[PoP 3.7063] In-depth guide to heavy cavalry units

[sher]

And what is, btw, your definition of a real battle?

From player perspective, real battle is when melee cavalry mess up enemy formations and scatter them, small number of mounted shooters distract opponent and lure his forces away further, strong mounted shooters like HA, Doomguide or CKO troops are actually killing suitable targets like shieldless troops. Cavalry acts in team with infantry which bind enemy in melee and even can include small number of shiledless troops. Infantry shooters capitalize on whole mess and kill with impunity. That's how balanced army works.

Player does not field 50 of something against 50 something, order them to charge and then does nothing. You assume that it doesn't matter and what matters is how many soldiers survive after that 50vs50 fight. Even called it "scientific approach" which is... well.

You wouldn't even progress to experiment phase if you'd follow scientific method since you'd had to determine comparative criterion that could be represented as number. Good luck with that with cavalry which primary role in real life and in PoP is to break formations and harass enemy forces, not to kill entire armies. IF you can determine such criterion then you can proceed to experiment phase to design such experiment scheme that would represent real phenomenon as close as it can be without or with calculated interference of tools you'll use for measurements and with some freedom of altering model parameters which would give you abilty to study correlation between criterion and these parameters. Then... On the other hand it's already boring and wasn't done properly right from the start.

Instead of this, "committing science" is much simpler proccess: "Whoah, we can run fights! Let's find out who'll beat crap out of who! Who'll win will be our champion, ultimate badass and we'll be sure that it's not our imagination cause we'll have them numbers which means science, or something like that."

 

[sher]

Which spawn armies have strong shooters other than noldor? Leonion even tested Lions against a noldor party and they did quite well. Not to mention AI shooters are easily disrupted and destroyed if you use any tactics at all.

Strong mounted shooters are HA and Doomguide like troops. Horse archery is very demanding if you really want to see kills frequently, so faction horse archers are too weak and just harass enemies killing only weak troops fast. Maybe RotCC can come close to real deal after 5 upgrades but I had no such experience, I don't rely on such troops in late game - harassment is more than enough.

Infantry strong shooters on the other hand are very widespread since after dehorsing 300 WP mediocre horse archer with good bow becomes elite.

If you'll introduce player's tactics then any test will be simply impossible.

Russian tests also revealed Lions dominance over silvermist rangers. Why don't you do this--give a situation and explain why a different melee KO would be better than Lions?

Melee cavalry in simple 50vs50 test will own any archers. They'll simply smash into firing line and then it'll become melee. If it happens without problems in your games just like that then such test is good for you. There's always will be deserters and Lions will own them for sure!

And which KOs scale better with upgrades? Presumably griffons due to PS8 and ebony swords, but even then, their weak armor is a real liability and it's not like Lions get zero benefit.

With lower starting pros but with good equipment and setups. Lions and SLC will hardly notice, not zero but +50 to 350 is significant while +50 to 440-450 and already fast weapons... Overall it depends on opponents they'll face but to test them unupgraded seems not very reasonable if you want to rate them for player to use - person who'll be interested in "power tests" will upgrade them for sure I think. It's not like you'll get very different results, it's just the matter of whole approach when you're testing I don't know what.

 

[Leonion]

Good luck with that with cavalry which primary role in real life and in PoP is to break formations and harass enemy forces, not to kill entire armies.

I'm not sure about RL, but in PoP all-cavalry army is the best one a player can field. And it can be used very effectively to destroy entire marshal forces, not just "break formations".

And, sure, what we do here is not the perfect science, and I've already said it multiple times. But it is as close to science as we can afford considering that we are not spending our time on an actual research which will result in one more article in our bibliographies, financial support, promotion opportunities and so on, but just comparing troops in a computer game.
So, yeah, once again, we do lack statistical significance because we don't have time to repeat every battle 50 times, and controlled variables are not exactly constant (random AI behavior can affect test results significantly), but it is still better than your blah-blah-blah-you-are-all-wrong-your-tests-prove-nothing-I-know-better-yet-I-haven't-shown-you-any-numbers-from-the-game-that-would-back-my-position-so-I'll-just-keep-theorizing.

 

[sher]

Good luck with that with cavalry which primary role in real life and in PoP is to break formations and harass enemy forces, not to kill entire armies.

I'm not sure about RL, but in PoP all-cavalry army is the best one a player can field. And it can be used very effectively to destroy entire marshal forces, not just "break formations".

No, it's not. Because cheaper and easier to maintain army will get the same job done. You didn't determine the criterion of "best", otherwise it would be clear that such tests are not very useful simply by looking at this creterion. In players army it doesn't matter that Queen's Guard will beat crap out of Veccavia Knight in your "tests" since in proper army they act as lancers primarily and as lancers they're pretty similar but Queen's Guards are more expensive so in time it becomes an issue as wages are rising but you don't get raise in power.

If it's "these troops prevail in fictional all charge 50vs50 fights" then it's "OK, whatever", but if it's "Now we know the best KO!" or something like that then it confuses people.

but it is still better than your

Wrong again. You cannot prove transition from your fictional tests to actual game except "This is obvious!" so it's only a matter of faith. Some people have faith simply by getting some numbers. Will it help them to improve their armies? Doubtful.

 

[gpower]

Look mate, obviously just becuase a unit does well in the test does not mean it is the best unit for every conceivable situation you can run into in the game. Even so, these tests are valuable because before this we don't even have this imperfect measure. We just have subjective field experience and opinions.

 

[sher]

Look mate, obviously just becuase a unit does well in the test does not mean it is the best unit for every conceivable situation you can run into in the game. Even so, these tests are valuable because before this we don't even have this imperfect measure. We just have subjective field experience and opinions.

If only it was true. What surprising results these tests gave you? That melee cavalry slaughters horse archers (except Lions vs HA because of 2h weapons)? That melee cavalry with better stats, more armor and damaging weapons will win against weaker melee cavalry? That melee cavalry will own infantry without tactics? Do you really need tests for that?

I can tell you what false results these tests gave you. That Lions are better than SLC. From player perspective they're definetely not, at least until latest small nerf of SLC but they (Lions) still were winning the "test" before that. Simply because SLC have lesser casualties rate in real game (they have shields and lances) and high burst damage in the beginning of a fight. They were the only troops worth to buy in quests because of this, with other troops you were just spending money on heavy cavalry similar to what you can get from prisoners.

Will you send your Lions or other melee cavalry in frontal attack on army full of Sarleon Halberdiers or Fierdsvain Berserkers because according to these tests they have nothing to fear, like at all?

So, what's the point then?

 

[Harding Grim]

You don't have a high opinion of any of us if you are throwing out strawmen like suggesting we'd order cavalry into halberdiers. To humor you, yes, Sarleon Knights actually would probably have better luck doing so than Heroine Adventurers. Why do think trolling like that is productive?

 

[Leonion]

I can tell you what false results these tests gave you. That Lions are better than SLC.

Will you send your Lions or other melee cavalry in frontal attack on army full of Sarleon Halberdiers or Fierdsvain Berserkers because according to these tests they have nothing to fear, like at all?

I'm having a deja vu.
I already spent some time in the past proving to another disbeliever that Lions do better against Empire Gladiators than Dawns (his argument was also "Lions don't have shields => they will fail").
Now it's you.
Alright...
Just one test, one try. it's probably randomly biased, but I'm not gonna waste my time doing more fights just to prove to you that you are wrong.
Just because Lions look vulnerable with their 2h swords, it doesn't mean they are easy kills for their enemies.
And that's what tests are for. To actually see how strong troops are, to no longer judge them by their looks, like you do.

 

[heckani]

Look mate, obviously just becuase a unit does well in the test does not mean it is the best unit for every conceivable situation you can run into in the game. Even so, these tests are valuable because before this we don't even have this imperfect measure. We just have subjective field experience and opinions.

If only it was true. What surprising results these tests gave you? That melee cavalry slaughters horse archers (except Lions vs HA because of 2h weapons)? That melee cavalry with better stats, more armor and damaging weapons will win against weaker melee cavalry? That melee cavalry will own infantry without tactics? Do you really need tests for that?

I can tell you what false results these tests gave you. That Lions are better than SLC. From player perspective they're definetely not, at least until latest small nerf of SLC but they (Lions) still were winning the "test" before that. Simply because SLC have lesser casualties rate in real game (they have shields and lances) and high burst damage in the beginning of a fight. They were the only troops worth to buy in quests because of this, with other troops you were just spending money on heavy cavalry similar to what you can get from prisoners.

Will you send your Lions or other melee cavalry in frontal attack on army full of Sarleon Halberdiers or Fierdsvain Berserkers because according to these tests they have nothing to fear, like at all?

So, what's the point then?

What unit do you from your personal experiences would say are the best Sher? Like a top 3 heavy cavalry whats the strongest? Not cost effecvtive just overall strongest? I´d be very curioys as i know you´re a smart player (not meant sarcastically)

Also, Sher does have many good points. The tests surely wouldnt prove anything in regards to scientific results that could be used in proper arguments. On the other hand testsers doesn´t have time to test real battles  1000 times. Where theyd´d ahve to lower/raiser amount of archers, and they´d also have to record the battles to compared the closest battle where both heavy cavlary behaved the same ways etc. it´d be hundreds if not thousands of hours of work and more. It could however definitely prove an "overall average" better troop. And equipment here would probably imo mean insanely much. Because in the end profiencies, PS all that is all good and dandy, but when we choose troops, equipment just means so much. Here im talking about when we see those idiots with a lance thats dehorsed just getting slaughtered.

In any case, lets try not to sound too hostile, and if someone wants to test something and that makes them happy. let them. Theres no doubt hat the sarleon knight is supreme IN THESE TESTS. Would i take them in the game over another knight? I doubt it, but that might be because of looks or weapons or my own persoanl experiences. But im staying neutral here. I appreciate the work put in from al the testers, and its not fun to have someone keep saying the work you do is pointless or berate you, on the other hand Sher realyl does have a point and i cimpletly agree.

I think the closest compromise we can get is perhaps, take these tests. And then compare them to our own personal experiences. If we all thought that lets say
Fierdsvain huscalr beats Sarleon recruit. The tests shows that. Okay, in our experience from the game, like when we play it normally. Does fierdsvain huscarls beat recrutis there?
9/10 said yes. 1/10 was a troll and said no. (gotta count in the trolls guys or this is not a realistic project).

Then we could conclude, that in our opinion, probably the huscarl is the better choice. The tests doesnt prove it, neither does our experiences since they are not able to be used in a "hey we proved it because i felt they were better way". But it can at least give us an idea, a factor to consider, an indicator as to whos better. Yet for all that said we´d still be very biased. So really coutn on your instincts, and they are all good at what they do, they asre not much better than to just pick whatever you like, any thing can get things done in pop. Its just when we get ot the top top tier troops, the margins are harder to see than comparing recruits to elite troops.

The tests and the video from Leonion is very neat. Just becasue i klike battles. But lets please try and close it here and just get a long with each other. I get its a sensitive topic, and sher´s "nasty truth" (well said) is a good eye opener, i doubti t coudld be said differently. SOmetimes you need to hammer on people whos deluded aor cant take the truth because it´d crush them emtnally so they atry block it out and get hurt when people keep trying to say they are wrong. Let them have their faith i say. No harm done. Unfortuantely harm might be done if too amny people would use thesese tests to say people like Sher is wrong, which he isn´t.

You all made very wise points though as as you Leonion said, you dont have time to take in all factors like you´d need to, to properly use it. But that doesnt mean you cant do these tests and if it gives you confidence in them being the better troop, then use them. No need for all this hostility. 

Back to my first question, sher, what top 3 cavalry would you say is the strongest from your experiences? And why if so?
Really curious as to who, i expect SLC might be one of them since you mentioned them earlier? Is that the shadow legion?

 

[Harding Grim]

I'm not sure what's being missed here. The tests have already demonstrably yielded valuable insight, such as in the difference between Immortals and Forlorn Hopes showing that the added offensive power of 2-handed weapons > the added defensive capability offered by shields against medium-heavy cavalry. It has now demonstrated that the Sarleon Knight is in fact superior to the rest of the non-Knighthood Order heavy cavalry vs. like units. This certainly was a valuable insight as I had no idea that they were that strong and that at best would be comparable to the Knights of the Dawn or Griffon.

I actually am in fact more likely to use Sarleon Knights now.

Sher, you deflect anything that comes out of these tests with non-sequiturs and straw-men, or target Leonion's throw-away casual aside on horse archers which was one sentence long.

Thermocline and leonion preemptively made your arguments redundant by explaining their controls and variables, and here's why. Both of them explained upfront that they just issued a charge command. It is not a huge leap to assume that no micro-management was involved (it was pretty clear to me) and that players who do micro heavily can take the results with a grain of salt.

I don't like to spell it out, but now it's clear I have to: these tests are useful for players who want to minimize their supervision of horse archers and lance cavalry throughout the battle, and want to know which units fight best independently while they focus on killing enemies themselves. Granted, not many radically different army compositions were there, but Kodan is very much like Meregan or Boris the Raven, and the battle vs. the mixed combined-armed army by a standard Ravenstern lord are all applicable for the player. Add in K'juda and the tests from other threads and we have a good understanding of the test units vs. at least 4 of the unique spawns in the game. 

A good player at around level 20 can consistently kill 50+ of the enemy's most dangerous units in a prioritized manner, singlehandedly (if he's focused on just that and not microing his units) so there is value in knowing what units to avoid and having a sense of the general hierarchy. Before this, I would have assumed Sarleon Knights were at most like another heavy knight, but it is clear from the results that they are very consistent and dominant vs. a wide array of enemies to a higher degree than what would otherwise appear to be peer units.

There is no need to caricaturize players who enjoy this style of play by implying they'd charge cavalry straight into a wall of halberds or pikes.

"The tests and the video from Leonion is very neat. Just becasue i klike battles. But lets please try and close it here and just get a long with each other. I get its a sensitive topic, and sher´s "nasty truth" (well said) is a good eye opener, i doubti t coudld be said differently. SOmetimes you need to hammer on people whos deluded aor cant take the truth because it´d crush them emtnally so they atry block it out and get hurt when people keep trying to say they are wrong. Let them have their faith i say. No harm done. Unfortuantely harm might be done if too amny people would use thesese tests to say people like Sher is wrong, which he isn´t. "

Pretty condescending there, but I agree we can drop this. We can all believe what we wish, and trust each other to be intelligent enough to draw what conclusions we may from the available data, and give it the proper weight.

 

[heckani]

I'm not sure what's being missed here. The tests have already demonstrably yielded valuable insight, such as in the difference between Immortals and Forlorn Hopes showing that the added offensive power of 2-handed weapons > the added defensive capability offered by shields against medium-heavy cavalry. It has now demonstrated that the Sarleon Knight is in fact superior to the rest of the non-Knighthood Order heavy cavalry vs. like units. This certainly was a valuable insight as I had no idea that they were that strong and that at best would be comparable to the Knights of the Dawn or Griffon.

I actually am in fact more likely to use Sarleon Knights now.

Sher, you deflect anything that comes out of these tests with non-sequiturs and straw-men, or target Leonion's throw-away casual aside on horse archers which was one sentence long.

Thermocline and leonion preemptively made your arguments redundant by explaining their controls and variables, and here's why. Both of them explained upfront that they just issued a charge command. It is not a huge leap to assume that no micro-management was involved (it was pretty clear to me) and that players who do micro heavily can take the results with a grain of salt.

I don't like to spell it out, but now it's clear I have to: these tests are useful for players who want to minimize their supervision of horse archers and lance cavalry throughout the battle, and want to know which units fight best independently while they focus on killing enemies themselves. Granted, not many radically different army compositions were there, but Kodan is very much like Meregan or Boris the Raven, and the battle vs. the mixed combined-armed army by a standard Ravenstern lord are all applicable for the player. Add in K'juda and the tests from other threads and we have a good understanding of the test units vs. at least 4 of the unique spawns in the game. 

A good player at around level 20 can consistently kill 50+ of the enemy's most dangerous units in a prioritized manner, singlehandedly (if he's focused on just that and not microing his units) so there is value in knowing what units to avoid and having a sense of the general hierarchy. Before this, I would have assumed Sarleon Knights were at most like another heavy knight, but it is clear from the results that they are very consistent and dominant vs. a wide array of enemies to a higher degree than what would otherwise appear to be peer units.

There is no need to caricaturize players who enjoy this style of play by implying they'd charge cavalry straight into a wall of halberds or pikes.

"The tests and the video from Leonion is very neat. Just becasue i klike battles. But lets please try and close it here and just get a long with each other. I get its a sensitive topic, and sher´s "nasty truth" (well said) is a good eye opener, i doubti t coudld be said differently. SOmetimes you need to hammer on people whos deluded aor cant take the truth because it´d crush them emtnally so they atry block it out and get hurt when people keep trying to say they are wrong. Let them have their faith i say. No harm done. Unfortuantely harm might be done if too amny people would use thesese tests to say people like Sher is wrong, which he isn´t. "

Pretty condescending there, but I agree we can drop this. We can all believe what we wish, and trust each other to be intelligent enough to draw what conclusions we may from the available data, and give it the proper weight.

I gotta say you´re acting a little sensitive. If you´re going to call it condescending, but then so be it. All i said is "let them have their faith" (both to you, to Sher, to anyone. For and Against, let them have their faith). Because this is just people arguging about the tests being stupid or not being stupid really. I mean when it comes down to it then, who are you to tell anyone that they have to accept these tests as facts? And who is anyone to tell you that you cant use these tests? Noone. Thats why i said no need to argue. Some will like the tests, others won´t. I like them, to me they just give me a factor/an indicator and i might try Sarleon knights to see if they really are this good. I can only be happily suprised. I dont see anything negative in this.


anyway I´ll leave you guys to it, i just tried ending this hostile discussion. Its all good.

Harding Grim, since you said the things about "offensive power of 2-handed weapons > the added defensive capability offered by shields against medium-heavy cavalry."

Does this mean that 2 handed are just better than giving them shields? So for instance your own custom knight hood order would also benefit from no shields, but just a 2 handed sword? I mean sarleon knights sometimes get shields, but do those units perform worse than those sarleon knights with shields?

In that case then as i thought equipment plays the biggest role if they are all about even in stats. Lancers are great for a charge or the beginning clash where they can instant kill units, thats the only time it´d be nice to have 5-10 untis with lances in front imo. But that thing you said about 2 handed beats the shield. Thats a huge gamebreaker if thats true. So do you think its better to give custom knights the best 2 handed possible and skip the shield?

Really curious about that.

Anyway overall i think the tests has been done quite substantionally, with alot of varied enemy parties. So thanks for all your work and pelase keep doing it. Im in for it to support it.  Loved doing this myself in the old m&b with the mod that allowed you to.

 

[thermocline]

@heckani - Best weapon for CKO has been tested extensively: https://forums.taleworlds.com/index.php?topic=344402.0

@Harding Grim - IMO, you are being too quick to draw a conclusion about the reason for KotL's performance. Falcons have almost the same weapon setup and perform considerably worse. Moreover, the silver claymore does about ~86 base damage per hit in the hands of a Lion after taking the 23.5% damage reduction for 2H on horse into account. 86 base damage is not much higher than what other KOs can do, and is significantly lower than a Griffon with an ebony sword.

 

[heckani]

@heckani - Best weapon for CKO has been tested extensively: https://forums.taleworlds.com/index.php?topic=344402.0

@Harding Grim - IMO, you are being too quick to draw a conclusion about the reason for KotL's performance. Falcons have almost the same weapon setup and perform considerably worse. Moreover, the silver claymore does about ~86 base damage per hit in the hands of a Lion after taking the 23.5% damage reduction for 2H on horse into account. 86 base damage is not much higher than what other KOs can do, and is significantly lower than a Griffon with an ebony sword.

Thank you very much for that link sir. 
Suprised you mentioned in that thread that people said the NWS was the best. Ive always used the RBS, but had some debates with people if the 50% (i believe it was) you lost when them using it as a 1 handed was worth it. I think it ws 50% speed+damage, not sure. Happy to see you tested it, great job and even more happy was using the best weapon all along, good judgement there. 

 

[sher]

Back to my first question, sher, what top 3 cavalry would you say is the strongest from your experiences? And why if so?
Really curious as to who, i expect SLC might be one of them since you mentioned them earlier? Is that the shadow legion?

Strongest is the cavalry on fast warhorse with good armor rating, without 2-handed weapons as this reduces survivability significantly, and with option of a lance (or a long spear like Black Iron) as this allows safest kills possible if enemy is scattered. Strongest melee cavalry troops are Maiden Nobles, Shadow Legion Centurions (SLC), Barclay Conquistadors. Only SLCs can be acquired in good numbers and its a big advantage of this order due to large number of empire noble troops you can rescue and upgrade very fast and cheap. As horse archers who not suck in melee HA have no equal. SLC + HA (60/40 or 40/60 ratio depending on opponent to face) is a dream team for pure cavalry PARTY, not army.

I had very limited experience with SLC since last nerf of their starting pros to 450, thought it should have minimal impact since their horses were buffed in speed but with 10 (which is a very small number to get definite conclusions) SLCs as reward for quests (it's not possible to buy them in large number anymore) I had an impression that they died faster than they're used to but it was very small party participating in hard fights so I'm not certain at all. Previously they were undoubtedly toughest to kill and you could end up with party full of them simply from quest rewards.

-----
As last page clearly indicates my "opponents" are arguing mostly with voices in their heads and do not read or memorize what I'm actually posting here, on the same bloody page, not days ago. So it feels like a waste of time. Readers can have a warning now, in the end it's not my problem if people prefer to confuse themselves and to frustrate people so much that they cannot understand even simplest logic was not my goal at all. So I will not discuss these "tests" anymore.

 

[Kelefane]

Are we really arguing real life vs a fantasy game in a Fantasy world setting? lol

 

[Harding Grim]

Not a graceful way to bow out sher, but thank you. I'm sure just as you did in the Fierdsvain balance thread you will quit the argument knowing you're in the right.

Are we really arguing real life vs a fantasy game in a Fantasy world setting? lol

If you derail a thread I put a good amount of work into for a full page, arguing that the tests do not accomplish what their parameters state that they are testing, then I will sooner or later put a stop to it.

Does this mean that 2 handed are just better than giving them shields? So for instance your own custom knight hood order would also benefit from no shields, but just a 2 handed sword? I mean sarleon knights sometimes get shields, but do those units perform worse than those sarleon knights with shields?

This piece of data suggests that 2-handed weapons is preferrable for infantry when facing cavalry, since the test was with Radiant Cross knights vs. Immortals & Forlorn Hopes. It does not make sense to apply this test to cavalry. To test whether the same holds true for cavalry means we have to take for example the Knights of the Dragon, and a KO with guaranteed 2 -handers (so the Runed bastard sword or ebony bastard sword with no shield to guarantee they wield it as a 2-hander and not a 1-handed/2-handed weapon).

However, if your sergeant is built as a heavy infantry unit, I would be more inclined to gie them a 2-handed weapon. In sieges you can cover them with your shielded cavalry, and in the field a standard or square formation kills elite cavalry noticeably faster than Huscarls and kill more than enough to offset any stray losses to archer fire. As you said and I agree with, you can make anything work in the game. I'd like to see a Forlorn/Immortal-level troop armed with 1-handed+shield though, since Huscarls have lesser stats than fully trained CKOs and don't make for a strict comparison.

@heckani - Best weapon for CKO has been tested extensively: https://forums.taleworlds.com/index.php?topic=344402.0

@Harding Grim - IMO, you are being too quick to draw a conclusion about the reason for KotL's performance. Falcons have almost the same weapon setup and perform considerably worse. Moreover, the silver claymore does about ~86 base damage per hit in the hands of a Lion after taking the 23.5% damage reduction for 2H on horse into account. 86 base damage is not much higher than what other KOs can do, and is significantly lower than a Griffon with an ebony sword.

I remember that thread. It's interesting that the 1-handed Rune Bastard Sword is still the best weapon. Do you know mechanics-wise if 2-handed weapons wielded 1-handed still suffer from the horseback damage penalty?

I don't think I stated a conclusion about why the KoTL are so effective. I've mainly said they arecapable of surviving a lot and that the tests bear out that they are objectively superior to at least some of the KO orders. Their armor is basically 56/18. They can still one-shot most enemies with the speed bonus, the morningstar does only marginally less base damage and suffers no damage penalty.

I'm guessing that Agility is the biggest factor. As far as I know, the KotL are the highest Agility KO from horseback. The St differential between KotL and most units is only 3 (27 vs 24), but the Agility differential is huge (27 vs 15). Almost twice the Ag that the other Orders have. The Shadow Legion are at 18. The claymore is a fairly cumbersome weapon which is why Falcons may not be as deadly. The Knights also have 440 melee weapon proficiencies, but similar to strength it's only about an average of 60 points above the other KOs.

I've always thought of agility as mattering to footmen in dueling situations, but for knights, that faster swing could translate to better timing? The claymore is not the fastest weapon so that higher weapon speed could be important. It's the only discernible advantage they have over the Griffons (who have comparable armor, have higher attributes, lower proficiencies, BETTER weapons, but are still only 24/15 in Strength and Agility).

 

[Leonion]

Strongest is the cavalry [...] without 2-handed weapons as this reduces survivability significantly, and with option of a lance

In other words, the opposite of what tests show.

I'll just remind you what the criterion we use to compare troops is a number of survivors.
When we say "Lions are the ultimate best KO, according to tests", the first thing we mean is "Lions have the highest survivability rate".
Yet you ignore all of it, preferring to judge them by their look. A wise approach, without a doubt.

I'm guessing that Agility is the biggest factor.

Agility is sure the most outstanding thing about Lions, but... it just can't matter.
It only affects running speed and, as one of Russian testers established, the bonus it gives to speed is (!) 0,2% per point (i.e. Lions' agility, which is as about 10 points higher than that of other KO knights, will only increase their running speed by 2%).

 

[Harding Grim]

How did they manage to determine that if you don't mind me asking? That's just a crazy test.

This is mysterious then. The Knights of the Lion consistently beat every single other Knighthood in direct matchups in addition to beating unique spawns like Kodan with higher number of survivors (useless according to sher though), yet if we recognize that Agility makes no difference, and equipment loadout is nothing special, then +3 additional Health with the same ironflesh and +60 weapon proficiencies are the only things that stand out, which would make them faster than most of the other Orders, though by how high a margin?

Riding skill? They may be more effective lancers. This doesn't contradict the finding that lancers make for weaker. Simply they may use them more effectively than the Griffons, or the Falcons. They only wield the Heavy lance, which is actually one of the shorter lances in the game. Better for hammering cavalry when in chase situations but not as good in that initial clash.

I don't have any Sarleon Knights in my party now, does anyone have the Riding skill of these guys? I pulled up Morgh's editor but it only shows up as 'F'. For comparison, Griffons/Falcons/Dawn/Eventide all have Riding of 7... presumably F is used in place of something higher than 7? It would be just stupid for TaleWorlds to use F instead of 1-6 for example. Is Riding mainly useful for the horse riding requirement and otherwise useless on its own like Agility, or does it actually increase the speed and maneuverability by greater than .1%? If it actually does what Warband advertises, then I can see that being impactful.

Just more devastating to infantry and archers if you can build up charge faster, faster speed damage bonus, then in the direct matchup vs. other cavalry you will always outpace the other riders and also be able to hunt them down even when they are being evasive. It would make the perfect subjective reason, but as we've learned, things that seem reasonable on face value may be disproven by the cold hard facts and objective results.

 

[thermocline]

It's a good question and I don't know. In sher's defense, the way to eliminate Riding as a possibility would be to run two sets of tests--say 5 trials of KotL vs something dangerous, and then 5 more trials using a nerfed KotL with only 5 riding and see if there is a big difference in performance.

Also, looking at the math, each upgrade is worth about 2-2.5% damage for most KOs. So 60 WPF would translate to a maybe 6% damage boost and a 3% attack speed boost, according to Leonion's tests.

It could also be the horse. Take away the Lion warhorse and give them a regular Riding 4 warhorse and see how they perform.

My suspicion about Lions' superior performance is that they are average or above average in every category, with no weaknesses. Average armor, above-average weapons, top-tier stats, above-average horse, above-average mobility, summed together, gives you a pretty good knight.

---

Also, 2h weapons used one-handed do suffer the damage penalty. See Leonion's post in the WPF vs PS thread--he just made a new post today or yesterday.

---

Also, Griffons don't have comparable armor. Griffon plate is only 53 base, making them among the weakest cavalry KOs in terms of raw protection. I think small differences in armor actually matter a lot, considering Radiant Cross's dominant performances in 3.7 (they are the most heavily armored KO in 3.7) and the absolutely terrible performance of the LVs (base armor ~48-50), even though 55 base armor CKOs with similar weapons do quite well.

I think armor in this game works like a lot of other RPGs... basically, it has increasing returns to scale, rather than diminishing. In a lot of games, going from 1 armor to 2 armor (or whatever the scale is) makes hardly any difference, but going from 98 armor to 99 armor doubles your survivability. I don't understand Warband's armor formula because of the math^exp function, but I suspect this has something to do with it. I've observed that CKO troops are very powerful even when their WPF is still in the 300 range--they usually have plate armor 55/17, mettenheim gauntlets +11, and Eventide helm 58/6 when I start using them, for a total of 72 body armor. I think the armor is the reason, they just don't die easily.

Also, I suspect the horse matters a lot, since it tends to tank a lot of the damage that would have hit the rider. I looted a Netherworld Charger early in my current game and gave it to Roland--he barely dies against small parties, even Heretics and Snake Cult. I think the NC is tanking a lot of his incoming damage. Lions do have a more heavily armored warhorse than most KOs.

 

[Leonion]

How did they manage to determine that if you don't mind me asking? That's just a crazy test.

What exactly?
Bonus to running speed?
By setting a timer and repeatedly running from point A to point B in the game using different agility and athletics values.

I pulled up Morgh's editor but it only shows up as 'F'.

F is 15, AFAIR. it's just a hexadecimal number.
Which is strange, though, considering that PoP only allows skills to be as high as 10, and we also see Lion's riding as 10 in the game.

This is mysterious then.

Lions are indeed a miracle.
My Excel table which I use for troop evaluation gave them a pretty average grade at first. Then I refined my formulas in multiple ways, and now Lions are the second strongest KO. Computer still believes that SLCs are better.
I'm doing more tests atm against normal lords from each factions, let's see if the result clarify something.
But Lions are already winning. Again. By a margin.