Counter Intuitive Mechanics - bulding Light Cavalry will make you lose fights
- Thread starter Avisian
- Start date
-
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.
Every single trooptype is better then light cavalry (even Light Infantry due to superior numbers)
The only real use I see for light cav, especially retinues, is for a mop up unit that is significantly smaller than the main army which has as few light cav as possible. You throw your two units together for the initial clash of doomstacks, then the light cav unit is the one that does the chasing while the rest of the army sieges.
This is far too specialized in my opinion. Perhaps if there is a tactic for light cav to disengage from the melee and put their flank back into skirmish, or, more interestingly, an ability of light cav and horse archers to change which flank they're on during combat. Instead of breaking from the battle when their flank collapses, they should instead switch to a different flank, and they should switch to a struggling flank if their flank is winning. This would simulate the increased maneuverability and battlefield role of lightly armored cavalry.
This is far too specialized in my opinion. Perhaps if there is a tactic for light cav to disengage from the melee and put their flank back into skirmish, or, more interestingly, an ability of light cav and horse archers to change which flank they're on during combat. Instead of breaking from the battle when their flank collapses, they should instead switch to a different flank, and they should switch to a struggling flank if their flank is winning. This would simulate the increased maneuverability and battlefield role of lightly armored cavalry.
This is a very interesting thread. I'm not sure what I think. Someone would have to do some controlled experiments to be sure. My instinct is that a larger army with light cav in it will outperform a smaller one without, despite the big difference in tactics. But I really don't know.
A few things to add however:
1. The problem seems to be that the skirmish tactics that benefit archers hurt light cav, and vice versa. That's really an annoying design. It would be much better if they just boosted one or the other. But damaging the other one at the same time? That creates really bad choices for the player. This issue only exits because of that ill considered design choice. Archers and light cav are the two main skirmish units. "Good" Skirmish tactics shouldn't harm either one of them. Only "bad" tactics from poor generals.
2. For retinues, I really like light cav. I don't build archers at all in retinues. I stack up on melee units, and toss in a good number of light cav to mop up in pursue. The melee units then pick defensive tactics in skirmish (shieldwall), take very little damage, and then destroy the enemy in melee and pursue. So light cav is still very useful if you avoid archers, which is a very valid tactic if you're not grouped with levies.
3. I'd really like to understand better how the bonuses work for cultural units. It really seems to me that Spanish light cav might indeed be trash. That would be unfortunate. All the other cultural units are very useful. It's disappointing if playing in Spain makes you gimped, but that may indeed be true.
A few things to add however:
1. The problem seems to be that the skirmish tactics that benefit archers hurt light cav, and vice versa. That's really an annoying design. It would be much better if they just boosted one or the other. But damaging the other one at the same time? That creates really bad choices for the player. This issue only exits because of that ill considered design choice. Archers and light cav are the two main skirmish units. "Good" Skirmish tactics shouldn't harm either one of them. Only "bad" tactics from poor generals.
2. For retinues, I really like light cav. I don't build archers at all in retinues. I stack up on melee units, and toss in a good number of light cav to mop up in pursue. The melee units then pick defensive tactics in skirmish (shieldwall), take very little damage, and then destroy the enemy in melee and pursue. So light cav is still very useful if you avoid archers, which is a very valid tactic if you're not grouped with levies.
3. I'd really like to understand better how the bonuses work for cultural units. It really seems to me that Spanish light cav might indeed be trash. That would be unfortunate. All the other cultural units are very useful. It's disappointing if playing in Spain makes you gimped, but that may indeed be true.
First I must say I don't know how each retinue type stacks up against other retinues. Maybe LC is weak, I wouldn't know.
That said... many things in this thread are wrong. Even if the premise was right, the calculations needed to draw conclusions would be way harder and better thought out than what we have here. Another thing, it's not like LC is some special unit screwing others... what about pikemen/heavy infranty? The levies are basically bunch of guys stepping on each others dicks but I'm totally fine with that. Also you guys are confusing swarm tactic with harass tactic, swarm is for HA not LC. Not a big deal but I'm gonna mention harass tactic next so don't get confused.
The wrong premise I'm talking about is that more troops of certain kind increases the odds of using certain tactics.
Here is what I did: I put one retinue of 400LC/100HC with 10000 HI. The odds of landing harass tactic was 45.7%. and shieldwall was only 40%. Does that seem right 400 LC vs 10000 HI? Well I added another LC retinue and the odds were still 45.7%. With only HI there is no harass tactic and shieldwall is 75%.
It seems the amount doesn't matter at all as long as you have at least some.
Conclusion:
1. Think about if you want to put mixed units in the same flank in your retinues.
2. Levies are a mess but I think it's fine and probably more realistic than single unit type retinue flanks. Unupgraded Castle has LC in it and you will end up with a mix anyway so adding more of unit X shouldn't screw you. So you shouldn't worry because you are already screwed.
Testing above and some other things I learned that Housecarl retinue vs random ilkhanate levy stack is pretty badass. Assault from sea on to mountains gave me two penalties but 30k vs 30k and the Housecarls won 90%++ average 10k lost 20k++ killed. Only loss was when they did heroic counter charge with two of their flanks. They had better generals too.
One final thing, am I supposed to have a typo in my file for the tactics? Does it read volly_tactic or have I accidentally erased a letter and saved? Might be a typo because I also noticed that powerful charge in game is called awesome charge in the file... but I'm paranoid now! :O
That said... many things in this thread are wrong. Even if the premise was right, the calculations needed to draw conclusions would be way harder and better thought out than what we have here. Another thing, it's not like LC is some special unit screwing others... what about pikemen/heavy infranty? The levies are basically bunch of guys stepping on each others dicks but I'm totally fine with that. Also you guys are confusing swarm tactic with harass tactic, swarm is for HA not LC. Not a big deal but I'm gonna mention harass tactic next so don't get confused.
The wrong premise I'm talking about is that more troops of certain kind increases the odds of using certain tactics.
Here is what I did: I put one retinue of 400LC/100HC with 10000 HI. The odds of landing harass tactic was 45.7%. and shieldwall was only 40%. Does that seem right 400 LC vs 10000 HI? Well I added another LC retinue and the odds were still 45.7%. With only HI there is no harass tactic and shieldwall is 75%.
It seems the amount doesn't matter at all as long as you have at least some.
Conclusion:
1. Think about if you want to put mixed units in the same flank in your retinues.
2. Levies are a mess but I think it's fine and probably more realistic than single unit type retinue flanks. Unupgraded Castle has LC in it and you will end up with a mix anyway so adding more of unit X shouldn't screw you. So you shouldn't worry because you are already screwed.
Testing above and some other things I learned that Housecarl retinue vs random ilkhanate levy stack is pretty badass. Assault from sea on to mountains gave me two penalties but 30k vs 30k and the Housecarls won 90%++ average 10k lost 20k++ killed. Only loss was when they did heroic counter charge with two of their flanks. They had better generals too.
One final thing, am I supposed to have a typo in my file for the tactics? Does it read volly_tactic or have I accidentally erased a letter and saved? Might be a typo because I also noticed that powerful charge in game is called awesome charge in the file... but I'm paranoid now! :O
It's a typo. Fortunately, they also put it that way in localization. They also misspelled harras_swarm 
There's also Awesomer Charge.
There's also Awesomer Charge.Now that I think of it there was fient too
Time to correct something what I said. I now remember the 400lc 10k hi stack never had raid as an option in melee. Raid has a trigger of 0.2 which is high. Most tactics have 0.01. I wonder if that means 20% and 1% of units in a flank?
It would mean this isn't accurate "It seems the amount doesn't matter at all as long as you have at least some." for raid tactic because 20% is more than just some.
Time to correct something what I said. I now remember the 400lc 10k hi stack never had raid as an option in melee. Raid has a trigger of 0.2 which is high. Most tactics have 0.01. I wonder if that means 20% and 1% of units in a flank?
It would mean this isn't accurate "It seems the amount doesn't matter at all as long as you have at least some." for raid tactic because 20% is more than just some.
Thank you for testing this. This implies that availability of a troop type affects if a tactic is available but its numbers don't affect how often it is used. (Aside: the percentages are not uniform across generals and fights, so there is SOMETHING affecting them). This was arguably OK when everybody was just wandering around with levies (because it was very hard to significantly impact army compositions), but is very much not OK with retinues (or put another way, I don't want 5000 archers doing diddly squat because i've also put in 400 LC and the general goes 'lets do harrass! herp derp'). This smells like good idea, badly implemented.
Valinn's comment that the very large penalties on some tactics isn't a great design choice is also a valid point.
The game also has an amazing knack of providing lots of feedback, but not enough useful feedback, so for example we get told the bonuses and penalties to our units, but not their base stats (thats in a game file), so we can't parse together what it all means.
NO. Contrary to how the game presents it, but medieval generals didn't just jumble their units together and pick a tactic at random (they could be uncreative and make bad decisions but that isn't jumble and pick). They were limited by the troops they had available, and the skills of those troops, but they weren't going to pick tactics relying on lots of light cavalry or archers if they had very few. And separating out infantry/cavalry and giving them roles was the basic that every general did (well tried, in some circumstances it was done for them, but even that was reversion to customary (or plain bad) tactics, not French fighting like Turks or Mongols because of dice rolls).
Yes, they would be percentages, but 1% of a flank is a really, really small no (eg to remove that 1% tactic as an option for your 400LC retinue, the flank would be need to 40,000+ (which equals attrition in almost all cases)
Last edited:
One extra variable is martial. Harass has different modifier at 8, 12 and 16 martial. I noticed my 17 and 18 generals had both 45.7 but when I switched a 15 martial one he had about 38 or so.
Good point about having low triggers is you don't need to stress over meeting the trigger or not because you'll have them anyway.
Good point about having low triggers is you don't need to stress over meeting the trigger or not because you'll have them anyway.
That's not what I think and that's also not the feel I get from the game. I don't think they are making "wrong" choises when not using the most damaging tactic. It's a simulation of a battle we can't see so no reason to stress over dice rolls. Maybe they had to send their cavalry because X or Y and then the archers had to hold.
I think the issue with this thread is that it assumes you are mixing LC with anything else, which is fantastically bad idea.
Make one flank nothing but light cav, give it a flanker or cav leader general with 20+ Martial, enjoy the slaughter as Harass and Raid are picked 99% of the time granting you +300% in skirmish and +240% in melee (during which time the enemy will probably either be already broken or near broken).
Then your LC flank with harass, from a flank (bonus damage), the centre.
When I was running my 20k sized retinue (Egyptian), it was 9000 LC on the right flank with my emporer who was Cav leader and martial skill of 25. The middle was 400 HI/100 Archer retinues and the left flank was the generic LC/HC flank (smallish). I routinely beat 25K sized byzantine stacks due to a quick flank-rout.
Make one flank nothing but light cav, give it a flanker or cav leader general with 20+ Martial, enjoy the slaughter as Harass and Raid are picked 99% of the time granting you +300% in skirmish and +240% in melee (during which time the enemy will probably either be already broken or near broken).
Then your LC flank with harass, from a flank (bonus damage), the centre.
When I was running my 20k sized retinue (Egyptian), it was 9000 LC on the right flank with my emporer who was Cav leader and martial skill of 25. The middle was 400 HI/100 Archer retinues and the left flank was the generic LC/HC flank (smallish). I routinely beat 25K sized byzantine stacks due to a quick flank-rout.
Thats only for retinues right ? As levies are always mixed if you don't prioritise.
Still impressive.