Barrage vs City Raider tests

[vicawoo]

Null hypothesis: City raider promotions are always better for siege attacking cities
AKA the most hammer efficient medieval warfare test

In a minimalist worldbuilder file:
Reload random seed of course.
Your forces:
30 CR1 swords
10 CR2 swords
? C1 elephants
30 C2 elephants
30 CR1 catapults
10 CR2 catapults
30 Barrage 1 catapults
10 Barrage 2 catapults
Defending:
20 CG2 longbows in a city
1 CG1 longbow
1 catapult

The city is across a river to simulate 25% fortify. It has 0 defense to simulate bombardment.

If you want more CG1 longbows, just edit the file with a text editor, search for longbow, and copy from the beginunit to endunit position.
Ditto with the catapult. Note that catapults are valid collateral targets but take 0 damage, if I recall, so increasing catapults will reduce collateral effectiveness.

Test procedure, open to anyone:
Copy additional units as desired.
Record what units and strategy you will use (aka I will use 5 cr2 catapults, as many cr1 catapults as desired, 5 cr2 swords, and as many cr1 swords).
Record losses and damaged, by type, including promotions.

Run multiple tests

Edit: For some reason my CR1 catapults aren't getting the promotion...

Results:

Spoiler :

Pending

 

[NihilZero]

Nice experiment, I await the results witrh interest.

CR increases survival chances, but if I know a siege unit is going to die I'll promote it with barrage and do more collateral damage.

 

[DampRain]

I'm going on the assumption the point is you want to make it as easy as possible for your own forces to take a city with little no losses?

Anyway you've missed out a possible case set that is quite important,
a) when siege is about same strength as defenders
b) siege substantially weaker
c) siege substantially stronger

Although it could make testing significantly more difficult :\

Does this also mean you're going with the foregone conclusion if the opponent has lots of siege you're better off with no collateral? (which would make sense unless someone can correct me).

I'll be watching results with interest of course.

 

[Gresharas]

One thing you might want to consider:

Defender with City Garrison

Defender with Drill


Drill is the only true way to stop a siege-heavy SoD, it will be intresting simulate for single player pourpose.

 

[TheMeInTeam]

One thing you might want to consider:

Defender with City Garrison

Defender with Drill


Drill is the only true way to stop a siege-heavy SoD, it will be intresting simulate for single player pourpose.

There is no "true" way to stop a siege-heavy SoD other than using MGs while they're more advanced or to attack that stack before it attacks you.

 

[mirthadir]

There is no "true" way to stop a siege-heavy SoD other than using MGs while they're more advanced or to attack that stack before it attacks you.

Sure there are. First you can have defenders strong enough to win redlined vs any attacking units. Maces and LBs vs swords and pults is often this type of fight; even with heavy collateral damage the swords have extreme difficulty taking down the maces. Alternatively you can ensure that all the attacking units are also damaged; using air it is quite possible to have the attackers in the same general state of ill health as siege depleted defenders.

Secondly you can use non-MG siege for defense. Both cannons and arty make reasonably decent defenders if you have a tech lead (cannons vs muskets/knights and arty vs riflemen). Suiciding cats and trebs against drill cannons is pretty much futile, cannons against arty is also a lousy trade.

Thirdly, DIV units are good if you can get enough of them, particularly with hills. For instance I could be wrong but I believe it is more to attack SB's DIV/CGI archers with flanking HA than pults.

 

[Gresharas]

There is no "true" way to stop a siege-heavy SoD other than using MGs while they're more advanced or to attack that stack before it attacks you.

I totally agree but, to name one nuisance at random, Shaka SoD pre-Mgs is a fair sight if you hadn't squished him earlier, and a 100+ unit SoD cannot be killed in the field, either because you where looking to an offence [so your troops are city busters] or on defense [and they are city-defenders].

Or, that might be me to follow the wrong strategy. You play higher difficulty than me so: do you keep many "field intercept" troops in times of peace?

I hate defense in mediaval warfare. The best I could do is to withdraw, lure them in one of my flat cities, and then city-raid miself a couple of times.


@mithrandir: I never take into account to be on a tech lead in this kind of questions. As a warmonger, if I am on the lead then I'm also on the offence and the game is pretty much to an end.

 

[TheMeInTeam]

Alternatively you can ensure that all the attacking units are also damaged; using air it is quite possible to have the attackers in the same general state of ill health as siege depleted defenders.

I do believe this constitutes "attacking the stack before it attacks you" .

I guess I overlooked a monstrous tech lead, I rarely consider war a threat if my tech position is so good that I can win with redlined units. I disagree on longbow/mace though...I've been on the sword attacker end of that, and the conclusion is that the maces do NOT want to wait to get raped down to <6 str by siege and then picked off by a >50% city attack sword (CR II and base ability). It wasn't a fun war, but I won it and the game (it was against the only real war threat I had).

I guess if you went hammer for hammer into nothing but longbow spam and stacked it up into the city, you could probably hold cats off reasonably. I'm not convinced this is a good strategy against trebs w/o hills though, as without protective CR II trebs won't have terrible odds there with even a tiny bit of damage.

The non-MG siege is a good point, but then you get nonsense like 1-3 cats hit to weaken pikes, and then you get spanked by combat II+ knights or cuirassers and the cannon's hammer trade is suddenly less favorable. Same thing for cavalry, though at least they're not inherently immune to 1st strikes.

And if you have that kind of siege, wouldn't one be tempted just to attack with it anyway? Unless you allowed the attacker onto some heavy defensive terrain, the odds won't be much different and you get collateral out of the deal.

 

[henrebotha]

This is relevant to my interests. I can't help with the testing, but I look forward to the conclusions drawn by some of the more experienced players. (Someone should tell PieceOfMind about this thread... )

 

[mirthadir]

I do believe this constitutes "attacking the stack before it attacks you" .

I guess I overlooked a monstrous tech lead, I rarely consider war a threat if my tech position is so good that I can win with redlined units. I disagree on longbow/mace though...I've been on the sword attacker end of that, and the conclusion is that the maces do NOT want to wait to get raped down to <6 str by siege and then picked off by a >50% city attack sword (CR II and base ability). It wasn't a fun war, but I won it and the game (it was against the only real war threat I had).

I guess if you went hammer for hammer into nothing but longbow spam and stacked it up into the city, you could probably hold cats off reasonably. I'm not convinced this is a good strategy against trebs w/o hills though, as without protective CR II trebs won't have terrible odds there with even a tiny bit of damage.

The non-MG siege is a good point, but then you get nonsense like 1-3 cats hit to weaken pikes, and then you get spanked by combat II+ knights or cuirassers and the cannon's hammer trade is suddenly less favorable. Same thing for cavalry, though at least they're not inherently immune to 1st strikes.

And if you have that kind of siege, wouldn't one be tempted just to attack with it anyway? Unless you allowed the attacker onto some heavy defensive terrain, the odds won't be much different and you get collateral out of the deal.

Swords vs redlined maces depends upon the promos being used. One promo swords (an AI hitting before they get theo and no earlier war) have trouble with maces; shock maces on a hill are simply that good. Remember this is not going to be an intelligently promoting human (charge anyone?), but an AI.

Using cannons and particularly arty as defense depends on how many moves I have left. I often try to stack units at central locations so I can use interior lines to bulk up defenses (or make an attack of opportunity in either direction). Not having the 1 move point more required to attack out of the city rather than just gaurd the city is reasonably common. Also the unit that gets completed the turn of the attack and air drops never get to get out of the city. An additional concern would be rivers. Collateral vs a 50% differential often makes the latter half a better option. Lastly, it comes down stack composition; if the AI has gone overboard with siege I'm most likely going to do no real collateral so that is a wash. If the AI has a few super units that I figure are going to kill my siege anyways (e.g. a GG lead C mace) I'd rather not double their attrition rate.

Most of the time it is a question of reinforcements; attacking piecemeal can often be far worse than enduring your licks by waiting for a turn (when the AI will ineffectually attack) and counter attacking en masse afterwards. Getting backstabbed can often require me to air/sea chain redeploy defenders in mass; arty own for that (and are often the ground unit I'm spamming anyways). Having shaka waltz across the border with mass cannon is not so scary if I can move or airlift in eight arty each turn.

Now, if the situation is normal, it is normally far better to flank away the AI siege, use siege initiative yourself, etc. but for those rare cases where you don't get to plan a specific defense, masses of arty or overpowered defenders both can defeat mass siege in a pinch. If I'm building a defensive stack it is a far different thing than "crap, Cathy just got bribed to declare so let me mass upgrade my HR catapults."