The final & only discussion thread for naval battle mechanics

[Knight of the Temple]

I'm fairly new to MTW, and I was wondering why caravels seem to get sunk consistently by barques, by ships that seem inferior stat wise. As soon as I was able to build caravals for the first time, they seemed like a definite improvement over the barque seeing as their attack and defence was higher, their support costs lower, and their range much greater, so I built hosts of caravels and retired all my barques. I've recently been experiencing my first naval war and I've been surprised by how poorly my caravels seem to be doing against barques. I've pretty much won the naval war now, and I'm just clearing up what's left of my enemy's ships, which all happen to be barques, but they certainly aren't easy to get rid of with caravels. If I attack 1 barque with 2 caravels it'll sail away and avoid the fight, and I accept that it's fast enough to do that, but just recently I decided to try and entice certain barques into a fight by splitting up my caravels and attacking one by one. Well, it did work in that the barques stayed to fight, but my caravels were defeated every time - I lost 4 caravels and sunk no barques at all. And most of my caravels had 1 or 2 stars. So, I don't get why caravels fair so poorly against barques. Not surprisingly, I've started to reintroduce some barques into my navy, and hopefully I'll be able to defeat the enemy barques with them, as I can't afford to just throw away all my caravels every turn! Have I just been unlucky or are barques generally better than caravels in a one on one situation?

[The Unknown Guy]

There are always surprises in naval battles. It´s not only a matter of attack and defense, but of speed and command, and (I think) whether you have a dock there or not.

My theory is that attack is the attack rating, "defense" is the hitpoints (how many attack points it can sustain), and speed regulates how many rounds of attack per turn has each ship (IE: a dromond fights four times in the time a wargalley fights one)

[Don Esteban]

There are always surprises in naval battles. It´s not only a matter of attack and defense, but of speed and command, and (I think) whether you have a dock there or not.

My theory is that attack is the attack rating, "defense" is the hitpoints (how many attack points it can sustain), and speed regulates how many rounds of attack per turn has each ship (IE: a dromond fights four times in the time a wargalley fights one)

Intersting theory but I think speed just has to do with evading battle rather than winning it. Not sure about docks but i'll look out for that - it would make ssense to have an advantage in "home waters".

[caravel]

Defence and attack stats are only part of it. A higher valour (referred to as command on the info pic) barque will usually be more effective than a 0 or 1 valour caravel. Speed is a stat that is used for ships that are fleeing. If you attack a ship one year and in that same year it moves to another sea zone, it's speed vs your ship's speed helps to determines if it escapes or has to fight (there is still a good chance it will escape though even if your ship is the same speed or faster). When you attack a ship, it's that ships defence stat vs your ships attack stat. This is how it works. Attack is usually higher than defence which is why in most cases the attacker has the advantage. When your new caravels at 0 valour are attacked by a stack of 3 barques with a 4 valour lead ship, that stack will probably win.

[Martok]

What Caravel said.

To be honest, though, the mechanics of MTW naval battles has always been a bit of a mystery to players. You'd be doing yourself a favor by heeding Caravel's words, but even then you're not guaranteed a favorable outcome, as there seems to be a high degree of randomness involved. I have a hunch that the random number generator for naval battles is particularly, well....random -- at least more so than "normal".

Oh, and welcome to the Org, btw!

[Sensei Warrior]

Not much more to add onto this one, so I'll reference a thread we had here a while back that has a bunch of good stuff about navies.

Click Here

Another ship thread only centered on trade, another sometimes confusing part
of MTW.

Click Here

I hope I did the clicky thing right, anyways have fun, oh and welcome to the Org.

[Agent Miles]

I still think that stacking ships together is a bad idea. One ship can sink the whole stack. In my last campaign as the HRE, I had about 81 ships (barques and caravels) and the Byz had a total of 43 ships (dromon, galleys and fire galleys). The command star distribution was about equal. I only attacked with about 60 of my ships. Although the AI ships were often stacked, none of mine were. I had 3 to 5 ships in a sea zone and launched multiple attacks against the stacks. I slaughtered the Byz navy in about 5 turns with fewer losses than what I inflicted. An advantage of one or two per sea zone is enough. The last attack was 5 separate caravels (speed 1) against a stack of 4 dromon (speed 4). I sank all of them with no losses. This implies that if the first attack wins, then the whole stack goes and that great speed can be overcome. I think that the greater movement radius of deep water ships gives a strategic advantage over coastal ships and I can beat them with massed attacks.

[caravel]

I still think that stacking ships together is a bad idea. One ship can sink the whole stack. In my last campaign as the HRE, I had about 81 ships (barques and caravels) and the Byz had a total of 43 ships (dromon, galleys and fire galleys). The command star distribution was about equal. I only attacked with about 60 of my ships. Although the AI ships were often stacked, none of mine were. I had 3 to 5 ships in a sea zone and launched multiple attacks against the stacks. I slaughtered the Byz navy in about 5 turns with fewer losses than what I inflicted. An advantage of one or two per sea zone is enough. The last attack was 5 separate caravels (speed 1) against a stack of 4 dromon (speed 4). I sank all of them with no losses. This implies that if the first attack wins, then the whole stack goes and that great speed can be overcome. I think that the greater movement radius of deep water ships gives a strategic advantage over coastal ships and I can beat them with massed attacks.

I know what you're saying. I always unstack my ships when attacking and always have better results than if I had stacked them. In my current egyptian campaign I'm practising stacking ships, as a result I'm losing alot more. My ships were all over the mediterranean and black sea, as were the byzantine ships. One year I decided to nip the byz in the bud and attacked all of their ships at once first. I lost about 1/3 of my ships, including entire stacks of 4 baggalas. I decided to load and try again doing the same. The result was much the same. Then I reloaded once more split my ships up and attacked with individuals. The result was that I lost two ships and wiped out probably 90% of the byz ships. The only ones left were one in the straits of sicily and another one in the black sea. A much better result. Since playing the game I've always found this to be the best method, others don't think it makes any difference. So I suppose the jury is still out on that one.

[Agent Miles]

I never leave my ships stacked, not even to move them. If I can take out a 4-stack of AI ships with one attacking ship, why would I ever let the AI do this to me?

[caravel]

I've never seen the benefit in stacking either for defending nor attacking. I tend to only unstack mine when I'm attacking though, for the sake of tidyness. The largest stack i ever use anyway is a three ship stack. I always try to keep more than 1 ship in a sea zone, otherwise the AI tries to pick them off (starting a war). Having unstacked ships means that if an AI stack attacks, only one of your ships will be engaged, meaning that you'll only lose one ship, if you lose any.

[Vladimir]

Use 3: One of your fastest and 2 of your best attack ships. I always leave them together when not attacking and never get attacked. When attacking, attack with your fastest first then your most powerful. I think ship combat is too random to know for sure wether or not single ship raids are more powerful than stacks.

[Enigma13]

While I have certainly seen the defensive advantage to not attacking (i.e. you do not lose whole stacks), I have had a much higher success rate on attacks using stacks, especially if the opponent was stacked. I tried the suggested methods for a while of separating my ships for attacks and was losing about 70% of the time. Using stacks (small stacks-3 ships) I have seen a much more balanced result. About 50/50 generally, and a small advantage (60/40) to me when my stack is larger than theirs.

[Sensei Warrior]

I have pretty good luck with stacks. For some reason I have very poor outcomes when I use single ship stack attacks. The only rule I hold to is I keep ships stacked together by type. All the barques in one stack the caravels in another.

The lack of consensus seems to point at how random naval battles are. Once I get my ship producing provences up and running they never stop producing ships of one kind or the other just to ensure I'm not on the losing end of any naval confrontations. It's worked for me so far.

[gaijinalways]

I also don't stack, and I have found generally it works well. I often use a mix of ships, and generally manage to eventually rule the seas. Pumping out ships and always spreading out your ships, though for keeping costs down, you should try to keep ships close to their origin ports (where they were made).

[Agent Miles]

Ship combat may be random, but it is not voodoo. I think that we can look at what we know and make some conclusions.
The combat has two phases. When you attack the AI, either they get away or you fight. I do not know of a situation where several ships attacked one AI ship and one of the attackers was lost but then the AI ship escaped. So the AI’s decision to fight is irrevocable. Also, if you fight a battle, it’s all or nothing. I have no knowledge of two stacks fighting and nobody being sunk.
The Decision to Run:
The AI makes a calculation and a result is applied. That’s how software works. I see three possibilities.
-The AI generates an entirely random number between 0 and 9 and either fights or runs away based on this number. If this were the case, then we would read of examples where a stack of ten AI ships fled from one attacker. Any combat at all should only happen about half the time. I submit that common experience shows that this situation is not the case.
-The AI cheats and knows that it will lose a battle and just runs away. I submit that one ship attacking a stack of four would appear less threatening than a five ship stack attacking, so single ship stacks would be better in this case.
-A percentage chance to escape is determined and applied, perhaps based on relative speed values. The AI gets away or they don’t, something like a 50/50 chance. In this case, many separate attacks win over one big attack. In the case I gave where five Caravels attack one stack of four Dromon, that would be 1/2*1/2*1/2*1/2*1/2=1/32 or about a 3% chance that the AI gets away compared to the 50/50 chance with one attacker. Even if it were a 90% chance to get away, five attackers would reduce this too. Also, a fast ship was not needed to catch the stack of Dromon, just lots of slow ships. Again, many single-ship stacks are better than one big stack.

The Fight: Stack A attacks stack B. Again, I see three possibilities:
-A number between 0 and 9 is generated and one side loses. If this is the case, then all ships are the same and there should only be generic coastal ships and deep water ships. Discussions of tactics or skill are worthless, as you would be just as well off taking advice from the Easter Bunny.
-The first attacker fights each defender in turn until it is destroyed and the next attacker continues this process until the battle is resolved. I have never destroyed only part of a stack, nor have I ever lost part of a stack of attackers (long ago when I stacked my ships) in a battle. Thus, I have no evidence to support this possibility.
-The number of ships in the stack is irrelevant. The best ship fights the best ship and it’s a one shot, one kill deal for the whole stack. This is what seems to be happening to me. No benefit accrues to stacking ships, as one attacker can sink them all.
This is my reasoning and is why I do not stack my ships, nor do I recommend doing so. Of course, I could simply out-produce the AI and win by sheer numbers. I think that, given the choice, people would rather build the right force and do the right thing, than build a large force and not know what to do. If I do the right thing with the right force, I may still occasionally lose. Such are the fortunes of war, but I won’t lose for lack of trying what is right.

[Sensei Warrior]

A good theroy, Agent Miles. Some other things to consider:

On the escape thing. I have never noticed a ship escaping combat, but staying in the same region, it always moves to an adjacent region. Which could mean that the fleeing formula might only be applied if the ship you were attacking was just moving through the region.

Since I always stack I can very easily say, I have had taken ship losses, but have not lost the whole stack. I can also say I have destoyed whole stacks (albeit small ones) when attacking as well as destroyed partial stacks while having part but not all of my stack survive.

It's always humourous to watch some advocates of single ship stack assume that only one ship in a stack fights or only one calculation is used, and its an all or nothing deal. Since I have seen every general variation of stack survival I can easily say (although not prove since I am not actively taking screen caps of my naval battles).

I am pretty sure in one of the other threads I pointed to also brought up another theroy. I will sum up the part of naval battles. A stack is ordered from most powerful (meaning valour) to least, the most powerful being the stack's 'general'. In the case of single ship stacks, it works exactly the same simply because there is only one ship. When attacking the lead (aka the most powerful) ship attacks and either wins or loses. If this ship wins then its part in the battle is over and the comp moves to the second ship in the stack if their are other enemies present. If the first ship loses then it is sunk and the process is repeated with the second ship.

In this case the attacking stack can lose all ships when the defenders don't lose any ships by failing to sink the first one. The attacking stack can also wipe out the defending stack while not taking any losses. And the attacking stack could lose some of its ships while the defending ships lose some of their ship as well.

The above outline seems logical and statistically random, and without CA telling us how they programmed ship combat, we could easily come up with 20 different ways on how they might have done it.

With that said, if you found a method that works for you then great, and you should use it.

EDIT: Cut out part of last paragraph.

[Agent Miles]

I play MTW/VI patched to 2.01 with no mods and I have never had a partial stack survive. What version do you play?

[Enigma13]

I play MTW/VI patched to 2.01 with no mods and I have never had a partial stack survive. What version do you play?

I also play MTW/VI 2.01 w/ no mods and have had partial stacks survive multiple times. Often times If I attack w/ a 3 ship stack against a 2 ship stack I can get a result such as: 1 ship lost, 2 ships sunk and achieve victory in that way.

Like I said in my earlier post I have had a much higher success rate on attacks using small stacks. However, defending in stacks seems to carry a much higher risk of losing the whole stack. Has anybody seen that the AI is less likely to attack a stack than single ships, or does it only look to the total number of ships in the region? I know that it will definitely attack isolated single ships at a higher rate.

[Sensei Warrior]

MTW/VI 2.01 as well.

[Knight of the Temple]

Thanks for the pointers (and thread links) . . . although I'm not sure that I'm any clearer than I was before! The only thing that seems pretty clear is that, well, there isn't much clarity to be had in the naval aspects of MTW; from all the discussion and theories it seems it's in agreement that there is no agreement on what the best tactics are! I guess we all have to go on our own experiences and observations.

Anyway, I decided to risk some more of my caravels to try to clear the seas of my enemy's remaining ships, and thankfully I was far more successful and I did sink the last 2 barques in one on one battles with caravels, so I guess I must have been very unlucky the last time. It may just be my imagination, but I seem to fair better when I have additional ships coming into the region on the same turn where a naval battle is taking place.

[Deus ret.]

I seem to fair better when I have additional ships coming into the region on the same turn where a naval battle is taking place.

Definitely! I almost always attack that way and the funny thing is, you don't have to issue a separate attack order to involve the arriving ship in the fight. although when defending the advantage is not that obvious to me outcome-wise...

btw I disagree with the strongest-ship-on-top-of-the-matchup-theory. more often than not I don't lose my precious admiral (=the strongest ship) when taking casualties in a naval battle. it happens, but it doesn't have to be.

in addition, I also have no insight in CA's concept behind naval battle but I find the 1-on-1-matchup theory overall somewhat unconvincing since it massively favours the weaker, but more experienced competitor. ok whole stacks ARE sunk by single ships (not too often in my experience, though) but I never found single 4-star wargalleys to be as invincible as they should be after that theory. after all, how often is a lone super-admiral be capable of fending off 4+ attacking ships? and, in relation to that number, how often does a single defender sink 3 attackers headed by a 4-star admiral? considering general naval randomness not much more seldomly, I daresay.

.....I wonder, how long will the dephts and mysteries of this game continue to occupy us? if only the successor TWs were as elaborately crafted....

[Adrian II]

The only thing that seems pretty clear is that, well, there isn't much clarity to be had in the naval aspects of MTW; from all the discussion and theories it seems it's in agreement that there is no agreement on what the best tactics are!

It is even worse. There is a lingering suspicion that the outcome of your naval battles is somehow tied to the outcome of your (preceding) land battles. This certainly seems to be the case when the odds in a given naval battle are even, e.g. it's your one-star barque commander against a one-star enemy barque commander. If you win on land, you lose at sea, and vice versa.

[Martok]

Geez, Adrian, you just had to go and mention that to him, didn't you? No need to confuse the poor guy even further -- he probably would've been better off in blissful ignorance.

[Agent Miles]

I ran some tests Friday and finally got some partial stack kills. I did ten battles for each test. I attacked with three longboats (a 2* and two 1*’s) versus a 3* barque, a 1* barque and a 1* Dromon. First was a stack of three attackers vs. a stack of three defenders. I lost a total of 21 ships and sank a total of 21 ships. 40% of the time I won control of the sea zone. Then I did three singles vs. a stack of three ships. I lost 19 ships and sank 18, but I gained control of the sea zone half the time. Finally, I attacked with only one ship against the stack of three. I lost 8 ships and sank 11, but only won control of the sea twice. The defenders never tried to flea. This is hardly conclusive, but seems to indicate that neither large stacks nor singles have a clear advantage.
Still, IMHO.

[gunslinger]

It is even worse. There is a lingering suspicion that the outcome of your naval battles is somehow tied to the outcome of your (preceding) land battles. This certainly seems to be the case when the odds in a given naval battle are even, e.g. it's your one-star barque commander against a one-star enemy barque commander. If you win on land, you lose at sea, and vice versa.

I have seen evidence of this in my own games. I definitely think the naval battles are programmed to "even out" the results of land battles. For example, I entered a battle where the odds were overwhelmingly against me, and had one of those amazing victories where I won with about a 10 to 1 kill ratio. Then I had a naval matchup where I had a huge advantage, but lost a big stack of ships anyway. Due to a bad savegame file, I had to reload my game and redo the big land battle. Not wanting to go through it again, I auto-calced and lost. That time I won the big naval battle. I didn't want to give up my victory in the land battle, so I reloaded again, played the land battle manually, and won. Guess what. After winning the land battle, I lost horribly in the naval battle. I've only noticed this in conjunction with land battles I've won against overwhelming odds.

[Deus ret.]

I've re-discovered the thread where the precious bit of discussion was going on before being renewed in the "my spies watch movies" thread. It's here.

Moderators, can anyone merge the indicated thread with the relevant (i.e. the latter) part of the "my spies watch movies" thread and this one? One day it might even form a mini-guide to naval warfare....Thanks a lot!

The most important point here is that the Almohad Dhow attacked and in my experience that makes all the difference. The attacker always seems to have the advantage over the defender. I think this must be because the valour stars are interpreted differently in attack and defence situations.

Funny, we seem to have rather different perceptions of the game. Not that this would be uncommon with MTW...anyway, as long as I only have barques I almost never attack with them due to them being so bad at it, and I usually lose much more when attacking with them than when my stack of barques is attacked. I can assert that my 3-star admirals are doing admirably well in defending against the annoying byzantine fleets, almost all of whom do have 3 stars as well.

Has anyone oberserved whether ships gain stars after each victorious battle or only once in a while? It seems to me that not every battle increases their rating, yet I had the impression that they get stars faster than agents (assassins, for once) do, and for sure MUCH faster than a comparable general.

Maybe we should initiate a petition to CA to finally reveal the secret of naval battles

edit:

Those big multiple ship fleets, if attacked, can give the enemy a field day. Once they sink a few and the lead ship gains valour, the rest of the 0 valour ships in the stack will sink very easily.

Very well, this would mean that my stack of 4 v0 and 1 v3 ship which is attacked by a stack of 3 v0 and 1 v4 ship (of the same type) should perish quickly once their leading ship manages to sink mine. More often than not, I remember winning these kinds of battles even though I lost my leading ship - and it even was the only ship I lost. According to the stack leader vs. stack leader approach, this should be highly improbable, for once the admiral is sunk the remaining ships should be facing a more or less invincible valoured-up leading ship, making further casualties on the side of the latter very unlikely. At least in my experience, this isn't as uncommon as it should be. Admittedly I have no model to present on the way in which naval clashes work, but I can't help feeling sceptical about the rather simple way it's currently presented.

....does some of this make sense?

[Agent Miles]

I believe that there is only one promotion system and I have found this to be true for generals, inquisitors, and assassins. The promotion system goes like this: from no stars to one star takes one victory as does the progress from one star to two; the number of victories needed to progress then grows exponentially, i.e., two victories for three stars, four victories for four stars, eight for five stars, etc. I actually trained a general from zero stars to six stars, and an assassin and an inquisitor from zero to four stars to test this. I have never trained an admiral to four stars (as they die off more easily), but his progress to three followed the above pattern.

I cannot speak for Caravel, but in the test that I ran in the earlier thread you linked to, stacks had virtually no advantage over single ships in an attack. However, an entire stack of ships can be sunk by an AI stack, or even a single AI ship. For this reason, I don’t see an advantage in stacking.

Run some tests yourself.

[Martok]

Threads merged.

[Deus ret.]

Scenario 1: - The 'stack' player

Each stack leader certainly gains stars from a series of victories but (as you pointed out earlier) his subordinates never gain any stars of their own. Sooner or later, despite slim odds (c.f. assassin success percentages), an AI fleet gets lucky and takes out the admiral. In the next combat round, it is just a string of 'greenhorns' versus a fleet with a leader who has just gained some stars.

I moved the reply to this thread

In any case, I still don't get to terms with this. Mainly on the basis that there are quite a few battles (I'm a 'stack' player), command stars and number of ships roughly equal, in which I actually win while only losing my leading (admiral) ship. The same happens the other way round: More than once my fleet was sunk by an enemy stack which had a two-star leader after the battle instead of a three-star, and the AI hadn't simply moved the ship to an adjacent sea square. With the top ship vs. top ship approach, this would be highly unlikely, far too unlikely for the relative frequency of such events.

On the other hand it is also quite possible to beat 8 green enemy ships with three of your own lead by a 4-star ship, which would suggest something along the lines of the above approach.

I believe that there is only one promotion system and I have found this to be true for generals, inquisitors, and assassins. The promotion system goes like this: from no stars to one star takes one victory as does the progress from one star to two; the number of victories needed to progress then grows exponentially, i.e., two victories for three stars, four victories for four stars, eight for five stars, etc.

AFAIK this doesn't count for naval victories, I had the impression that the rate was faster on the seas, although I never saw an admiral gain more than one star in battle.

....maybe each enemy ship sunk counts as once victory, meaning that it's enough to sink 4 enemy ships to be instantly promoted from 3 to 4 stars?

*blink*

wait, that's a good hint! Caravel / EatYerGreens, you may well be right in that only the stack leader actually fights, given the rate at which he, and no ship besides him, gains valour! now that's something which convinces me...

so,
1. Ships are some kind of agents. Seems quite probable given the system of promotion, among other factors.
2. Only the leading ship of a stack fights versus the leading ship of another stack. Additional ships are treated as 'hit points' and only come into combar themselves once (and if) the leading ship should be randomly sunk.

Still, there is the issue I mentioned above. Can't really incorporate that....

[Voivode of Romania]

My naval battles seem to turn out for the best. I don't know if it's luck or not. I played the Russians once and had the largest navy. Even when i had 2 barques vs. a caravel I won. Once I defeated three dormons with a caravel and a barque.