first scientist at Writing, and no sometimes they dont actually go Writing at the start. They go for archer tech, then usually Masonry for its walls. Then usually goes for Writing where it bulbs its first scientist (not plant it!)

But Venice is hands down the worst AI in the game. Mainly because it goes thousands of miles across the map with its Merchant of Venice to buy out/trade with the City state, which it usually gets eaten by barbs on the way there, or Venice will get so confused that it will constantly move its Merchant of Venice in its territory and keep it in OR PLANT IT IN ITS OWN TERRITORY! The trade with Venice is good, but AI Venice is even weaker because it can't figure out what to do with its Merchant of Venices.

I agree with Songhai and Venice, and want to add Sweden as one of the worst AI's. If Sweden is in my game it's probably the first to get beaten to crap by other AI's.

Babylon is disappointing as an AI indeed. They are ok in most of the games but they never manage to accelerate at science.

For some reason Polynesia is always one of the better AI's. I guess fractal maps give them just enough coast to make it work.

I'm not sure about Sweden. I know they have a spot taken up on their unique unit. It really isn't a unique unit, well not without a general and even then it is bad..... and its a Lancer. Hakkapettilas if I recall their name.

Careolean is pretty nice, and I've seen Sweden use these for rushes, having marched Riflemen is pretty good.

Now the confusing thing about Sweden is how it wants to war often, but their unique ability wants them to be peaceful, stack up those bonuses and making friends. This is probably the weakest point of Sweden. In CivData.com they're ranked a 7 for war (which is one of the highest). Sweden would be better if they we're more Ghandi like (without the nuke joke). This will allow Sweden a better chance in some games they play.

They also can gift unique units, in my experience Sweden doesn't often gift these Great People. When I mean often I mean in 1/3 of the games they're in. I have seen them give a City State a Great Scientist and Great Engineer one time, lol. If Sweden could use this ability better, it would make them great.

But over-all. They have a terrible Lancer replacement unit, a situational/good Rifleman replacement and very weird AI flavors.

Enrico Dandalo (Venice)

-He is excellent candidate for forward settling because he does not have the ability to get a settler out quickly to counter this early move in the game.

-Take out his first Merchant of Venice, and he is pretty much done. Many times he doesn't support this UU, and it is a easy capture. If he does somehow get away..Then ally the CS when he arrives, and backdoor him.....Goodbye!

-Normally he is playing the top of the Tech Tree to get to "Optics", which means no early Archers, Spearmen, Catapults, or Swordsman. (Unless he hits a Tech. Ruin,..Of Course)

-Most of his early production goes into Work boats, and Cargo Ships instead of Military. Pillage, Pillage, Pillage,...Plunder, Plunder,....Now what does he have left?...Some Gold,...I would imagine, but usually he can't counter if you get him early enough, and were concentrating on building up your military.



-

ThorHammerz said:

EDIT: Just use the link Balkoth posted.
- A high expansion/low growth flavour means the AI will be more inclined to expand.
- low expansion/high growth (i.e. gandhi) means the AI will be less inclined to expand
- a high bias towards ignoring minor civs will...... make the AI ignore city states.
As for picking Honor first, leaders who have high offense and military_training flavors will be more likely to go for honor.

Click to expand...

After I got tired of the AI spamming trash cities, I started using civdata.com to pick opponents with low expansion flavors. That worked pretty well. I am really struggling with Deity, so I am all for using the civdata.com to stack the deck in my favor!

Is there a way to tell from flavors which other trees the AI will favor or avoid? And how reliable is this? Will leaders with high offense and military training flavors always open honor or is RNG a pretty big factor? It seems pretty random to me, but it would be nice to have a shot at Hanging Gardens...

Sharples said:

Songhai is one of the worst AIs out there.

Click to expand...

With the AI's tactic of throwing units in the water, I actually feel like Songhai does okay. The only AI that I have noticed consistently underperforming and is never the run-away is Byzantium. She only founds in maybe 1 game in 4, and does not exploit the Dromon, so basically she is an AI with no UA, UB, UU. I agree that AI Venice also has very mediocre behavior (and also never a run-away), but the MoV will at least is used if the player doesn't stalk it. I hate seeing Venice in my games, as in addition to the annoying but underpowered UA, it means the other civs have more space...

Really, are any other civs as terrible for the AI as Venice and Byzantium?

Mongolia is pretty bad. I often see ghenghis not settle any cities while he tries to attack CS's. If someone like Alex is around spamming units and gifting them i've watched ghenghis fail to take it and sit on his lone city until everyone else hits industrial

beetle said:

Is there a way to tell from flavors which other trees the AI will favor or avoid? And how reliable is this? Will leaders with high offense and military training flavors always open honor or is RNG a pretty big factor? It seems pretty random to me, but it would be nice to have a shot at Hanging Gardens...

Click to expand...

EDIT (disregard what I wrote previously):

You can sort of guess which trees the AI will take based on the flavour distributions in the table Policy_Flavors, and compare that with Leader_Flavors. The PolicyAI stuff makes my eyes hurt, so I'll leave the explanation at that.

It is sometimes hard to guess however, as policy flavours will propagate (meaning, later policies will have their flavours "trickle down" into their pre-reqs, this is the same with technologies) down the policy tree, so the AI will also consider a bit of what the later ones unlock.

Grand strategy also has an effect later on for choosing/changing ideologies.

Ive seen a couple of civilizations lose their capitals to casimir. Theodora was one of them and i dont really remember who else couldve lost their capital. It couldve been kamehameha, Ive seen that guy do good and do bad in different plays.

That is because Poland is overpowered and can easily take on half of the Civs in the game. Especially when they run off in the game.

(I wonder why the devs thought Poland was balanced ).

Poland really is OP, i was trying to tell that to many people before and ive even made a poll about that and people still didnt think casimir was OP. Other players have tried saying that casimir was overpowered too and also failed.

kickthecommie said:

Among the AI leaders, who would you consider to be the "worst" at playing the game optimally?

Behaviors like picking Honor and Exploration, slow expanders, ignores city states etc.

Click to expand...

Is picking Honor really considered that bad? I like to adopt it early just to help with the barbs, and get culture from them. I feel that helps a lot, especially if you went Liberty.

If the AIs miss the early opporunities by going Honor, they are dead. They won't recover in the long game compared to Tradition/Liberty Civs. Honor is strictly a policy tree for rushes, and the AI will have almost no benefit to growth or production from choosing Honor. If an Civ that choose honor takes out another Civ, they are fine. I've seen AIs go honor and they really do fly away in score. But it really depends on if they can get more of a benefit from choosing honor when they capture cities. Because no having early growth and production will really hamper a Civilizations ability to catch up. This is why Songhai is so bad. It basically gets no bonuses if it fails to do anything with Honor.

Poland really needs a nerf. I mean it makes no sense to give Poland an incredibly strong lancer while Sweden gets just another Lancer (or basically one if it doesn't have a GG and even then its bad!). And I know for a matter of fact that Poland is miles better than Sweden. For a game that tries to balance Civs, this is an obvious one that needs doing.

I think the most decisive factor that determines AI success is what opening policies they take. Songhai is probably weaker than most because it favors religion and domination so it is always going to start with Piety or Honor.
Is say the best AI is Zulus providing they invest in Liberty, but they usually run out of steam once the impi rush is over. I'm not sure about Poland I've never seen them do well.

But Husaria/Winged Hussars were in reality that good! Just read up on them, they are awesome in the game and usually were so in real history. They saved Vienna from Turkish conquest, under king Sobieski, something Austria doesn't want to admit.
They were deployed to crush defending forces in field battles, they did quite often put them on the run, the representation of the Winged Hussar is very accurate.
Sweden was no match for Poland when it came to Cavalry, infantry and artillery vice versa.
I really enjoy playing as Poland, not for the social policy bonus, but for the battlefield excellence of their unique cavalry unit.
Thumbs up to Firaxis for their research, wish they'd do the same about Celtic and Iroquois city/village names in the game, among many other things.

ps. I was responding to posts from Sharples and reddishrecue

ps. We can talk about Poland being overpowered in the social policy department. Although I think recognizing the Polish Solidarity Movement was another excellent move by the design team, only it's effects are a bit too strong.

In a historic sense its hard to balance Civilizations, unfortunely you need to sacfirice some historical records in order to make sure there is balance in the game. This is not Europa Universalis.

The winged hussar really is neat because the unit withdraws the defending unit back a tile space. Like berber cavalry, winged hussars are good at defending its borders and pushing units back into where they came from. Upgraded into helicopters is another story though because i havent tried it out, but i guess that its doing the same thing as a helicopter if the promotion stays with the unit. However, it can still remain a mystery because some promotions get lost when the unit gets upgraded such as african forest elephant scare promotion that gets lost when upgraded into a knight.