Swordfigters vs. Pikemen

[Ben]

~Mod: Continued from here: viewtopic.php?f=22&t=1797&p=30883#p30883 ~

Why? Pikemen are way inferior to swordfighters...

[Esthlos]

Why? Pikemen are way inferior to swordfighters...

Cheaper.

[Ben]

Not really. You'll need about 2.5x the amount of pikemen to match swordmen. Swordsmen cost only 1 more iron and coal, plus another building. Easily worth it.

[Esthlos]

Not really. You'll need about 2.5x the amount of pikemen to match swordmen.

Actually, it depends on how many ranged troops are there... 15 Pikemen (30 iron needed) easily and utterly destroy 10 Sword Fighters (30 iron needed).
Though, adding in a few Crossbowmen changes this... with 5 for each team, things begin to get more balanced, and it sometimes actually happens that the team with Sword Fighters wins.
(Again, this has all been tested in a test map, AI vs AI... it's quite hard to run tests in an actual game)

[pawel95]

XD² Well in a game without ranged you may be correct. However in the current release with that "op" bowmen that are just the best choice against xbows and co (in combination with swords). So like 20 pikemen won´t even reach the army of the enemy when this guy will be smart and have like 40+ bowmen. When they reach it, because you have way more then this 20 pikemen(then the swordmen will just kill them without any death propably )

[T*AnTi-V!RuZz]

Pikes talk cooler than swordsmen.

The admin has spoken.

[dicsoupcan]

Pikes talk cooler than swordsmen.

The admin has spoken.

Pikes speak like heavy smokers who hardly have any lungs left...

[Ben]

Not really. You'll need about 2.5x the amount of pikemen to match swordmen.

Actually, it depends on how many ranged troops are there... 15 Pikemen (30 iron needed) easily and utterly destroy 10 Sword Fighters (30 iron needed).
Though, adding in a few Crossbowmen changes this... with 5 for each team, things begin to get more balanced, and it sometimes actually happens that the team with Sword Fighters wins.
(Again, this has all been tested in a test map, AI vs AI... it's quite hard to run tests in an actual game)

I can see that there is no point in arguing with you. Do what you want, I don't really care if you deliberately make the wrong decision

@Tom LOL! Totally agree

[dicsoupcan]

@Tom LOL! Totally agree

NO, the swords have a way cooler voice

[Esthlos]

I can see that there is no point in arguing with you.

Cute.
Arguing has a point as long as everyone brings logically valid arguments; I'm running repeated and repeatable tests, and reporting that the results don't actually validate the idea that Pikemen are way inferior to Sword Fighters.
In fact, they suggest that for that to be true an at least noticeable group of ranged soldiers has to take part in the battle.

Which was expected, since Sword Fighters were changed so that they could be a counter to ranged troops, similarly to how Pikemen are a counter to cavalry units.

So, why did you state that "there is no point in arguing with" me?
I sure hope it's not because I still disagree after you brought up that ponderous argument where... oh wait, you didn't.

Very cute that you consider it pointless to argue with me. Ironic, too.

[dicsoupcan]

I can see that there is no point in arguing with you.

Cute.
Arguing has a point as long as everyone brings logically valid arguments; I'm running repeated and repeatable tests, and reporting that the results don't actually validate the idea that Pikemen are way inferior to Sword Fighters.
In fact, they suggest that for that to be true an at least noticeable group of ranged soldiers has to take part in the battle.

Which was expected, since Sword Fighters were changed so that they could be a counter to ranged troops, similarly to how Pikemen are a counter to cavalry units.

So, why did you state that "there is no point in arguing with" me?
I sure hope it's not because I still disagree after you brought up that ponderous argument where... oh wait, you didn't.

Very cute that you consider it pointless to argue with me. Ironic, too.

you can run all the maptests you want, but in practical terms you would just die because there are always ranged units on the field. pikes are inferior to swords as individual combat units. pikes need a number advantage to win over swords, but 1 vs 1 swords are superior. besides that in actual online games ranged units are always present, so pikes die faster due to the amount of ranged damge so the advantage they provide in numbers are no math for the disadvantages it has. maptests prove nothing, try to make it in actual gameplay and if it works there you can say pikes can be more effective.

[Esthlos]

pikes are inferior to swords as individual combat units. pikes need a number advantage to win over swords, but 1 vs 1 swords are superior.

This is where the "they're cheaper" factors in... if you train Pikemen, you have a numeric advantage simply because you can make more with pretty much the same materials and time (actually, you'd need more gold, but I never saw that becoming a noticeable problem).

you can run all the maptests you want, but in practical terms you would just die because there are always ranged units on the field. [...] besides that in actual online games ranged units are always present, so pikes die faster due to the amount of ranged damge so the advantage they provide in numbers are no math for the disadvantages it has.

Let's see: test map, 24 battles for each "configuration", every troop had the same number of columns (that is, 10 - except when there were less than 10 soldiers, of course).

51 Pikemen vs 34 Sword Fighters (102 iron per side): Pikemen won 12 times, Sword Fighters won 12 times;
50 Pikemen vs 30 Sword Fighters and 5 Crossbowmen (100 iron per side): Pikemen won 15 times, Sword Fighters won 9 times;
50 Pikemen vs 20 Sword Fighters and 20 Crossbowmen (100 iron per side): Pikemen won 22 times, Sword Fighters won 2 times;
50 Pikemen vs 10 Sword Fighters and 35 Crossbowmen (100 iron per side): Pikemen won 23 times, Sword Fighters won 1 time;
50 Pikemen vs 0 Sword Fighters and 50 Crossbowmen (100 iron per side): Pikemen always won.

Even factoring in the fact that Pikemen can't run and are melee soldiers, you still can't say that they are "way inferior to Sword Fighters".

I can upload the map if you'd like to do the tests again (it is a pretty simple map, with a 10x10 grid of impassable terrain where 8 battles for each of the above configurations are ready to start; each cell of the grid is a square with 24-tiles-long edges).

Out of curiosity, I also tried to put Pikemen in 12 columns while leaving the other troops how they were and ran 8 battles for each "configuration"... Pikemen pretty much dominated, only losing 2 battles (1 of the "51 Pikemen vs 34 Sword Fighters" and 1 of the "50 Pikemen vs 30 Sword Fighters and 5 Crossbowmen").

maptests prove nothing, try to make it in actual gameplay and if it works there you can say pikes can be more effective.

Test maps are the only way to go, since to prove anything you have to make many repeatable tests where you weed out as many confounding factors as possible; this method is also known as the "scientific method".

One of the confounding factors that would make actual gameplay reports meaningless is, for example, each player's town building skills: a battle between 2 Sword Fighters and 50 Pikemen would prove nothing about their worth, nor would a battle between 90 Sword Fighters and 5 Pikemen, since these differences are very likely due to different town building skills.

[dicsoupcan]

Yet again, maptests prove nothing here. swordfighters are superior to pikemen in 1 vs 1, and there are always ranged units here. i can also make a maptest of 100 swordfighters vs 20 pikemen and then say swordfighters are superior in any way.

If you really want to test it try to use them in multiplayer battles, i would even go 1 vs 1 against you where you use pikemen and i use swordsmen and i am sure i will win all the time. becasue those ''scientific tests'' really leave out many factors that are being use din actual gameplay, like flanking and ranged unit tactics. These ''tests'' are flawed if you really want it to be ''scientific''.


and this is why ben said there is no point arguing with you, since you have an opinion that you will not change and back it up by useless tests.

[Ben]

Esthlos, please, stop wasting your time on these tests! You've already proved that swordsmen are superior in one of your earlier tests. Why continue this?


EDIT: I figure that I better be more clear, as to avoid any confusion. Before, you said that swordfighters start shining when ranged units are put into the mix. This is what makes swordfighters better than pikemen. See, your "scientific method" is miscalculated because you are not testing anything meaningful. In fact, you might as well be testing Long-Swordsmen vs. Halberdiers in AoK! These tests would result in the same accuracy as the ones you have already done.

Lets go back a bit, I said that swordsmen are superior to pikemen. What you have tested, is that in open battlefields, equally supplied iron armies of pikemen will beat the equivalent amount of iron in swordfighters when they (the pikemen) are able to flank the swordfighters. This may be true, but again, meaningless.

You have forgot to factor many things:
1) Cost of gold to train the pikemen (pikemen armies cost more gold than swordfighter armies)
2) Time of training pikemen armies (more recruits = more schools + time)
3) Bonuses of swordfighters (shield bonus and (less important) storm attack)
4) Battles in closed fields
5) Battles in cities
6) Addition of leather units (especially the bowman)
7) logic

In a real game, you it isn't likely that you will see pikemen-only against swordfighters-only. However, if this did happen (doubtful), the player isn't going to attack you in an open field.
But what really makes swordfighters so good is the bowman. Bowmen and swordfighters work so well together. Swordfighters can take a lot of damage well doing a lot of their own, well bowmen cut the pikemen ranks to sheds. It is unlikely that bowmen and pikemen armies (especially pikemen and crossbowmen armies) will cause significant casualties to a swordfighter and bowmen army.

and that is why swordfighters are better than pikemen

[Esthlos]

Yet again, maptests prove nothing here. swordfighters are superior to pikemen in 1 vs 1,

If you are comparing them in a 1v1 you are not taking into account the price difference.
If two players of equal skill face each other, with one only training Sword Fighters and the other training Pikemen, then a 1v1 between their soldiers will never happen: the player training Pikemen will always have 50% more soldiers (if we don't factor in the higher gold cost - if you have any idea to quantify it, I'll happily run the tests again using this information)

and there are always ranged units here.

In 75% of the test battles there were ranged troops.
The last tests indicate that the presence of ranged troops actually makes Pikemen better, as long as Pikemen have no ranged allies (that is, as long as every resource is used to train Pikemen).

i can also make a maptest of 100 swordfighters vs 20 pikemen and then say swordfighters are superior in any way.

The numbers in the test battles are not random: every side is assigned an amount of Iron, and then possible configurations are computed.
In a battle where there are 100 Sword Fighters vs 20 Pikemen, the side with Sword Fighters has been assigned 300 Iron (you need 3 Iron to train one Sword Fighter) while the side with Pikemen has been assigned only 40 Iron (you need 2 Iron to train one Pikemen)
Again, if you have any idea to quantify the higher gold cost difference, I'll happily run the tests again using this information.

If you really want to test it try to use them in multiplayer battles, i would even go 1 vs 1 against you where you use pikemen and i use swordsmen and i am sure i will win all the time. becasue those ''scientific tests'' really leave out many factors that are being use din actual gameplay, like flanking and ranged unit tactics. These ''tests'' are flawed if you really want it to be ''scientific''.

If you want to play 1v1 with me I'd be happy to do it, but it wouldn't prove anything: too many confounding factors would be brought in, with tactics being just one of them.

Using a test map and assigning the sides to humans instead of AIs would make the tests better, since it would factor in tactics... but it'd need a way to make sure the tactics skills of the two players are at least similar.

and this is why ben said there is no point arguing with you, since you have an opinion that you will not change

I have an opinion that will change when and if proven wrong.
Not anytime sooner.
Outrageous, isn't it?

and back it up by useless tests.

...

Then help me build better tests.

Before, you said that swordfighters start shining when ranged units are put into the mix.

(No, I wrote that Sword Fighters start shining when ranged units are put into the mix on both sides -and this is likely due to their bonus vs arrows and darts-... but nevermind)

This is what makes swordfighters better than pikemen. See, your "scientific method" is miscalculated because you are not testing anything meaningful. In fact, you might as well be testing Long-Swordsmen vs. Halberdiers in AoK! These tests would result in the same accuracy as the ones you have already done.

Lets go back a bit, I said that swordsmen are superior to pikemen. What you have tested, is that in open battlefields, equally supplied iron armies of pikemen will beat the equivalent amount of iron in swordfighters when they (the pikemen) are able to flank the swordfighters. This may be true, but again, meaningless.

You have forgot to factor many things:
1) Cost of gold to train the pikemen (pikemen armies cost more gold than swordfighter armies)
2) Time of training pikemen armies (more recruits = more schools + time)
3) Bonuses of swordfighters (shield bonus and (less important) storm attack)
4) Battles in closed fields
5) Battles in cities
6) Addition of leather units (especially the bowman)
7) logic

In a real game, you it isn't likely that you will see pikemen-only against swordfighters-only. However, if this did happen (doubtful), the player isn't going to attack you in an open field.
But what really makes swordfighters so good is the bowman. Bowmen and swordfighters work so well together. Swordfighters can take a lot of damage well doing a lot of their own, well bowmen cut the pikemen ranks to sheds. It is unlikely that bowmen and pikemen armies (especially pikemen and crossbowmen armies) will cause significant casualties to a swordfighter and bowmen army.

and that is why swordfighters are better than pikemen

This is actually a good point.

Now, that "It is unlikely that bowmen and pikemen armies (especially pikemen and crossbowmen armies) will cause significant casualties to a swordfighter and bowmen army" wouldn't be a surprise... Sword Fighters have a bonus vs ranged units, thus if you are training ranged units to face Sword Fighters you are countering yourself and deserve to lose the battle.

What I was wondering with the last tests is whether a full melee army could beat an army of ranged units + Sword Fighters...
In the main tests Pikemen were not flanking nor being flanked by Sword Fighters (10 columns each troop, perfectly matching each other's first line... ok, I'm attaching the test map), but I didn't try to let them fight in a narrow passage. Let's see if anything changes there...

(the battles where Pikemen were flanking the Sword Fighters where mostly for fun, and to see if such a small change would have a significant impact. It had.)