Are there any tricks to naval combat? Why do my battleships suck?

[ConfusedCounsel]

Come on now; all posters here have surely played enough hours to know what is truly meant. Of course things will even out over 3000 battles. But if you have for instance 5 battles with winnings odds in the early game, and lose 3-4 of 5, then there is a good chance that rush is over, you can't take the city, defenders promotion heal, new guys are whipped, reinforcements come from nearby cities, maybe even they'll counter and kill your surviving units.

Yes. And this right. You are taking a risk attacking, even in 5 battles with winning odds. On average, you will lose 3-4 out of those 5 a number of times.
You might called that 'screwed by the rng', but it happens and it happens as often as the odds say it should. You are taking a risk that happens, say 10% of the time and it happens 10% of the time. How is that screwed? You are taking a risk, even a small one and it may backfire.
Occasionally, you will also get lucky, and survive 10-15 battles with 80% odds with no casualties, leading you to your HoF game or PR or finally cracking that difficult NC game or whatever, but I don't hear anyone complaining about how well their game went because they were just too lucky.

It is of course feasible to think of a system with more deterministic features. The more random rolls determine something, the higher the likelyhood of the expected average or something close to it. The question is, should it? Personally, I think no. If you are that annoyed by a bit of bad luck, you can always reload (with random seed option if you want). If your pride or your gamer spirit prohibits you from doing that, you essentially agreed to a game where randomness will occasionally kick your butt.

In terms of realism this thread makes no sense. The game is FAR more deterministic then the real world. Battle is messy and for large portion of history, open battles were avoided like the plague because of their inherent unreliability.

With more rounds needed to win each battle, the battles in the early game should in general be closer to the stated odds too, with less deviation due to "bad rolls".

Euh, no they wouldn't. As multiple people have already shown, the shown odds are correct. More rounds will simply change those odds in favor of the favored party.
An example: I play a game of tennis with a friend. I'm favored to win a set by a constant 60% or .6 . In a best of 1, I win 60% of the time. In a best of 3, I win .36 + .288 = .648. I don't feel like calculating the odds for best of five, but they go up. In a best of 10001 I'm approaching astronomical unlikelyhoods of actually losing.
The game would simply show 60% or 64.8%. Those would also be the odds. The 'deviation', would still be the same, ie: you will still lose 40% or 35.2%.
It would make the game easier though. Assuming the player is more often the one with >50% odds, you will simply win more combats. You could also just play at a lower difficulty, though.

 

[rbj2001]

I understand the desire for more predictability, but IMO that would be less fun, and would be unrealistic. There are plenty of historical battles where the lesser army has prevailed due to luck (for a nautical example, the invasion of the Spanish Armada, which got routed due to freak weather). And armies don't generally go into battle knowing they're going to lose. There's always chance involved.

The logical conclusion would be to remove the RNG altogether, and make it so that a strength 5 axeman beats a strength 4.99 axeman 100% of the time, finishing with a deterministic number of hit points. You could march into battle with an army only 1% stronger than your opponent's, and know you'll always win.

Where would be the fun in that? Humans like to gamble and enjoy games of chance.

It causes great luck and great happiness but also great frustration, one can lose a whole game in one attack and that without having made a single fault or having made a bad decision and that goes too far for me.

It comes down to being a good loser. If you stopped playing Civ and played other games with real people, you'd learn that the give and take is part of the fun. There's no fun in sport if the better team or player always wins. Your aim should be to on average win more than you lose, not to win every time.

It's like life. Unpredictable things happen, and you don't always win, even if you do everything right. Except that playing Civ doesn't need to be taken as seriously. It's only a game, so don't allow yourself to get frustrated by it.

 

[Lexicus]

Occasionally, you will also get lucky, and survive 10-15 battles with 80% odds with no casualties, leading you to your HoF game or PR or finally cracking that difficult NC game or whatever, but I don't hear anyone complaining about how well their game went because they were just too lucky.

Actually, I was playing a game not long ago where I think I won 3 in a row at ~30% odds (not sure because stack attack was on) when I was attacking my first AI, which allowed me to take a city I probably shouldn't have been able to take.

Knowing I had gotten so lucky did kind of annoy me the whole rest of the game.

 

[Dhoomstriker]

It is frustating when battle ends with oposite unit having 1/100 HP left just because each round my unit damaged was 33 not 33,3(3) (so it should won battle) but AI unit got lucky 4 winning strikes in a row).

Using the Combat Odds Calculator, I came up with an example where the attacker does 33 damage per round:
A(ttacker) original str(ength): 8
A(ttacker) generic bonus: 10
D(efender) original str(ength): 3
D(efender) first strikes: 1

That example represents your Combat I War Elephant attacking an enemy Archer with its built-in First Strike and who is defending on flat terrain, not across of a River, and not inside of a City or a Fort.

Here are the values that the Combat Odds Calculator provides:
Combat odds: 99.94%
Retreat odds: 0%
Survival odds: 99.94%
A experience points: 1
D experience points: 5
A modified strength: 8.8
D modified strength: 3
A hit per round: 33 hps
D hit per round: 12 hps
A odds to win a round: 75%
D odds to win a round: 25%

In particular, notice that when you do 33 hps (hitpoints) of damage, the defender is doing 12 hitpoints of damage. So, even if the defender was able to get "lucky 4 winning strikes in a row," and let's say that it also got its First Strike attack for a total of 5 attacks on you in a row, that would still only be 5 * 12 = 60 hitpoints.

In order for the Archer to win, it would have to win ceiling( 100 / 12 ) = ceiling( 8.3 ) = 9 rounds of combat. Assuming that it won its First Strike attack, that would mean that it would have to win an additional 8 rounds, while you'd only have to win 4 rounds.

And, yet, according to the calculator, the odds for the defender to win a given round are only 25%.

So, not only does the Archer have to damage you in 9 rounds before you manage to damage it in 4 rounds, it also only has a 25% chance of winning each given round, with only the First Strike round being (if I understand things correctly) a 50% of doing either 0 damage or one round-worth of damage (12 hitpoints, in this example) to your War Elephant.


If we plug in equal values into the Combat Odds Calculator, say, for two unpromoted Axemen in open-field combat:
A(ttacker) original str(ength): 5
D(efender) original str(ength): 5

then we get these values:
Combat odds: 50%
Retreat odds: 0%
Survival odds: 50%
A experience points: 4
D experience points: 2
A modified strength: 5
D modified strength: 5
A hit per round: 20 hps
D hit per round: 20 hps
A odds to win a round: 50%
D odds to win a round: 50%

As you can see, a "normal" round will deal 20 hitpoints of damage. So, it's only when you're already at a large advantage in the combat when you will deal 33 hitpoints of damage per round.


Note that if we plug in values as the following, basically giving your War Elephant Combat II against the unpromoted Archer with no Promotions and with no defensive bonuses:
A(ttacker) original str(ength): 8
A(ttacker) generic bonus: 20
D(efender) original str(ength): 3
D(efender) first strikes: 1

then we reach the break point of the War Elephant only needing to win 3 combat rounds in order to win the battle:
Combat odds: 100%
Retreat odds: 0%
Survival odds: 100%
A experience points: 1
D experience points: 6
A modified strength: 9.6
D modified strength: 3
A hit per round: 34 hps
D hit per round: 11 hps
A odds to win a round: 76%
D odds to win a round: 24%

since ceiling( 100 / 34 ) = ceiling( 2.94 ) = 3 rounds of combat.

Also, the Archer now needs to win ceiling( 100 / 11 ) = ceiling( 9.09 ) = 10 rounds (or 1 First Strike round and 9 regular rounds) in order to win the combat, with a slightly-lower (24%) chance of winning any given regular round.

Note that if the Archer did win, it would now receive 6 Experience Points instead of just 5 Experience Points; I believe that it was troytheface who posted about Attacko's suggestion of fighting combats with units even at low odds with the hopes of getting good Promotions. Given the right context (say, toss in unpromoted units when assaulting a City early in the game before siege units as your first attackers and where you have a large stack of units to back them up so that you will actually win the battle for the City), this advice is actually quite sage: your unpromoted Archers, Chariots, or Axemen who go in first don't expect to come out of the battle alive, but if they do, you very well might end up with being able to assign those units 2 Promotions on the following turn. Of course, bring enough other units to actually win the battle for the City, otherwise you might lose your wounded units in the AI's counter-attack before your units can promote on the following turn.


Using a boat-based example:
An unpromoted AI Destroyer attacking your defending, unpromoted (or Flanking I) Battleship:
Combat odds: 22.07%
Retreat odds: 0%
Survival odds: 22.07%
A experience points: 5
D experience points: 1
A modified strength: 30
D modified strength: 40
A hit per round: 17 hps
D hit per round: 23 hps
A odds to win a round: 43%
D odds to win a round: 57%

Give your Battleship Combat I:
Combat odds: 11.21%
Retreat odds: 0%
Survival odds: 11.21%
A experience points: 5
D experience points: 1
A modified strength: 30
D modified strength: 44
A hit per round: 16 hps
D hit per round: 24 hps
A odds to win a round: 41%
D odds to win a round: 59%

Give your Battleship Combat II:
Combat odds: 4.13%
Retreat odds: 0%
Survival odds: 4.13%
A experience points: 6
D experience points: 1
A modified strength: 30
D modified strength: 48
A hit per round: 15 hps
D hit per round: 25 hps
A odds to win a round: 38%
D odds to win a round: 62%

Give the attacking Destroyer Combat I and your Battleship Combat III:
Combat odds: 4.77%
Retreat odds: 0%
Survival odds: 4.77%
A experience points: 6
D experience points: 1
A modified strength: 33
D modified strength: 52
A hit per round: 15 hps
D hit per round: 25 hps
A odds to win a round: 39%
D odds to win a round: 61%

Attacker hitting 17 hitpoints per round = ceiling( 5.8 ) = 6 rounds needed to win
Attacker hitting 16 hitpoints per round = ceiling( 6.25 ) = up to 7 rounds needed to win
Attacker hitting 15 hitpoints per round = ceiling( 6.67 ) = also 7 rounds needed to win

Defender hitting 23 hitpoints per round = ceiling( 4.35 ) = 5 rounds needed to win
Defender hitting 24 hitpoints per round = ceiling( 4.17 ) = 5 rounds needed to win
Defender hitting 25 hitpoints per round = ceiling( 4 ) = down to only 4 rounds needed to win

So, promoting to Combat I means that the attacker needs 1 more round to win the battle.

Promoting to Combat II means that the attacker needs 1 more round to win the battle and you need 1 less round to win the battle (relative to you have 0 Promotions or having the Flanking II Promotion*). You promoting to Combat III with the defender being at Combat I just so happens to give similar results in this case, although with the attacker having just slightly higher odds of winning any given round.


If you are defending from the Coast, I believe that you would need to fill in:
D(efender) situational bonus: 10

If my understanding of applying the Coastal defensive bonus is correct, then doing so appears to be the equivalent of an extra Combat Promotion for the defender, according to the odds shown from the Combat Odds Calculator. So, looking at the numbers above, it means that you can have 1 less Combat Promotion and still get the same results if you end your turn on a Coast square.


* Of course, if you are playing multiplayer, Flanking II has a sneaky effect of making your unit immune to First Strikes, so if someone has been making Drill IV Privateers and then upgraded them to Destroyers, Flanking II can be of use, even on defence, as Drill IV gives between 3 and 6 First Strikes. But, AIs don't do that; at most, you'll see Drill II for between 1 and 2 First Strikes.

Let's run those odds (the AI having Drill II), just to see how the numbers look:
Attacking Destroyer has Drill II (1 First Strike chance and 1 First Strike = 1 to 2 First Strikes)
Defending Battleship has Combat II while not being on the Coast:
Combat odds: 6.47%
Retreat odds: 0%
Survival odds: 6.47%
A experience points: 6
D experience points: 1
A modified strength: 30
D modified strength: 48
A hit per round: 15 hps
D hit per round: 25 hps
A odds to win a round: 38%
D odds to win a round: 62%

Attacking Destroyer has Drill II (1 First Strike chance and 1 First Strike = 1 to 2 First Strikes)
Defending Battleship has Flanking II (no combat bonus but we remove the attacker's First Strikes, meaning the same case as our first example of no Promotions on either side) while not being on the Coast:
Combat odds: 22.07%
Retreat odds: 0%
Survival odds: 22.07%
A experience points: 5
D experience points: 1
A modified strength: 30
D modified strength: 40
A hit per round: 17 hps
D hit per round: 23 hps
A odds to win a round: 43 %
D odds to win a round: 57 %

Here, we have to look at the Combat odds, since First Strikes are involved, and we see that Combat II pulls far ahead with the attacker only having a 6.47% chance to win, instead of a 22.07% chance to win when you promoted up to Flanking II.

If the attacking Destroyer has Drill IV and the defending Battleship has Combat II when not on a Coast:
Combat odds: 14.46%
Retreat odds: 0%
Survival odds: 14.46%
A experience points: 6
D experience points: 1
A modified strength: 30
D modified strength: 48
A hit per round: 15 hps
D hit per round: 25 hps
A odds to win a round: 38%
D odds to win a round: 62%

Even in this case, the First Strikes are not enough to overcome the gains associated from hitting the breakpoints of the attacker needing an extra round and the defender needing one less round to win.

So, even in the contrived example, Flanking II isn't doing a lot for you on defence. If there was going to be a lot of stack-versus-stack boat combat in your game, you could argue that Flanking II on a Battleship would allow it to do some Collateral Damage to the entire stack of defenders while having a chance to retreat from the battle. But, in practice, this type of a battle is quite rare and you're more likely to be fighting small skirmishes with small numbers of enemy boats. You're probably most likely to fight a larger number of AI boats in a single turn of combat when on defence when an AI throws boats from multiple squares against your same square. Thus, go with the Combat Promos.

Flanking I is nice for transport ships, though, as, if you are Charismatic with a Drydock, you have a Great General settled in a City with a Drydock, or you can have your Drydock-built boat earn 1 Experience Point, you can get your Galleon or Transport up to 5 Experience Points so that it can promote to Navigation I for +1 movement point, which is a Promotion that gets unlocked by Flanking I. I would not aim to get Navigation I for your combat boats, though; those boats can already move far enough on their own and you want them to have Combat (or maybe Drill) Promotions for doing the actual fighting.

 

[lindsay40k]

Three consecutive wins at 30% is not that extreme. Over the course of a conquest game, you'll probably have at least thirty battles. If each one has three throwaway vanguard units whose role is to soften up a defender and die, statistically you're going to have one where the Banzai charge actually works. That's not a faulty RNG, rather that's exactly what you expect from a combat system where units are all willing to fight to the death.

Spearman beats tank? Of course they do. If the tank's engine stalls or a tread gets stuck in mud with cover nearby, anyone with the use of their limbs and access to fuel, fire and a bottle can get an armour kill.

Transport kills Battleship after Destroyer hits it? Sure. "Captain, our escort's sinking. But the enemy ship is aflame. I estimate they'll extinguish the fires and sink us within fifteen minutes." "GIVE ME RAMMING SPEED! ALL HANDS, BRACE FOR IMPACT!"

 

[Kid R]

Adding rounds would not have any effect winning streaks or losing streaks.
....
In other words, it would have an indirect effect on losing streaks.

So, indirect but still definitely an effect. I suspect it's the indirect effect, i.e. from the variance, that's more off-pissing. OK you lose 4 90% fights in a row, it happens, but it's the fact the game made you even have those 4 fights instead of one at 90% and then 3 at better odds, because the funking top defender keeps escaping with no damage. I think there's something especially maddening about it when you lose a whole unit and its death makes absolutely no difference at all.

Probably realistic, but in gameplay terms maybe less ideal? The early game (i.e. few units, so more vulnerable to swingy effects) is so important maybe it should be more predictable, and swing brought in gradually through the game. Same principle as goody huts I guess - people do like the wild swing from them so it's hard to say one right answer.

 

[Luthor_Saxburg]

So why are my battleships getting sunk by destroyers? Why are my destroyers getting seriously hurt by ironclads? What am I doing wrong? What should I be doing?

I have read some of the comments on this thread and it seems that the thread has gone in the direction of discussing probabilities and the civ combat system. And that is fine. But I will focus more on the "what should I be doing" part.

I am not much of an expert on naval warfare! But I can give a (obvious?) hint or two:

Avoid having your ships out in the open
After playing a lot of modern single player games, I've realized that the AI does attack at unlikely odds! And it often wins, as you've noticed. So I just try to keep ships in the ports as much as possible. This way I attack when I choose and not be subject to the strange naval combat odds.

Used the full extent of combined forces: sea and air
I choose to keep the option of attacking instead of defending whenever I can. Off course a good modern naval warfare, combined forces should be used:
- Use airforce to hurt the enemy ships before attacking
- Use missiles to damage the ships (I use missiles mainly for naval warfare and very few for land combats)
- Use units with good withdraw abilities (submarines) to try and hurt them a little more without much damage
- After all of the above possibilities have been used, Battleships with strong colateral damage abilities should be used
- In the end I then use all the other ships for battle, including (empty) Transports and Carriers if the odds are looking good.

.

 

[rah]

Come on now; all posters here have surely played enough hours to know what is truly meant. Of course things will even out over 3000 battles. But if you have for instance 5 battles with winnings odds in the early game, and lose 3-4 of 5, then there is a good chance that rush is over, you can't take the city, defenders promotion heal, new guys are whipped, reinforcements come from nearby cities, maybe even they'll counter and kill your surviving units.

I couldn't agree more. The odds are what they are.

 

[Lexicus]

Three consecutive wins at 30% is not that extreme.

But, it seems extreme to me because my puny human brain isn't good with probability

Also, to be clear I'm not faulting the RNG, I'm faulting the way combat works.

I agree with those who think that changing the way combat works to "smooth out" the results given, without changing the RNG, would be a good idea. I can understand those who disagree though.

 

[lindsay40k]

Yes, let's bring this back around to OP.

The 'fight to the death' mechanic meant that you could plonk a Trireme at each end of your territory and forget about barbarian galleys. Whipping two would deal with AI incursions. With Chemistry, a Privateer in each city could snipe Caravels, a Frigate neutralised the rare AI Privateers, and a fleet of them could kill an entire invasion force before it even declared war.

When Oil starts flowing, all bets are off. There's little point in attempting point defence; a small battle group will easily overpower your sentry, and take out a city's seafood. Naval warfare is brutal, expensive, and fast. Subtlety comes in when you decide the order in which units attack, and strategy comes in when you decide a route and target for your giant armada.

Aside from Submarines, your ships will not run away, and with no decent defensive terrain to exploit you'll have to accept that underdogs will frequently get a lucky shot below the waterline. Even if you overpower them, a sole survivor will often be within another hostile's threat range and be sunk.

Pick your battles, move units in groups, and keep reinforcements and medics safe.

 

[Lexicus]

And for the love of God don't leave transports or carriers unprotected.

 

[Pir Lan Tota]

...

In a nutshell what Lindsay says, but some more tips.

1. Fight from save Harbours, e.g. leave your city to engage enemies and retreat into your city again before the end of the turn, so enemy cant counter attack

2. As mentioned before, use bombers and/or carriers with fighters to soften up enemy stacks. Alternatively also use cruise missiles.

3. Flanking Submarines are great. They wont win the battle, but with 80% withdraw chance, they will most likely survive for the big boys to win the next battle.

4. Use "Throw-away" battleships. Battleships with max stack-attack damage. Yes it will die, but do so much damage that the follow up battles are easier to win for the rest of your fleet.

5. Never ever defend if possible, always attack or hide in cities. You must control the battle.

 

[Tatran]

Over the years I've learnt to ignore the combat odds, I just look at how many HPs my unit will do, will receive and make a decision to attack, defend of withdraw.
Something I found more annoying is the randomness in catching a spy in the early game.
I've lost 6 spies in a row (2 failures + 4 for travelling in foreign lands) while the opponent had't Alphabet + I had more EPs. A huge delay in bulbing Engineering + rushing the Notre Dame with a GE. Not fun, so end of game. Just like the game in the spoiler. Excellent start. You can't see the floodplains + river corn + gold + 10 forests.

Btw, after all these years I never knew that only coastal tiles give +10% bonus when defending.

Spoiler :

 

[Pangaea]

^^ Wow!

This one isn't as shocking, and with a positive result, but 6 rounds in a row with losing odds, that's pretty wicked.

Spoiler :

This isn't just like getting the same side of a coin 6 times btw, because with losing (or winning) odds, there is a higher or lower chance to win each round. It's not a straight 50/50.

 

[Tatran]

6 rounds in a row with losing odds, that's pretty wicked.

Wait until you lose 17 rounds (iirc) in row.
Those screenshot(s) must floating around somewhere at this board.

 

[Pangaea]

If this calculator is correct, this would be pretty close.



Though I'd say the chance of losing 15 times in a row (or 15 of 16) with a 96% win chance per round isn't very high!


(The MechInf has Combat V btw, against a plains Warrior)

 

[0ttovonBismarck]

The combat odds for ships are not great... I often will lose a highly promoted ship just by a bad roll of the dice.

 

[Seraiel]

It comes down to being a good loser. If you stopped playing Civ and played other games with real people, you'd learn that the give and take is part of the fun. There's no fun in sport if the better team or player always wins. Your aim should be to on average win more than you lose, not to win every time.

It's like life. Unpredictable things happen, and you don't always win, even if you do everything right. Except that playing Civ doesn't need to be taken as seriously. It's only a game, so don't allow yourself to get frustrated by it.

It's a long time since I read this thread and I haven't read anything but on this I feel I need to comment:

I played player vs. player and group vs. group PvP in MMO's for over 10y hardcore and after 1 year I was in a team that won 80% of the battles. In the end I was able to easily beat 10-15 people solo and when we ran as a group the other side needed to come with their complete force (hundreds of players) to wipe us and while they were doing that we killed more than 50% of them. If less than everybody came we often killed zergs up to 250 people in size with 6-10 players.

So very seriously, I invest a lot of time in games and I don't have any problem with losing but it's the situation that Pangaea described that makes me really seriously angry because when I play 8h into a game and don't make a single fault and then attack with a number of units that only a very limited number of players in this forum would be able to have at that time and then lose all of those units because the RNG decides that i. e. the first 3 units make 0 damage because of too little combat rounds or when it decides that losing 4 fights in row if having 80% chances each... That's what can make a game unplayable because this forces to always take risks because attacking with completely safe numbers wouldn't be competetive anymore and those scenarios imo. just happen too often and that's why I wrote that I'd appreciate if the outcomes were closer towards the odds. I never wrote "make the game deterministic and kill the RNG" or anything because I know that that would take fun but the results imo. need to be closer towards the results shown and that for a small number of fights because early game is about small numbers. Late game with hundreds of units the RNG is of no matter anymore but in the early game and with it's great deviation that is possible it is too powerful imo.

 

[sampsa]

Big thumbs up for elitetroops. Good posts and good manners.

My deepest sympathies for the whiners.

 

[rah]

Yeah, eventually I'd love to repeat his test using GGs, to see if the results are the same.