• We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.
I find part of the problem has to do with nation sizes. Leaders with martial under 10, or with bad traits like craven, are much more prone to make major mistakes choosing terrible combat options, or get hesitant commander. But once you get limited crown authority to easily choose commanders, and a medium sized nation, it's easy to ensure that all your commanders are decent.
Super commanders with 20 martial and good traits for it don't seem that powerful, they're somewaht better, but not dramatically. Not like what a 6/6/6 leader can do in EU.
Partly this is because there isn't a stacking limit on combat; again like in eu3 where even if outnumbered, they only get 4 extra units of cav hitting you (and cannons, if any); without that superior numbers get to freely use their advantage, even in terrains like mountains where chokepoints are common.
 
ok... how about this then? There's not enough variance of the probable and improbable effects. When I was playing EU3 and I held Venice, I could sometimes hold off armies 4-5x my size simply by holding them off at Treviso. If I rolled a 9 and the enemy rolled a 1, on top of my mountain bonus this would decimate the enemy. In this game that would never happen because picking the 3 blues in your example hardly do anything different than if you picked the 7 reds. There is hardly any outliers in outcome.

As we can say if we plotted the occurrences of a battle 9000 troops vs 12000 on plains there'd be no deviation whatsoever. The 12k troops win every time, and casualties always range around 5000 for the bigger side and 8000-9000 for the defender. If the battle was between 15000 to 8000, the bigger force instantly routes a flank, the battle's outcome is set in stone, and the attacker loses 6000 of its troops while the bigger force loses maybe 1500. Wheras if we plotted what happens when Austria-Hungry invades my troops on Treviso with 4-1 advantage I'd win a heroic victory about 1/10 of the time with few casualties, get absolutely slaugtered 1/3 the time, lose while still inflicting mass casualties on the enemy another third of the time, and win by routing the enemy but still losing almost all of my troops the other third of the time. It sounds like its not that you don't want the game to not be random (for some reason you insist on calling it probable or improbable) you just want the different outcomes of each trial to hardly deviate from the norm, low standard deviation so to say.
 
If you're not trolling, Prime, then you are deluded/stupid. This has no relation to the discussion whatsoever. Random factor is present in battles. Save/reload and result will be different. Randomness is what causes this. Just because there is no -3 displayed does not mean dice are not being rolled.

People are dissatisfied there are no ways for small armies to win vs large armies. Adding dice that changes every 5 days won't fix this: depending on how much influence dice has, either battles will still be numbers dependent, or they will be just random with 3k obliterating 9k a common business. And then the next month these 3k fall to 500.
 
People are dissatisfied there are no ways for small armies to win vs large armies. Adding dice that changes every 5 days won't fix this: depending on how much influence dice has, either battles will still be numbers dependent, or they will be just random with 3k obliterating 9k a common business. And then the next month these 3k fall to 500.
As I already said - great nations should be great because they can afford to lose a lot of battles, not because they win (almost) every one of them.
If more randomness is introduced, the small nation may win a few battles, but in a long term it would still be losing it.
As I see it, there is a certain problem with predictable battles - it promotes doomstacking.
Even AI does doomstack and then proceeds to ravage it's enemy until (if ever) it can assemble even larger stack.
 
There's already a random element in combat, it determines what kind of troop will do damage today.

Besides, luck only occurs in history because of the lack of information (commander skills, enemy position, ambush), which will never happen in this game.

I'm all for random events as long as it's not something ridiculous like 1k army shrugs off the 10k, and the damage value depends on the army size of the triggerer, not the triggeree. (a 200 garrison village with 10 martial defender manages to assault a 10k army 20 martial and deals 5% -> 500 dmg? really?)
 
Besides, luck only occurs in history because of the lack of information (commander skills, enemy position, ambush), which will never happen in this game.
Really?
So your commanders know exactly which commanders lead the enemy army and the local terrain?
Why should it be so?
 
Because it'll require rewriting the entire combat and stat system if you want to add in the scouting element.

Because it's a game that's focused more on kingdom development and intrigues. (it's pretty clear when you realized you're not allowed to take down enemy transports, or the non-existence of naval battles)

Seriously, does anyone really want to try and random number until you can finally know how good your vassals are at this particular job years after he's instated?

And to my knowledge, martial does or should include the commanding ability to scout enemy territories and enemy armies. So if the armies are evenly matched on commander skill and troop training (kind of wish we have this too, but oh well), it's still going to come down to numbers.

Though to be fair, if both forces doomstack to the point where the entire province should've already been covered in people and dead bodies, what's there left to scout?

With that said, I'm all for at least adding events that remove or reverse terrain advantage/disadvantage through random numbers or reinforcements, or events that deal extra damage but doesn't completely determine the outcome of the battle.
 
There's already a random element in combat, it determines what kind of troop will do damage today.

Besides, luck only occurs in history because of the lack of information (commander skills, enemy position, ambush), which will never happen in this game.

I'm all for random events as long as it's not something ridiculous like 1k army shrugs off the 10k, and the damage value depends on the army size of the triggerer, not the triggeree. (a 200 garrison village with 10 martial defender manages to assault a 10k army 20 martial and deals 5% -> 500 dmg? really?)

Exactly my point. Thank you cywang86. :)
 
Think of this, a medieval, (and thus, more applicable), example of how random factors have granted a numerically inferior army victory; Agincourt. Would we ever see an Agincourt in CKII as it is? Probably not. Did a real Agincourt occur, and give the small, tired, inferior English army victory? Yes, and for this reason I think luck should be more prominent.

But the real reason why Agincourt happened as it did had very little to do with luck (unless you count England warring a deeply divided France lucky), the disparity in quality of command, and the unity of that command had a lot more to do with it. Simply put the English command had the ability to know what they needed to do, and the confidence of their staff and soldiers to be able to ensure this be carried out. The French command was deeply divided and badly lead. Many of the French commanders were actually more interested in winning personal honour than ensuring a French victory. They were also overconfident and thought that the first charge of their knights would carry the field against a peasant army.
The other big factor was terrain. The fact that the French had to cross a muddy defile which had recently been churned up by ploughs was probably the single biggest factor in the English victory. It protected their longbows, and gave them time to shoot multiple volleys into a densely packed and vulnerable mass of men-at-arms and knights. This factor was again due to the fact of good generalship, Henry having picked out the spot to give battle to the French, noting its obvious advantages if he could persuade them to attack his defensive positions.

Yes, random elements do sometimes affect a battle, such as a veteran unit inexplicably breaking just as victory seems certain or a bunch of green conscripts holding out far longer than anyone would expect. But the fact is that numbers, generalship, training, equipment and morale are far greater determinants of the outcome of a battle than any random luck factor, and these are all items which can be controlled or influenced in a deliberate manner.
 
Random elements do occur in real combat that could not possibly be simulated by a computer game. I understand that the obvious factors, (numbers, troop quality, terrain, leader, etc.), are included in the game, even if only superficially, but there are still anomalies that should be represented IMO. Take Washington's victory at Trenton, for instance. Washington's exhausted and hungry men who had spent the winter at Valley Forge were up against a fresh, well equipped, numerically superior army, yet they won due to variables that a computer game could never have accounted for. These variables; a fog settling over the patriots as they crossed the Delaware River, then leaving shortly after they got off of their boats, the Hessian mercenaries who were hung over due to the previous day being Christmas Eve, and routed. It is to these sorts of things, (which I'm sure history provides more example of), that I would attribute luck to. In this way, I think that luck should play a more prominent role in battles.

Think of this, a medieval, (and thus, more applicable), example of how random factors have granted a numerically inferior army victory; Agincourt. Would we ever see an Agincourt in CKII as it is? Probably not. Did a real Agincourt occur, and give the small, tired, inferior English army victory? Yes, and for this reason I think luck should be more prominent.
There was nothing "random" about the French defeat at Agincourt. A simple description of the battle makes it perfectly obvious why the French lost:

The French chose, after recent heavy rain, to launch a frontal assault through recently-ploughed fields (i.e. a sea of mud) with steep slopes on either side towards a prepared defensive position well-provided with archers.

This would not be the last time a French military commander underestimated the devastating disruptive power of mud, of course.

Edit: Bah. Emu'd.
 
Last edited:
Methinks most of you are talking about two different things. Some are arguing against randomness because the real-life battles listed as examples of how decisive randomness can be were not actually all that random (Agincourt being the main example discussed). On the other side are people who are arguing for the system to account for more of the variables that are not accounted for and calling it "randomness" for lack of a better term. But I think it's the term itself that is throwing off the discussion.

There are any number of variables that effect the outcomes of a battle that are not actually random in the strictest sense of the term, but are still highly important to the outcome and are almost completely unknown and/or out of the hands of the battlefield commanders themselves. Examples range anywhere from obvious things like the quality of lower-level leadership like sergeants and corporals or how well-rested the armies were (i.e. have they been lying in wait or on a forced march?) or the army's veterancy, to less obvious things like the quality of the ore that was used to get the metal used in weapons and armor or the availability of grazing land for the oxen carrying soldiers and equipment or the presence of enough blacksmiths to re-shoe horses, and potentially to things that may be completely unrelated like what the soldiers had for breakfast that day. I suppose it would be possible for Paradox to add these multiple layers to the game in the form of more lists in the ledger. And if they wanted to be even crazier they could allow the player to directly control these variables. But each step the developers take in that direction makes the game less and less about managing a dynasty and more and more about micromanaging armies, something that is highly unrealistic, immersion-breaking, and (in my humble opinion) not fun at all. Much better to attempt to account for all of these important yet uncontrollable variables than to completely ignore them. I don't know enough about computer programming to offer a better alternative, so I'll just say that some sort of dice roll is necessary, and that "randomness" in this case is really just shorthand for "all of the important stuff the game doesn't account for directly but still ought to be included".

If one is worried that this will make the outcome of battles too unpredictable, I have another suggestion (again, coming from a guy who understands statistics pretty well but knows absolutely nothing about programming): more than one dice roll related to "randomness". It might be that the massive and overconfident army is suddenly faced with a perfect storm of bad variables (e.g. the veteran sergeants all ate at the same tavern where they got food poisoning, while the oxen pulling the wagons with all of the gear all caught some kind of weird disease, on the same day that a panic spread through the ranks about ghosts or some other such medieval nonsense), but, through even more "randomness", this ends up not effecting the outcome of the battle all that much.

Or, for people who are more numerically-minded:
Assume two regular six-sided dice. Roll them to see the initial randomness variable. A result of 6 or 7 would provide a 1 for the "randomness" variable (i.e. no net benefit or loss). This would sufficiently simulate some of the randomness, but it would not really simulate the aforementioned scenario where all of the bad luck amounted to little effect. To do that, another two six-sided dice are rolled to either amplify or attenuate the effect of the initial "randomness" variable. This would simulate how strongly the "randomness" could potentially effect the outcome. If you wanted to make it even more complex, you could add another variable based on commanders' skills and how well they exploited or made up for these "random" factors. But basically, I think a system like this (when combined with the existing system or something closely resembling it) would make massive upsets possible while also making sure they're about as likely as a royal flush in poker. Then you have something pretty close to real life where battles were largely predictable though not as predictable as they are in the game presently, while also keeping what creates the unpredictability out of the hands of the player/AI.
 
You seem to be misunderstanding. I'll try to explain my Agincourt example, and how it applies to this game, with the utmost brevity and clarity:

The French lost Agincourt for reasons, yes; choosing to charge over sloppy mud against an encouraged band of English longbow men hidden behind stakes, or what have you. Know that I understand that all outcomes in combat occur for reasons, as do all things in reality, but in a computer game, all of these reasons cannot be accounted for. In Crusader Kings 2, a game which seeks to be relatively realistic, these 'reasons' for oddities in combat that could affect the outcome, which exist beyond the programming of the game, should still be included, and included as luck in one form or another. These reasons should be included because this is a game that seeks to resemble history, and history is so full of these elusive, sometimes seemingly inexplicable oddities that the game could never include in a mathematical or theorized sense, that it must add (in its quest for historical plausibility), them in as a slight bit of luck atop the more important values that are already programmed.

I didn't say that as clearly as I could have, (there is a lot of noise and distraction around me right now), but surely anyone reading this understands what I'm saying.
 
You seem to be misunderstanding. I'll try to explain my Agincourt example, and how it applies to this game, with the utmost brevity and clarity:

The French lost Agincourt for reasons, yes; choosing to charge over sloppy mud against an encouraged band of English longbow men hidden behind stakes, or what have you. Know that I understand that all outcomes in combat occur for reasons, as do all things in reality, but in a computer game, all of these reasons cannot be accounted for. In Crusader Kings 2, a game which seeks to be relatively realistic, these 'reasons' for oddities in combat that could affect the outcome, which exist beyond the programming of the game, should still be included, and included as luck in one form or another. These reasons should be included because this is a game that seeks to resemble history, and history is so full of these elusive, sometimes seemingly inexplicable oddities that the game could never include in a mathematical or theorized sense, that it must add (in its quest for historical plausibility), them in as a slight bit of luck atop the more important values that are already programmed.

I didn't say that as clearly as I could have, (there is a lot of noise and distraction around me right now), but surely anyone reading this understands what I'm saying.

Basically what I said, just that you did it with more brevity and clarity. :)
 
I have never lost a battle in which I outnumbered the enemy by more than 10%, and I have never won a battle in which the enemy outnumbered me by more than 10%. The system needs fixing.
 
Didn't read through the entire thread, so possible someone else mentioned something along these lines...

The Art of Defense just seems vastly underpowered in this game. The only time I can really tell much of a difference is when an army crosses a river to attack, in which case it is possible to win as a slightly outnumbered defender, but in pretty much every other case, forget it. This just isn't reflective of warfare in any time period, particularly this one. Someone mentioned the old "spearman takes tank" problem from Civ 4, due to RNG, but those of us who have played Civ 5 know how hard it is to take a smaller nation that uses the terrain and the proper doctrines to defend its territory. That simply isn't reflected here. You'll never see the Byzantine Empire in danger from Sicily in 1085, as they were historically. Byzantium's armies simply could not handle the Normans in the field, and had to resort to bribing the HRE to attack Sicily and force them to withdraw their troops. Small nations such as Georgia and Alania are easily overwhelmed by Muslims. It just isn't possible for an outnumbered nation to defend itself against someone that has more troops. That just doesn't make for a very satisfactory game. I've tried so many times to play as the Kingdom of Jerusalem starting in 1099 but have concluded that it is simply impossible in the current game system.
 
Either you havent done that many battles or you are "pulling" numbers out of your memory with little or no fact behind them.

I have seem pletny of battles where the smaller army wins. I mostly fight christians vs muslims and there are alot of factors that can lead to this. Muslim armies with big decadence for example get trampled by high martial christian armies.
 
EDIT: And how does a dice roll make it more realistic? Physics is not random at all. Everything can be predicted.

At a level of abstraction that basically says:

"A 'battle' stands for one or a series of battles that may take place at some point over the shown period of time in-game at some place somewhere in a province spanning a couple hundred square miles."

I think you need to abstract "physics" and other chaos factors a bit.