Warrior rushing Deity (with pictures!)

No use of MSPaint for entertainment value, though. Sorry. It's clear that aimlessgun has me on sense of humor. No sense contending with a master.

So there seems to be some polite skepticism in various threads about the viability of Warrior rushing on Deity, given that the AI starts out with oodles of units. This thread is intended to explain how you go about solving that problem. I intend to deal with three cases: good, bad and ugly. I have included screenies, as requested. I have now reduced the size down to something like 1680x950, which should work on most monitors. Thanks for the tip on batch jobs in Irfanview; didn't know about that feature.

The good case is a cakewalk. The starting position is good, the AI makes a boneheaded play, I can exploit the AI's inability to deal with rivers, and the result is a flipped city on turn 30 without loss. This is the outcome you want to see irrespective of strategy. If you're going for an early puppet strategy, you can press on to the capital in a few turns and flip it. If you want to annex and Settler pump, then raze and self-replace, that will also be fine.

The bad case is a situation where I can flip the city, but it takes some work. The game has provided an undesirable capital roll, I don't find the target civ until late, and the terrain is a bear. I flip the city without loss, but this results from planning and a little luck.

The ugly case is a situation where there is no way to pull off a Warrior rush. The player is best served to recognize this quickly and audible away to a strategy that will find success.

I'll take the good case on here, and attack the bad and the ugly in succeeding posts. For these games I am playing France, which is probably the most flexible civ. The extra SPs mean that any strategy you can imagine is sustainable, ranging from pure ICS with Meritocracy to pure warmonger puppeting using saved SPs and a Dynamite beeline to open Communism, Police State and then mass annex. (Police State makes all of your annexed cities nearly comparable to India's without building a Courthouse. Thank you, Nunya, for discovering this. I now have an excuse to warmonger again, albeit with a single civ.)

In the good case, I start with the following for a capital. I moved one tile to the NE to settle on Ivory. I'm taking the +2 GPT, since I can work a 2/1/1 right away and leverage France's fast border expansion to get the second tile before growing. The Tundra tiles mean that neighbors are most likely to be found to the North, so the Warrior heads that way.



The Warrior quickly finds Songhai's borders. (Oops, thought it was Siam. They do look the same when you get them at the edge.) Either way, good civ to cripple. I am pleased.



Upon arrival, I discover that the AI is bad at CiV. Notice that it thinks that the Warrior to the east can cover that Worker. I know that it believes this to be true, because I saw a Warrior on the Hills tile east of the city that really could protect that Worker move off to the northeast. The result is a 100% safe steal.

It's also worth noting that no good play goes unpunished. On the way up here, I popped a +1 pop ruin. Had I stayed in place, I'd be collecting two 2/1/1 tiles right now. Alas. It isn't worth spending 50G to get +1F/-2G for four turns, so I live with it until the border pop.



Now notice that the Warrior to the east never moved. Also notice that there is a Hills tile out of range of city shots, but two tiles away from that Warrior. We'll exploit this in a moment. If you're wondering why I chose to eat the city shot rather than an attack from the Warrior (given damage to the Warrior and double XP for my unit), it's because I don't want that unit able to attack once more and burn a promotion on a self-heal. The AI is pretty intelligent about using promotions for self-heals, and I hate it when the AI turns losses into wins.



Now this start begins to get silly. I'd have written it this way, but a publisher would tell me that I should learn to write fiction without relying on ridiculously improbable events:



Now I take advantage of the less-than-intelligent Warrior stationed on a bad Hill. Notice that I retreated the badly damaged unit back to the capital. It will heal (marginally) faster that way. Ordinarily, I would leave this damaged unit in place, but another Warrior shows up on that Hills tile and I have to run.



Please notice what I did with this Worker. I could Mine the Hill to the northeast of the Wheat, but that is the wrong play. The best thing I can do is take the 20H from the Forest now in order to deploy the fifth Warrior earlier. It only saves three turns, but turns this early are gold:



Hello India. Good to see that your Scouts are pro at Ruin pops. Remind me not to mess with you until I have Longswords.



Very little of interest happened for a good long while. At this point, the Warrior in the yellow killed a Scout turned Archer, and will have a saved promotion after surviving the remaining Warrior's attempt to kill it. The Warrior in red took out the Warrior that was on the Hill earlier, and has a saved promotion as well. I will heal both when they are standing just outside the city I wish to flip. Notice that I have spread out to engage the city from all sides, and that no Warrior is in danger of its life.



Now everything is ready to roll. I will simply run from the wounded barb, and capture the city with the Warrior that it endangers:



Same turn after burning two heal promotions and surrounding the target:



And that's all she wrote, leaving a very strong position going forward. It's worth noting that Gao had a Chariot Archer that never bothered me, because it was the garrison unit for that city. I'd be happier if I'd been able to snag the second Worker, but I'm hardly complaining given that I picked up Bronze Working from a ruin...



I've had better starts, but not many.

OK, welcome to the bad case. First game I spun up today. I find the target very late. Although it turns out to be surprisingly close, it has nasty terrain and the AI aggressively deploys units to defend it. Fortunately, the AI is rather boneheaded about its approach, and I manage to keep all of my Warriors. I expected to lose one, but did not.

The start is:



which I am decidedly unenthusiastic about. Being Hammer poor does not make me happy, and there's a giant time sink sitting on my Sugar. On the plus side, there are Cows and Marble. At least my Food tiles don't stink, right? I moved one tile NW to get here. I couldn't see the Cows until moving; those were an added bonus. The way I saw it, no sense in settling the river and being 2 tiles from both luxuries.

Warrior heads out SE and finds some Gold, which ends up being a lifesaver:



Unfortunately, I reach the sea and there's no one down here. I move NE now.

I discover a ruin, which pops a map. This delivers bad news; nothing but city-states to the NE. I carefully skirt the barb encampment, and it leaves me alone. There really isn't any sense in dealing with it until it can harass Workers; killing them will waste time that I could use to find a target. I head north into the fog.



On a side note, I am really beginning to dislike this start. There aren't a lot of Hills, the only available luxuries I've seen are more Sugar tiles, and they are profaned with more Marsh. Both city-states have Ivory, which is also a downer. The Forests will eventually be good, so there is some endgame potential, but right now any cities I found south and east are going to be rotten.

The Warrior I built (ten turns, bah) quickly discovers Suleiman. The good news is that I can execute a fast rush. The bad news is that the approach is rotten. There are a maximum of four possible tiles for attack, and I can't see how hard the ones in back are to reach.

You can see that I had more fun with Barbs up north. This time they attacked me, although now that I know where Suleiman is, I prefer it that way. The ruin pops +30 Culture. That's a good thing if I go Meritocracy; otherwise it's a waste of a ruin.



Suleiman has a serious moment of idiocy here. I also learn that getting to those back tiles is going to be a beast. The best option would be to station a Warrior on that Hill on the ocean, then move two tiles West to reach the final tile. That causes me to eat an extra city shot on the approach, but there's nothing for it.



The question is now what to do. I can steal both Workers and hope for the best. I don't have a promotion. If I did, I could put Shock on the Warrior and have some chance of surviving a city shot in the open and the Warrior's counterattack. Even then, it's a risk. I conclude that this is a bad idea, and take the near Worker. Also, I rush a Warrior. I'll need it anyway, and looting Suleiman's treasury of 19 Gold (woo!) puts me over 200.



Notice the road that I discovered there. This left me with some poor options. I can stay put and risk getting blocked by ZOC from the Warrior to the SE, or I can move one tile East and risk losing the Worker to something that comes roaring down that road. I decide to risk a steal rather than a Warrior death. It looks like the capital is 4 tiles away, so unless Suleiman has a unit sitting on that road in a spot I can't see, I should be fine.

It turns out that Suleiman doesn't even bat an eyelash. Not the response I was expecting.



Now it's time to exploit that Warrior's ability to see things he can't kill. Also notice the glacial pace of Warrior construction in the capital, because the best I can do is work that Forest tile. Maybe I should have gone AH instead of Masonry?



Advancing to the Hill three tiles from Edirne lured out another Warrior to its death. I really don't know what Suleiman was thinking here with that Archer, which came from the Hill two tiles East of Edirne. I have a moment of idiocy and forget to move the full health Warrior from the hill to the Forest next to the Archer for flanking. The Archer survives with 1 HP, and ends up dealing an extra point of damage to my unit next turn. Ironically, this misplay saves me a lot of grief later.



OK, so now I'm feeling really stupid about not going for AH. But I feel a lot better about the start now. I've got three high quality production cities ready to go here (I'm thinking 3W, 1 NW of Paris and 1 W, 2 SW from Paris) and I know where the Maritime is. If I can find somebody else to offload this Marble on to, I can get a boatload of Settlers out in a hurry. Happiness is going to be a huge issue, but I won't need Iron, so at least I can grab Calendar quickly. I'm starting to lean Meritocracy, but I'm deferring SP spending. I've attached a save file, and I assume you'd prefer that I not spend SPs for you.



Spearmen show up in the capital, and a Scout shows up one tile SE of Edirne. I kill the Scout and promote Shock in the hopes of surviving the Spearmen counterattack, then running for it. I also move the slightly damaged Warrior onto that Hill to reach the Forest tile behind the city. Suleiman's response surprises me:



Fortunately, I have a heal promotion to burn on the Warrior on the Hill. Suleiman offers me 45% Flanking in open terrain on the Warrior; what gives there? Somehow, it survives the first attack, which makes a bit of a hash of my attack plan. I had been intending to scoot that Warrior two tiles East of the city around the back, but now I can't do that. I stare at this position for a good long while:



My response may surprise you:



Attacking with either yellow Warrior this turn means certain death. One or the other will eat a city shot. I will retreat that one to safety, and attack with the other next turn. The city should fall, so I can scoot the very very vulnerable Warrior in back to safety:



All in all, a much better outcome than I expected. I figured that I would lose at least one Warrior, possibly two if unlucky. Instead, the AI gave me a couple of serious gifts, ran out of units (I killed its entire starting complement of 4x War and 2x Scout), and pretty much gave the city away. The two Horse pops mean that I'm looking at some production juggernauts and fast Universities, and I should be able to block Suleiman long enough to deploy a Settler to snag that tasty looking site to the NW of Edirne. The really wretched looking area to the SE doesn't look so bad now that I know there's a Plains Horses tile down there. This looks like a quality game going forward.

I'll probably post another ugly example, but this one isn't bad. The start looks like this:



Not great, not awful. There should be another luxury lurking around somewhere. I could settle in place, but moving the Warrior east reveals Plains Wheat next to Silver. That's good enough to waste a turn on.



Silver tends to spawn near the poles, so I guess South. This is pretty thin data, though. I want this city site that I discover:



Just wow. After grabbing the ruin, I discover a wall of Mountains to the South. So much for that idea. It looks like the only possibility is northwest:



The continued existence of this ruin on turn 9 is a very bad sign:



And at this point it's pretty clear I'm screwed. I have no idea where Washington is, there is a ton of rough terrain between me and him, my fifth Warrior will finish on turn 23 or so even if I rush a Warrior, and it will take 10+ turns to get into position. If I want to continue the Warrior rush line, my best option is probably to utilize aimlessgun's idea and hit the city state (which turns out to be Militaristic anyway). But I would be better served to realize that this is futile, start making a Settler and send it to the hot site to the south, take out a mortgage with Washington for Gold, rush a Worker, and steal a Worker from the city-state.




Impregnable fortress is impregnable:



That isn't as bad as the times I've had when a civ's capital was only accessible through a single land tile, but you're not taking that with Warriors. And you'd better take it before Janissaries appear.

Martin Alvito said:

I have included screenies, as requested. I apologize for the odd resolution (2048x1152) that is probably too wide for your monitor. I blame Samsung for making an excellent monitor (2343BWX) with a goofy resolution at a bargain basement price ($200 new last year).

Click to expand...

You do know that it's possible to resize images
Irfanview can even do it as a batch job, really convenient

It's hard to follow even on my 1920x1200 screen

Great post, but you need to sort the pictures out, imageshack has a resize option, 17" is usuall good for the forums.

A little work around; on firefox use ctrl + / - to zoom in / out.

I always confuse the borders of Siam and Songhai, too. How often do you tend to get a "good" start? For me, it's mostly bad or ugly

That was a bear to read even on dual monitors.

Excellent writeup, thank you for noting all the minor tricks.
Nonetheless this confirms what I've pretty much concluded from my own games.

You finished up on turn 30 with a perfect setup.

Any of the following basically results in the loss of a warrior and would have slowed you down.
1. Barbarian Camp being closer and two barbs showing up instead of one.
2. Second City being closer(2tile road)
3. Not finding someone until turn 10
4. Siam getting that pro ruin pop instead of gandhi
5. Crap terrain
6. Being unable to steal a worker (bad luck with terrain again in a sense)
7. Target being Alex or Monty

Now that I've got the idea down I embrace this tactic for the start, and so long as only one of the above happens, then you can bull through anyway for the loss of a warrior which is worth it. The moment two or more of the above are true however, I scrap the thing wholesale, and immediately switch the warrior building to a settler.

Basically odds of success go something like:
0=100%
1=75%
2=50%
3=25%
4 or more=why the hell are you even trying?

Alex and Monty aren't that bad. Alex's warriors are the same as anyone else's, and the Jaguars are usually only dangerous if they win. Jungles are too rare to matter much. Unless they start with more stuff in Deity...?

I'm surprised to see they just had 2 warriors. Usually it seems like they have at least 4 warriors to defend with.

Nice write-up! But, couldn't you have grabbed the 2nd worker on turn 29 and your warrior would still be in a position to attack the city? Is there a reason to be next to the city rather than 2 squares away?

Gus_Smedstad said:

Alex and Monty aren't that bad. Alex's warriors are the same as anyone else's, and the Jaguars are usually only dangerous if they win. Jungles are too rare to matter much. Unless they start with more stuff in Deity...?

Click to expand...

They're both dangerous for a whole different reason.

Monty's jaguars aren't worse than normal, however his build order is Jaguar->Jaguar->Jaguar->Jaguar->Did I mention Jaguar?->Jaguar->Jaguar->Some more Jaguar

And Alex will often start popping hoplites at turn 20...

pi-r8 said:

I'm surprised to see they just had 2 warriors. Usually it seems like they have at least 4 warriors to defend with.

Click to expand...

With at least 2 archers as well, from my experience.

Depends on how soon you find them, but on standard map size it's never been anywhere near as early as turn 3 - I might come across a "foreign" unit by turn 5 if I'm lucky, cities (outside CS) never before 10, and often a lot longer. Are you using some special settings here MA?

Nice writeup (though I think several people have already given some suggestions on the size front I use Paint.net but that's because I'm also editing them)

I am also wondering what the actual unitcount for Askia was/is. Are we seeing few units because they were defending his capital (other than the chariot archer). That's what has happened to me. Or did you kill more units than I thought in the field before the assault.

EDIT: I'd like to mention an audible you can considering switching to: taking down a city state with the warrior rush.

If you're going the total conquest route, taking a city state frankly isn't going to hurt you much down the road. The city defenses are a little tougher than the 2nd city of another civ, but if the CS is surroundable it is easily capturable. You aren't crippling another civ, but it's still a 2nd city very early (I'd steal workers from a nearby AI regardless though, that should be possible almost every game). Something to consider if it's turn 25, you've built nothing but warriors, but you're looking at your neighbor and thinking "ugh this isn't going to work at all".

KingMorgan said:

A little work around; on firefox use ctrl + / - to zoom in / out.

Click to expand...

Works on Chrome too.

aimlessgun said:

Nice writeup.

I am also wondering what the actual unitcount for Askia was/is. Are we seeing few units because they were defending his capital (other than the chariot archer). That's what has happened to me. Or did you kill more units than I thought in the field before the assault.

EDIT: I'd like to mention an audible you can considering switching to: taking down a city state with the warrior rush.

If you're going the total conquest route, taking a city state frankly isn't going to hurt you much down the road. The city defenses are a little tougher than the 2nd city of another civ, but if the CS is surroundable it is easily capturable. You aren't crippling another civ, but it's still a 2nd city very early (I'd steal workers from a nearby AI regardless though, that should be possible almost every game).

Click to expand...

That's why I put distance to capital as one of the more important things to the rush working or not from my experience.

With a bit of further thought, it is THE most important thing. I've found that the AI doesn't really redirect, and it seems to defend on a per-city basis. So if you're far enough away from the capital, those units won't show up to the fight.

On a side note, this is also critically important when defending. I discovered that when I put a forward defense up and actively hunted archers in front of my cities, I drew half their f-ing army in and lost cities. But drawing back and only attacking when absolutely critical, immediately withdrawing back made it waaayyyy easier.

Basically Civ V AI is like something from Diablo. It has aggro range (visible range 2-3 tiles usually) and so long as you avoid aggro, it won't attack.

I'll resize the pictures for you all this evening when I have time. Some quick responses:

No special settings (Standard, forget whether Pangaea/Continents), he just happened to be that close. I'd say ten to twelve tiles is a lot more common than 6-8. Remember, that's a "good" start where things break exactly as I'd like to see them. Great initial city, quick theft, quick Warrior kills, an open city and the Tundra tiles gave me a strong signal about where the nearest neighbor probably was.

Map settings matter. A Small map tends to be very compressed, and ideal for Warrior rushes. Tiny and Large tend to have your adversaries a long way away. Continents seems to result in closer adversaries than Pangaea.

I killed three Warriors and an Archer that probably started as a Scout. A second Warrior came out and attacked me on that same Hill as the first, and I forgot to take a picture of that kill. Killing the Archer while crossing the river is how the unit in the Jungle got to yellow.

There was a Chariot Archer in Gao that never joined the fight. It could have bagged one of my Warriors easily and safely, but the game seems to insist on having a garrison unit these days.

Monty is a pain, but the good news is he's aggressive. If you declare early, he will eventually clean out his cities and send a Jaguar force your way. You can kill the army, then counterattack. Alex is a lot more dangerous if he pops a couple of Hoplites from ruins.

dkrussian said:

That's why I put distance to capital as one of the more important things to the rush working or not from my experience.

With a bit of further thought, it is THE most important thing. I've found that the AI doesn't really redirect, and it seems to defend on a per-city basis. So if you're far enough away from the capital, those units won't show up to the fight.

On a side note, this is also critically important when defending. I discovered that when I put a forward defense up and actively hunted archers in front of my cities, I drew half their f-ing army in and lost cities. But drawing back and only attacking when absolutely critical, immediately withdrawing back made it waaayyyy easier.

Basically Civ V AI is like something from Diablo. It has aggro range (visible range 2-3 tiles usually) and so long as you avoid aggro, it won't attack.

Click to expand...

Hmm. How often is the 2nd city 2 tiles from the cap? It looks like MA's example works with 3 tile separation, though it is 3 tiles of rough terrain (would that matter to the AI?). I know for sure it works with 4 tiles distance.

But yeah I think the aggro analogy is pretty accurate for the AI on defense, because on defense it doesn't really have a focused plan like "take X city" or whatever.

playshogi said:

Nice write-up! But, couldn't you have grabbed the 2nd worker on turn 29 and your warrior would still be in a position to attack the city? Is there a reason to be next to the city rather than 2 squares away?

Click to expand...

I'd have eaten a city shot from Gao and a shot from that Chariot Archer. The damage would have meant that the city wouldn't have fallen on turn 30, and I'd have lost at least the Warrior SE of the city to the barb. I'd probably also lose the Warrior that stole the Worker to the turn 30 city shot.