Is it impossible to defend islands in multiplayer games?

[PhilPeter]

I signed up to this forum because I need your help. I keep losing against a friend, and I think it’s because there is a strategic imbalance in CIV4.

Here’s the problem: In many situations, a defensive strategy is way inferior to an offensive strategy. Assume that two human players are on two different islands. Both players are roughly equally advanced in all respects (technology, number of cities, infrastructure, etc.). Now, each player can choose between either

a) pursuing an attacking strategy: build ships and attacking units to attack the other player on the other island.
b) pursuing a defensive strategy: build defensive units and do not attack.

I like playing the defensive strategy and focusing on infrastructure and economy. But I noticed that playing a defensive strategy against an aggressive human player who is at least equally advanced is never a winning strategy. Here is why: To launch a successful attack that causes significant damage to your opponent (destroy a major city), you need much fewer units than you need to preempt such an attack. For instance, if the attacker attacks with 15 units, you’d need 6-8 units in each of your cities to defeat the attacker, because you do not know where he’ll strike. If you have more than just a handful of cities (say, 10 or 15) that could potentially be attacked, you need a lot more than 15 units to successfully defend your cities. The defensive strategy loses out.

One way to overcome this disadvantage would be to anticipate and preempt the naval attack. I have tried two ways of doing this, but both have failed:

- Espionage. If you know where your opponent’s units board the ships, you can sink them with a fleet before they arrive at your island. The problem with this strategy is that the intelligence provided by espionage is imperfect. Even if you can see everything that happens within the radius of your opponent’s cities, he could still board his troops somewhere else (e.g. outside his own territory) without you noticing. Also, you might miss the enemy’s ships as they sail towards your island.

- Sea blockade. If you position naval units all around your opponent’s island, you’ll notice when and where his naval attack takes place. The problem with this strategy is that you’ll need loads of ships. You need many ships only to encircle his island, and you need even more ships for several strategically positioned fleets that can sink the opponent’s ships as their pierce your blockade. This strategy is hardly any better than building plenty of defensive land units in your cities.

Another option might be to build defensive city improvements (walls, castles, etc.). But building city improvements in each of your cities is just like building plenty of units: It’s very expensive. Defending your cities becomes easier when you have railroads (because you can quickly move your troops to where they are needed) and when you have very few cities. Also, defending against an opponent who does not attack from the sea is easier, because land units do not move as quickly as ships. You have more time to move your defensive units to where they are needed. But in many other situations that match the above description, the defensive strategy seems entirely hopeless, or at least extremely inefficient, compared to the aggressive one.

How do you guys deal with this problem? Is there any efficient way of preempting or defending against an attack from the sea that I haven’t thought of?

 

[Fippy]

Hiya, it's mostly a sp forum part, but i have some mp experience.

For islands / naval attacks we have to approach this by tech era.
On sea we first have Frigates (or Ship of Line, depending on tech path), then Destroyers and finally Battleships.
On island maps, defending will never work without a strong navy

Walls or Castles are useless, those ships can bombard defenses quickly.
Losing one or more coastal cities can happen, but your navy must be able to attack main stacks.
It's mostly about cutting off reinforcements, which can only be done if transport ships lose protection.
Without new units arriving, the attacker cannot keep your cities.

Scouting should most likely be done with spread out ships,
before air units come into play. Airships can cover large areas every turn.
Scouting only helps thou if you already worked on your navy, and should be done around your waters.
I see little use in sending ships over to your friend's side, if they are not an invasion force.

On land i would focus on an army that can easily take cities back (if your friend keeps them),
should be nps with siege and no cultural defense.
Ofc some basic defenders are needed, but you are right imo..too many would be needed for safety.

It's an interesting concept, defenders no longer have an advantage against sea attacks.
So your only chance will be building an attack force, both on land and sea (with navy being far more important).
Not for attacking his island at all cost, but for defending with offensive units.

 

[PhilPeter]

Thanks for your response, Fippy!
The problem is that the attacker likes to destroy (rather than conquer) cities. So it's crucial to prevent attacks. Of course, having a strong navy helps. But my experience has been that it is very difficult to intercept a naval attack. The attacking ships move very quickly, and you often do not know where they are going to attack. And you need several stacks of naval units at different strategic places to intercept an attack. Successful defense is so much more difficult and expensive than successful attack. That's why it seems to be always better to build an attacking force/navy than to invest in defense. Or am I wrong?

 

[rah]

Normally the ability to retake a city will neutralize this type of strategy but if he's going to sack the city there is less you can do. But keep in mind that a duel is a contest of attrition. An invasion force is more expensive that a defensive force so if you destroy the invasion force you can end up in the positive. Have a fleet ready to destroy his navy and any land forces will eventually be doomed.

Build all your major wonder inland and keep your seaside cities more basic so you risk less. Eventually he will figure out the math.

You can also keep a less expensive task force near him. If all of his resources are being used to take your island, they're not at home defending his.
A little paranoia can go a long way.

 

[lymond]

Yep, mainly SP stuff here, but I did play a little MP years ago and understand that there are some striking differences in playing an MP game vs. humans. However, I would point out that it would benefit you greatly to improve your SP game here. I don't know your level of experience nor your friends, but I can assure you that improving your optimization in SP will help with MP.

With that said, yes, offense is the best defense in this game.

 

[Fippy]

Rah makes a great point about duels and attrition.
1vs1 is not very popular overall for that reason, declaring war or being in peace barely exist..more like always war, since declaring has no consequences.
With more players involved, diplo plays a role and they also must look at gains when planning a war, not just leaving burned earth.

You should probably talk with your friend about that, cos yup it's a bit like playing hack n slash if you duel and he comes in razing your cities.

 

[Bibor]

In my experience in multiplayer games, it's best to keep the enemy constantly distracted, forcing him to deal with apparently trivial things that divert his attention and resources.

For example, not to be taken literally, tech metal casting fast, build two triremes, pillage his sea resources then retreat to a seazone he can't see. Now he can't possibly know when you'll be back or if you simply went away.
Build a galley, embark a chariot and a hill archer, snipe a worker, pillage his copper...

Generally be a pain in the butt.

As with every strategy, the question is where is the pressure point. You don't need a big navy to relieve pressure from your cities, what you need is constant pressure so he can't focus his attack.

And then there's morale. If you can tank his morale, you don't need to win at that point. For example, if you can raze only one city, but one that contains a critical economic wonder like Pyramids or The Great Lighthouse, you basically already won, because he can't recover psychologically from that defeat.

Also, be diverse in this. Once you win, play the next game slightly differently. Perhaps don't send a chariot and and an archer, but six horse archers next time. Or play Mansa Musa and dot a few skirmishers on his hills. You'll see him sweating soon enough

With astronomy contact, I guess its a bit more tricky, because it's based on tech pace and who's first to galleons. But again same principles apply. Send a few galleons just to pester him, perhaps put a spy or two to scout him out, take out whatever hurts most in his capital (marketplace, forge etc) then send a galley with two horse archers just to pillage his remote resources.

It doesn't matter if you "waste" hammers, these hammers are not wasted, they are used to let you get to chemistry and land before he does. Because you are free to do what you want, while he's forced to constantly react. Invest these hammers as if they were a "national wonder"

 

[6K Man]

(I’ll preface my comment by saying that while I haven’t played a lot of MP, my experience of MP games is that they are largely military contests. When you say you focus on infrastructure and economy, that’s a bit of a red flag. The player who builds 10 more Axemen is going to beat the player who builds the Pyramids instead, if the former can reach the latter in the BCs. Overbuilding infrastructure – unnecessary Libraries, Courthouses, Temples, etc – will get you killed by a more practical opponent)

Really, it depends on what point in the game you are at. As Fippy said, air units do an excellent job of scouting – but I suspect you aren’t nearly that far into the tech tree, if you are playing a duel.

In general - If you don't already, consider building coastal cities on hills, clear surrounding forests ASAP, and build mobile defenders – a force of 6-10 Horse Archers makes more sense than keeping 3 Archers in every city.

For early (pre-Astro) games, your opponent will be constrained by three things: boat capacity, crossing places, and movement rate. If you’re on separate islands, it pays to have a Scout or ship placed in a way where you can see the invasion fleet coming, and this is easier when your opponent can only cross Coast tiles. Galleys move 2 and carry 2, so they’re slow and the attacker automatically needs 1/3 of the invasion force to be Galleys, which don’t add anything to the attack. So your defense can be comparatively economical – you can have parity in land units for 2/3 the hammers your opponent has to spend. Figure out how much warning (in turns) your scouts will give you of a pending attack, and have hard counters (e.g. Spears, Axemen) within that many turns of your coastal cities.

Post Astro: This is harder ; there’s no crossing point issue, Frigates can take away cultural defense quickly. It’s especially tough before Airships give you 2+ turns of lead time to reinforce trouble spots. Mobile defense is still the way to go, IMO – Cuirs/Cavalry with Pinch promo. In most cases your opponent won’t have Amphibious-promoted units, so you get an extra turn to defend while they land before attacking, which will allow you to move some defenders into place, and whip others.

 

[PhilPeter]

You guys are great. Thanks for all the ideas. I will definitely try to be more of a pain in the butt
I was mostly thinking about the post-astronomy era.

 

[Gavagay]

An MP veteran here. Some suggestions:

1) It is true that defending against naval attacks is tricky. Overall, there is an enormous defender advantage in civ4 but when it comes to naval warfare, it all but evaporates.

2) You need a sentry net of ships around the whole coast of your island. By the time naval invasions are a concern, the cost of these ships should be negligible, if you have 10-15 coastal cities. Consider building privateers for this purpose, they come with a built-in sentry promotion.

3) After physics airships are your best friend. They make any kind of surprise attack pretty much impossible.

4) Naval warfare is one of the reasons I believe that protective trait is a bit underrated, though it still sucks. This is because static defense, as you have rightly pointed out, becomes important. With protective trait, you should be able to spam something like CG3 Muskets/Rifles easily and breaking through them is a huge pain.

5) Keep in mind that if your enemy wants to take a roundabout way and attack your cities from an exotic angle, it means that he would need to essentially put a large chunk of his army out of action for a few turns. This is a weakness which you can exploit.

6) Battleships can take barrage which gives an advantage in naval warfare back to the defender. Air units also make defending easier. But, of course, few MP games last that long.

7) Generally, you want your enemy to be afraid of your counter-invasion, even if you are not planning one. A popular trick is to have empty transports taking threatening positions. It would make your opponent reluctant to put his army on boats and swim to the sun.

 

[PhilPeter]

Good post, thanks a lot, Gavagay! You make many good points, but it does not change the fact that the attacker has a huge advantage. The attacking strategy is much more efficient.
Assume the attacker approaches with 1 frigate and 5 galleons, each carrying 3 attacking units (21 units in total). Such an attacking force would presumably suffice to destroy at least one of my cities. Assume I set up a sentry net of ships around my island. If I have 10-15 coastal cities, this requires at least (!) 20 ships. But these ships will not suffice to sink the attacking fleet, because they are too dispersed. So I will need even more ships to not only detect but also sink the attacking fleet. I would need maybe another, say, 4 fleets of 3 frigates each, strategically positioned around my island. In total, I would thus need at least 32 ships to detect and defeat the attacking fleet consisting of only 21 units. And this has been an optimistic estimate, and it is still a risky strategy. Also, assume that the attacker arrives with 4 frigates instead of 1. It would not be enough to build an additional 3 or 4 frigates to defeat this more powerful fleet. Rather, EACH of my 4 defensive fleets would have to be bolstered. The attacking force is very concentrated, but your defensive efforts are necessarily de-centralized. This is why preempting an attack is so much more costly than launching an attack.
Or am I wrong?

 

[Gumbolt]

Why wait for stack to arrive? Caravels can scout enemy seas.

Best form of defence is attack. No reason for you not to be razing his cities. Don't wait around for an attack. You will lose every game on that basis.

Check also power graphs in game to see if the other player is spamming units.

 

[Gavagay]

Good post, thanks a lot, Gavagay! You make many good points, but it does not change the fact that the attacker has a huge advantage. The attacking strategy is much more efficient.
Assume the attacker approaches with 1 frigate and 5 galleons, each carrying 3 attacking units (21 units in total). Such an attacking force would presumably suffice to destroy at least one of my cities. Assume I set up a sentry net of ships around my island. If I have 10-15 coastal cities, this requires at least (!) 20 ships. But these ships will not suffice to sink the attacking fleet, because they are too dispersed. So I will need even more ships to not only detect but also sink the attacking fleet. I would need maybe another, say, 4 fleets of 3 frigates each, strategically positioned around my island. In total, I would thus need at least 32 ships to detect and defeat the attacking fleet consisting of only 21 units. And this has been an optimistic estimate, and it is still a risky strategy. Also, assume that the attacker arrives with 4 frigates instead of 1. It would not be enough to build an additional 3 or 4 frigates to defeat this more powerful fleet. Rather, EACH of my 4 defensive fleets would have to be bolstered. The attacking force is very concentrated, but your defensive efforts are necessarily de-centralized. This is why preempting an attack is so much more costly than launching an attack.
Or am I wrong?

Three points.
1) You are right that purely defensive strategy wouldn't work in naval conflict, especially when your enemy knows that you only play defensively. But if he does not, then he would not be able to send 21 unit towards your land by long, roundabout road. While these units are inroad, they cannot be used, neither defensively, nor offensively and this creates for you a window for your own counter-attack. In practice in a situation of power parity such attacks are too risky which means that you only need to worry about the shortest approach to your shore.
2) Cities are not equal in value. Your strong cities should always be overprotected and I do not think that burning a fishing village worth risking 21 units plus transports. You are talking about having 10 coastal cities which is quite a lot. Losing one of them cannot be the end of the game but while losing 20 units potentially can.
3) I forgot to mention last time - you can increase the mobility of your own units greatly if you constantly keep your reserves in water. Do you know chain-transport trick? You can put your units on a transport and move this transport to a tile with another transport. After that, you can move your units to a new transport with free movement points, even if those units have used up their own movement points. Then you can move the transport to the next transport and repeat the trick. In this way, units can be ferried through indefinite distances in one turn, as long, as you have enough ships for an uninterrupted chain. It gives you a possibility to reinforce your coastal cities if you have at least one turn warning which you should get just from culture-vision.

 

[Gavagay]

In other words, while you are generally right that on a sea-heavy map there is no usual defender advantage, you are overestimating the attacker advantage because you do not try to put yourself into attacker's shoes and do not appreciate problems the attacker faces. From attacker's perspective putting units on boats and moving them away from his own coast is always a considerable risk. And you only consider the best-case scenario for the attacker - the one in which he managed to slip undetected and raze something valuable enough. The attacker himself is afraid of the scenario in which his fleet is tracked down and destroyed. This is why he may be more reluctant to do this thing.
Also, we are talking about a duel scenario which is not really the situation the game is balanced for. In a normal game, naval invasions are complicated by the possibility of your other neighbors backstabbing you while your army is far away.

 

[PhilPeter]

@Gumbolt: Exactly, offence is the best defence. That's what I've been saying. But my question was whether it is possible to improve one's defensive strategy in such a way that the defensive strategy is better than the offensive one.

@Gavagy: I did not know about that chain-transport trick! This is really cool! Thanks!

Regarding 1) In my experience, it is not true that the attacker picks the nearest city for the attack. Naval units move so quickly that he can basically attack wherever he wants. Its true that there is always the danger of a counter-attack. But the probability of a counter-attack during the period in which your attacking fleet is on its way (say, 10-15 turns) is relatively low. And once you've attacked, your opponent won't have the resources to launch a counter-attack.

Regarding 2) Good point. I just hate loosing cities, though, even if it is just a fishing town. I just love my cities so much that this is something that I really struggle to recover from psychologically. But you're right. Defending is easier if you focus your efforts on your most important cities. Also, since the attacker is likely to target your most important cities, your more likely to intercept his fleet. Unless of course he knows your strategy, in which case he will target your small cities, knowing that they are virtually defenseless.

I guess the best way to correct for this bias in favor of an attacking strategy would be to turn off the city razing option.

 

[sinimusta]

One thing that hasn't been mentioned here is the astro bulb path with great scientists. If you get astro way earlier than your friend you have a huge naval advantage against him and could get a nice surprise if your friend doesn't pay too much attention to tech situation. You just need to remember the teches you need to avoid in order to be able to bulb that way, I think at least medi and CS are to be avoided, but the path should be found in deity isolation thread or somewhere else here.

 

[Windsor]

- Sea blockade. If you position naval units all around your opponent’s island, you’ll notice when and where his naval attack takes place. The problem with this strategy is that you’ll need loads of ships. You need many ships only to encircle his island, and you need even more ships for several strategically positioned fleets that can sink the opponent’s ships as their pierce your blockade. This strategy is hardly any better than building plenty of defensive land units in your cities.

I totally disagree. You don't need many ships and this is a very effective strategy.

If he's supposed to launch an effective attack he first needs as many warships as you do. Then he needs transports and then units.

If you manage to win the initial fight over the seas, a comeback is very hard

 

[PhilPeter]

Can you elaborate on how you pull this off, Windsor? For instance, how do you make sure that you even detect the opponent's fleet? On a large map with, say, 10 coastal cities per player, it seems very difficult to keep track of the enemy's activities, and even more difficult to negate an attack. It seems so much easier to attack oneself than to preempt an enemy attack. Or am I just too clumsy? As a matter of fact, I have always found ways of circumventing my opponent's naval defense system.

Also, does your strategy work if you and your opponent discover astronomy at roughly the same time? I can see how it might work if you develop naval technologies earlier than your opponent (as Sinimusta suggests). But what if you're equally advanced?

 

[Bibor]

It's a question of economy. 10 frigates cost 900 , an invasion fleet with 20 units costs in the ballpark of 3000 (galleons+rifles+cannons) .

A coastal city focused on production at this time can yield, what, 20, perhaps 30 hammers per turn? Lets say he does has have 5 coastal production cities, that's 100 per turn, so he'd need 40 or so turns to build up the invasion force (less with whipping/drafting of course). A force that can be sunk in one turn, if one's not careful.

 

[TheMeInTeam]

Attacker advantage is pretty much unique to naval maps and nukes. For PvP collateral initiative using roads/vision inside territory is decisive w/o major tech/unit lead.

In the water the script flips. Put a navy between 2-3 cities, and the embarked units can threaten any of those 2-3 cities' garrisons after bombarding them. Any city that gets taken can be burned same-turn, so there's no retaking. Getting most of your coastal cities burned while opponent doesn't nearly guarantees a loss. You have no choice but to properly invest in navy and leave cheap boats out there as screens so you can see ships coming and react.

If you *do* invest in navy, allocating more production to that than your opponent (who built military units) gives you the advantage in theory. In practice you still need equal or better ships and to see it coming.

Winning at sea and blockading someone can hurt them significantly if it's a map depending on seafood.