Would you abandon this siege?

[YuriiH]

What I've understood is that high cavalry armies are not efficient in the early game.

First of all, efficient in what way? money? time? manpower? wining battles? vs other country?

Moreover, this all depends on your techgroup and game years.
E.g., At Military Tech 6-9, Morocco (Muslim) has SIX pips in cavalry, while Castile (Western) has only THREE.
Moreover, at Miltech 5-8, Morocco's infantry has FOUR pips vs. THREE pips in Castile's units.
Logically, Morocco's units will be more efficient than Castile's at certain points of the game.

Plus cavalry has a bunch of issues (combat position, cost, replenishment, attrition, etc.) that make horses less desirable than a no-brainer solution for the same price: mercs.

 

[Kergan]

What I've understood is that high cavalry armies are not efficient in the early game. Sure, cavalry - because of their shock benefit - rule the battlefield, but primarily as a flanking unit. This has been reinforced by a couple of reddit spreadsheets on army composition, suggesting no more than two or at the most four units of cavalry per stack of 12 to 14.

My army composition, during the early game when shock heavily impacts most battles, is something like 8 to 10 infantry with 2 to 4 cavalry (in stacks of 10 to 14). Even with that, I feel like I'm building more cavalry than I should. I used to build high cavalry armies (I like cavalry ) but people told me I was "doing it wrong."

So is this "internet wisdom" wrong? I'm happy to build standing armies with more cavalry if it's worth the money and it really does make a difference in battles. Another thing I was told by the player who taught me to play the game: (1) cavalry adds no benefit to a siege and (2) cavalry does not perform well in mountain provinces. Is this incorrect, because it's been driving a lot of decisions I make.

There's a cavalry sweet spot in the very early game where it's really really powerful if you can afford it. The thing is flanking. In a 22 vs 18 battle, with 5 cavalry you've 2 horses on each side flanking one or two units without taking damage on each side unless they were flanking horses, and then doing nothing, plus one horse towards the very end of one side bashing three infantry while taking damage. Morocco, by contrast, likely had 6-8 cavalry, each dealing damage to 5 of your units at the time instead of 3 for infantry, and Muslims have much better horses than Western countries in the early game.

The usefulness of cavalry progressively goes downhill until Schwarter Reiter (tech 10 if memory serves), to the point of being borderline useless IMO except for flanking purposes at one point between when you get Condottieri units and when you get Schwater Reiters (infantry units basically fight well enough that it's not worth the extra ducats to get the basic western infantry).

In the mid game there's another sweet spot when enemies don't systematically have full combat width or an inferior one. It's great to throw in 2 or 3 horses on each side (depending on how good you're at flanking). Or more, for that matter, if you can afford it: the shock damage they deal is monstrous (always pick the horse that maxes it when given a choice), and don't forget that if a horse has 5 units (7 later on) in front of it, it'll happily bash all 5 (7) at each combat phase. But keep in mind that other nations have better horses than you do.

Late game I still build them in SP games - like 6-8 per unit, sometimes as many as 12 for the same reason. Even in a full width vs full width battle with a full back row of canons, the western horses are better than other horses at this point, and will massacre infantry because of this flanking damage insofar as I've experimented. But there's something to be said about the convenience of having 40/0/40 stacks all over the place to split combine them as needed, especially if the infantry units are all mercenaries.

 

[rinehime]

and don't forget that if a horse has 5 units (7 later on) in front of it, it'll happily bash all 5 (7) at each combat phase.

Really? In that case infantry should be attacking 3 units post tech 23 as well... I know the UI shows all the units that a unit can target, but I assumed all casualties were dealt to only a single unit on the opposing side. Are you saying each unit targeted by cav gets all C (as defined in the wiki) casualties? Or are the total casualties C divided between the units? Only the later would make sense. The former would make cav ridiculously powerful (as in worth 3/5/7 times a single unit) and I imagine people would be using them more often...

 

[IrishGirl]

First of all, efficient in what way? money? time? manpower? wining battles? vs other country?

I'll answer your question this way, Yuri. I'll post a link to the spreadsheet that I've been sent twice now.

https://docs.google.com/spreadsheet...5LEKiON0k2rZJ82YbYaB0/htmlview?sle=true#gid=0

 

[IrishGirl]

And I'll just add this: the author does make exceptions for (1) playing as Poland - which I wasn't - and (2) facing hordes.

I know that, like most recommendations, it needs to be evaluated against experience. I just want you to know that (1) I've tried to do the best research I can and (2) I'm not just pulling information out of the air. (Instead, I'm just pulling information off the Internet - which is only slightly better, I suppose.)

 

[rinehime]

Cavalry is quite strong early game because of the difference in shock vs fire modifiers (and unit pips). The reason most people say they don't use cav is because they're also quite expensive and, if recruited as regulars, cost manpower so it's a cost-benefit analysis.

Yuri also hit on some other points above, like the difference in units between Western / Muslim tech groups, but I think these misconceptions should be addressed:

(1) cavalry adds no benefit to a siege and (2) cavalry does not perform well in mountain provinces. Is this incorrect, because it's been driving a lot of decisions I make.

1) Cav is just as useful for sieging as infantry. It counts toward the total units needed to progress a siege [NB: I can't remember if this was always the case, maybe it wasn't in old patches or EU3?], but doesn't give any extra benefits like arty does. Cav is not used if you decided to assault, however.

2) If you're not a horde, there's no special penalty for cav due to terrain. The die penalty is the same for the whole army. If you are a horde, there's a -25% shock damage penalty in non-flat terrain, and a +25% shock damage bonus in flat terrain. The "bad in mountains" advice might also be due to older patches where combat width was terrain dependent. Mountains used to have reduced combat width, making it harder to get the flanking advantage from cav.

 

[Kergan]

Really? In that case infantry should be attacking 3 units post tech 23 as well... I know the UI shows all the units that a unit can target, but I assumed all casualties were dealt to only a single unit on the opposing side. Are you saying each unit targeted by cav gets all C (as defined in the wiki) casualties? Or are the total casualties C divided between the units? Only the later would make sense. The former would make cav ridiculously powerful (as in worth 3/5/7 times a single unit) and I imagine people would be using them more often...

Insofar as I'm aware, yes, they trash everything they can hit. And yes, cavalry is very powerful because of this. The problem is manpower and how they get wrecked when they've many units facing them. (Which is why in mid- to late-game I used to like to hire merc cavalry. With professionalism, I'm not too sure yet.)

 

[IrishGirl]

I think these misconceptions should be addressed:

1) Cav is just as useful for sieging as infantry. It counts toward the total units needed to progress a siege [NB: I can't remember if this was always the case, maybe it wasn't in old patches or EU3?], but doesn't give any extra benefits like arty does. Cav is not used if you decided to assault, however.

2) If you're not a horde, there's no special penalty for cav due to terrain. The die penalty is the same for the whole army. If you are a horde, there's a -25% shock damage penalty in non-flat terrain, and a +25% shock damage bonus in flat terrain. The "bad in mountains" advice might also be due to older patches where combat width was terrain dependent. Mountains used to have reduced combat width, making it harder to get the flanking advantage from cav.

This makes sense, as I received this information well over a year ago (and maybe shortly after EU4 was released). Thanks for the information.

 

[True_Heir_of_Rome]

One of the issues with using Cav to siege is they cost more to reinforce than infantry. I don't know about your sieges, but I roll five 1s(disease outbreak) for every other number combined, so my units spend a lot of time in hostile territory slowly replenishing troops.

They still do function perfectly fine in general though, but using infantry and artillery is the way to go when possible.

 

[rinehime]

Insofar as I'm aware, yes, they trash everything they can hit,

Ok, I had to test this b/c it would make cav seriously OP.
TL;DR: cavalry only attack one unit at a time, they don't attack all of the units they have the range for.

Here's a battle a few days in (started as France, rivaled then attacked Burgandy), on the last Fire tick before the beginning of the shock ticks. I was trying to get a 1-on-1 battle (10-4 vs 10-4), but one of BUR's subjects hopped into the middle of the fight...

Note the BUR cav has 1000 men before the shock phase (Cav fire modifier @ tech 3 =0):

Spoiler: Jan 1, 1000 men

Now, on the next day, we have a die=9, terrain=-1, attack_pips=1, defense_pips=0 for a base roll 9. They took a total of 124 casualties (1000 to 876)

Spoiler: Jan 2: 876 men

From the wiki:

C_base = 15 + base_roll*5 = 60
Us_atk =1.000 (full strength)
Um_atk = 1.0 (Cav shock, lvl 3)
No combat ability bonus
disc/tactics_def = 1.00 / 0.5 = 2
Casualties = 60*0.938*1.000*1.0*1.00 /0.5 = 120

Now, 120 is not exactly 124, but it's close enough for paradox math .

On the other hand, if my 2 cav were attacking all units they could at once, then the damage to the BUR cav would be doubled, and that's not what happened, so I think it's fair to say that they only target one unit at a time.
EDIT: Now, maybe, just maybe, they're something going on with the total casualties being divided amongst the units they can attack, but I don't think that's the case... I don't think it would explain the discrepancy either, as my units were all near full strength.

 

[YuriiH]

I'll answer your question this way, Yuri. I'll post a link to the spreadsheet that I've been sent twice now.

Well… the spreadsheet is just awful:
1) it does not calculate anything,
2) it simply states that 4 cav is enough (without clear and calculated reasoning why),
3) it refers to Zwirbaum's guide which tells only about the units and their types, and does not say anything about composition of 4 cav max

Certainly, this "spreadsheet" may be used as reference to units, but not as an evidence that 4 cav is enough early on.

 

[True_Heir_of_Rome]

I'm near certain that units can only attack one unit at a time. Cav attack the nearest unit to them, then move in as units retreat.

 

[Shinkuro Yukinari]

The answer is simple. Use the Backup exploit and crash the game if it goes badly!

 

[rinehime]

I'm near certain that units can only attack one unit at a time. Cav attack the nearest unit to them, then move in as units retreat.

Yeah same here, I just had to test it. There was that bug with infantry attacking from the backline that just got fixed in 1.23, so sometimes things aren't always so "certain".