Investigation of the Phalanx formation

[therother]

Placeholder.

(This thread is a pruned merging of two threads transferred into the Ludus Magna from the Colosseum. The threads were frogbeastegg's Step two three, shuffle two three, dance two three : the big phalanx question, and Doug-Thompson's A thin line of spears. Click the appropriate links to see the originals.)

[Doug-Thompson]

Custom battle. Medium difficulty. Greek hoplites vs. Juliaii Haspati. Grassy flatlands. Human player deploys the Greeks. Large unit size.

1.Left the hoplites on default formation rank and file. Turned on phalanx. Turned off guard mode. Ordered an attack with one click. No other human intervention.

Phalanx won with 44 men remaining out of 81, including recovered wounded.

2. Put the hoplites in a formation only two ranks deep. Everything else the same. (Formation was slightly closer to enemy because it was put into its new formation slightly ahead of the starting position.)

Phalanx won with 71 men remaining, including recovered wounded.

Also of note: Phalanx drift to the right was serious in the default depth and negligible in the shallow depth. In fact, the shallow phalanx drifted to the left for quite a while after running out of Romans to kill on the right.

=========

Same thing only with the Romans having Equites.

1. Regular formation; Roman defeated. 71 hoplites left.

2. Thin formation. Romans defeated. No losses. Not even wounded.

Against the regular formation, the Romans charged, got beat, retreated, circled around a bit and charged again, hitting a corner. They only charged the thin formation once.

[Ziu]

Interesting. Does the thin formation wrap around the enemy lines?

[Orvis Tertia]

This is quite interesting indeed. I think it's pretty obvious that--in terms of realism and historical accuracy--the results should have been just the opposite.

[Sin Qua Non]

A thin line in front of several 8+ ranked lines works wonders against cavalry and sometimes even chariots. The first row screens the formation, distrupting the cavalry, while the heavy formation behind don't have the charge bonus to worry about.

[Doug-Thompson]

A thin line in front of several 8+ ranked lines works wonders against cavalry and sometimes even chariots. The first row screens the formation, distrupting the cavalry, while the heavy formation behind don't have the charge bonus to worry about.

What about a thin line backed by another thin line?

[Jeanne d'arc]

In default formation the hastati wraps around both left and right flank of the phalanx and maybe this is why there are more losses.
In a two rank deep formation the hastati have no chanche to make it to the flanks of the phalanx, they just get impaled by spears and cant even get close enough to make any casualties on the phalanx.

[GodsPetMonkey]

More interested in finding why its happening (although I had a few good ideas) then actually see it happen I decided to try it out.

Using pricepes rather then hastati vs normal hoplites, with the default 5 ranks, its pretty easy to spot the main problem, the pricepes work arround the hoplites, so whilst from the front, they get slaughted, the real fun in on the sides, and ultimately thats where the losses come from.
http://www.users.on.net/~roehr/flanked.jpg

But apart from that, it seems as if most of the time, the 2nd and 3rd rows never get to fight with their spears! http://www.users.on.net/~roehr/rows.jpg. The circles show that the red and blue (3rd and 2nd) rows all pretty much line up, whilst the green (1st) row are doing tho fighting, so while we do have the potential of more then one spear attacking a target at the same time, the target doesnt seem to get close enough for the other rows to be in range. Thus the 'new' rank bonus is negated by tubby romans unable to pass between the first row of spears!

Of course, this same problem happens with the thin ranks. What really gives thin ranks more power is the fact that the AI isnt smart enough to attack the flank. http://www.users.on.net/~roehr/thinrow.JPG, so they end up trying to push though the ranks of spears infront of them, a virtual date with Death.

[Sin Qua Non]

What about a thin line backed by another thin line?

Yes, this also works, but sometimes cavalry can break through two thin lines. No real matter, since most barring cataphracts will be distrupted and cut down. I just like the full phalanx behind the thin line because I find it easier to reform the line as the battle dictates.

I've even been tinkering with stringing a thin line of archers in the open space between the pike points and the first rank of a phalanx, so that they are protected by the pikes, but don't distrupt the formation or cause friendly casualties. It's been mixed results. If the archers survive the initial charge (and don't run at the enemy like idiots), then they can cause significant casualties in the perfect section - the front row of the enemy's melee.

[Sinner]

I've even been tinkering with stringing a thin line of archers in the open space between the pike points and the first rank of a phalanx, so that they are protected by the pikes, but don't distrupt the formation or cause friendly casualties. It's been mixed results. If the archers survive the initial charge (and don't run at the enemy like idiots), then they can cause significant casualties in the perfect section - the front row of the enemy's melee.

Which would potentially work very well with the Germans using their Chosen Archers and Spearband or the Egyptians with their Pharoh's Archers and Nile Spearmen. Both types of archer are better than average when it comes to melee, so they should hold up a lot better.

[Doug-Thompson]

Interesting. Does the thin formation wrap around the enemy lines?

No. I tried switching off phalanx in the middle of one battle so it would, and the result was disaster.

Experience with lousy, non-phalanx spears (Eastern Infantry) does result in lapping around.

This is quite interesting indeed. I think it's pretty obvious that--in terms of realism and historical accuracy--the results should have been just the opposite.

I couldn't agree more. However, I repeated this little stunt four more times with the hastati tonight. The smallest number of phalanx survivors after five tries was 58 men. The largest was 78.

In order, the number of phalanx survivors were: 71, 58, 71(again), 66 and 78. This includes the healed wounded.

[Red Harvest]

CA nerfed the formation depth effects in RTW, either that or they omitted them altogether. Missile units are another example of the problem (as your horse archer tests have shown, as well as my own tests of jav, slingers, and archers.)

I've been deploying my phalangites 4 deep because I quickly learned that deep phalanx formations were not of any use and were instead a liability.

There is one other aspect that is harder to judge: multiple units. I suspect that two multi-rank lines side by side will perform better than two thin lines one behind the other when facing two or more attackers. I suspect the thin line will suffer casualties quickly allowing the two to be defeated in detail. However, the two might behave differently if they "stack." Hmm, I'll have to try this.

I really hope the major problems can be fixed when the patch(es) arrive(s). Right now the fighting side of the engine pales in comparison to MTW.

[Doug-Thompson]

CA nerfed the formation depth effects in RTW, either that or they omitted them altogether.

Agreed --Which pretty much voids years of hard-won experience on how to fight with spears in the Total War series.

Missile units are another example of the problem (as your horse archer tests have shown, as well as my own tests of jav, slingers, and archers.)

Small correction here. You're giving me credit for tests by Frogbeastegg and Louis de la Ferte Ste Colombe. I just applied their findings, as did other players.

I'm interested in the results of any "stacking," as you mention.

In default formation the hastati wraps around both left and right flank of the phalanx and maybe this is why there are more losses.
In a two rank deep formation the hastati have no chance to make it to the flanks of the phalanx, they just get impaled by spears and cant even get close enough to make any casualties on the phalanx.

That's very true, especially in the initial stages. By the time the phalanx drifts in the thin formation and there is some wrapping around the edge, there aren't enough hastati left to make a difference.

The bizzare thing, though, is that the thin line drifts to the left. It keeps doing this. It wasn't a fluke, as I thought.

Much seems to depend on whether the spears hit on-center or off, but that's just how it appears.

[Paul Peru]

One more finding in the "useful to know, slightly cheesy to exploit"-bag, then.

The bizzare thing, though, is that the thin line drifts to the left. It keeps doing this. It wasn't a fluke, as I thought.

Perhaps a medium depth can be found where they stay put?

[Doug-Thompson]

One more finding in the "useful to know, slightly cheesy to exploit"-bag, then.

Guilty as charged, but at least my intentions were good. I wanted to give a simple, clear demonstration of what's wrong with spear units in R:TW. I think it worked.

Perhaps a medium depth can be found where they stay put?

Wondered that myself, but would bet that a three-rank formation would get wrapped again.

========

Fridge and KyodaiSteeleye;

I will soon know more about how brittle a thin line is when charged by cavalry. Beyond that, I just don't know how all this plays out in a real fight. These are not the type of questions that can really get answered in a simplistic custom-battle demonstration with only one human. I know that long, thin formations are much harder to maneuver and to change facing.

[therother]

This thread is a pruned merging of frogbeastegg's Step two three, shuffle two three, dance two three : the big phalanx question, and Doug-Thompson's A thin line of spears. Click the appropriate links to see the originals.

[Jambo]

This may seem a little controversial to some, but I reckon phalanx units really ruin the challenge of the siege and city aspects of this game. It's almost impossible to lose a city to the AI when you're defending it with phalanx units. Simply align them in the city plaza at strategic choke points and wait for the AI to launch itself on the units in whatever random and inane order it chooses. I will sometimes lose as little as 1000 when defending cities with phalanx units. I'm not even talking about the top phalanxes either, simply militia hoplites or levy pikemen will do.

Granted, this is as much to do with the strategy the AI employs for city assaults as it is for the power of a phalanx in tight choke spots. The AI fails to use archers properly once inside the city - instead of firing on stationary units like phalanxes, it will use all archers and skirmishers in melee to get to the city plaza.. In fact, as soon as a city wall is breached or once inside a city, all AI units do is charge in an ill-thought-out manner to the city plaza.

The trouble is, trying to find a suitable solution to keeping the game challenging when using a nation with phalanx units? In the field I'm aware phalanx units are far more susceptible and major nerfing will impact on this part too much. Does anyone have any ideas?

So far, what I've decided to do is try reducing the "mass" of phalanx units a little (by 0.2), so that in city fights there will be a greater chance that a massed mob of attackers can disrupt their formations a little more.

[Epistolary Richard]

Granted, this is as much to do with the strategy the AI employs for city assaults as it is for the power of a phalanx in tight choke spots. The AI fails to use archers properly once inside the city - instead of firing on stationary units like phalanxes, it will use all archers and skirmishers in melee to get to the city plaza.. In fact, as soon as a city wall is breached or once inside a city, all AI units do is charge in an ill-thought-out manner to the city plaza.

I don't think there's anyway to stop the AI from charging headlong forwards with all their troops if it's written into their siege AI. You could however, try them out using their normal battle AI and see if it has anything other than catastrophic effects.

I think this could be done by doing the following. Create a normal battle in the battle editor and put a settlement into it. Mark the settlement as ambient so the AI won't think it's fighting a siege battle and the other siege special rules won't apply. Redefine the starting positions to match that of a normal siege battle and put some troops in there and see what happens.

EDIT: Okay, I did and the AI couldn't handle it. It just stood outside and tried to demolish the town with its onagers and, when they ran out of ammunition, they just stood there. Even when I was taunting them by nipping in and out of the gaping holes in my walls.

[Bromley]

Firstly, sorry - I haven't read through the whole thread, so this might have been mentioned.

Whilst looking for something else, I found this in the pre-patch readme:

Unit Linking
The Rome: Total War manual contains a reference to the unit linking functionality that has been removed from the 3D battle user interface. On page 53 the description of the AI assistance button states "This button replaces the linking button when a group is selected." This linking button has been removed from the game.

So it looks like it might have been originally planned to have an ability to link units into one long line.

[MajorFreak]

uhm...okay, let's try to keep this on topic. lol.

[SpawnOfEbil]

I'm not sure if this has been mentioned before, but I've found out how to get a single unbroken phalanx line on vanilla RTW.

Edit the data_formation file, find the single line entry, and change the unit spacing from 2.0 to 0.0.

Hey presto, a perfect phalanx line.

[Puzz3D]

This may seem a little controversial to some, but I reckon phalanx units really ruin the challenge of the siege and city aspects of this game. It's almost impossible to lose a city to the AI when you're defending it with phalanx units. Simply align them in the city plaza at strategic choke points and wait for the AI to launch itself on the units in whatever random and inane order it chooses. I will sometimes lose as little as <10 soldiers and kill >1000 when defending cities with phalanx units. I'm not even talking about the top phalanxes either, simply militia hoplites or levy pikemen will do.

Granted, this is as much to do with the strategy the AI employs for city assaults as it is for the power of a phalanx in tight choke spots. The AI fails to use archers properly once inside the city - instead of firing on stationary units like phalanxes, it will use all archers and skirmishers in melee to get to the city plaza.. In fact, as soon as a city wall is breached or once inside a city, all AI units do is charge in an ill-thought-out manner to the city plaza.

The trouble is, trying to find a suitable solution to keeping the game challenging when using a nation with phalanx units? In the field I'm aware phalanx units are far more susceptible and major nerfing will impact on this part too much. Does anyone have any ideas?

So far, what I've decided to do is try reducing the "mass" of phalanx units a little (by 0.2), so that in city fights there will be a greater chance that a massed mob of attackers can disrupt their formations a little more.

I'm using Mordred's mod where all non-spear units are reduced to 0.75 lethality, all units get a +1 to their defensive value and a +4 to their morale, movement speed is set at 0.9. This works well in the field battles. The AI's single minded selection of "best individual matchup" works better because there is more time during the fighting for those matchups to play out. Its cav flanking moves make contact with better timing as well. Although, that tactic of chasing cav with infantry didn't work well in the battle where an AI 1000 man infantry army chased my 3 units of cav all over the battlefield and I was able to inflict 800 casualties before loosing. The AI did win the battle and some of the casualties healed.

In city battles with this mod, it is easy to defend by blocking city streets with phalanx units. If you look at the layout of a typical city, there are very few streets leading to the central plaza. For cities without walls with large enough plazas, you could defend on the central plaza itself so that the option of blocking choke points is not available to you. I didn't actually have the courage to do this in my Carthaginian campaign since I didn't know if I could hold cities that way. For cities with walls, I simply sally right away when sieged thereby turning it into a field battle. This also means I have to maintain decent sized garrisons which slows down my economic expansion, and I only train or retrain one unit per turn per city. So far this has worked to make the strategic game more interesting, and the battles are already more interesting with the mod. I keep phalanx units in guard mode all the time. I would have no chance against the Roman's better armies without plenty of sacred band infantry holding the line. I suffered 2:1 casualties overall when I took on SPQR, and lost about 40 sarcred band infantry units in the 12 years it took me to defeat them. The Carthaginian campaign I'm playing is still interesting at 134 BC where I've managed to control 24 provinces and the next 26 are not going to be that easy to get. Since I got to this point with only 2 saves, the AI factions have not been overrun with rebel provinces since they always move to recapture any that revolt.

[Silver Rusher]

I have read back here and have seen that there are quite a few people who actually think it is a good idea to pull your units out of phalanx, then move them in and put it back. While it looks good on paper, this is actually a really bad tactic. It works against infantry armies, but no...

I am currently playing through a campaign on Mundus Magnus, a mod which dramatically improves the campaign map and a few other things (if you still play vanilla RTW, I recommend you get it now), as the Parthians. The Seleucids are my biggest enemies, but I can easily crush their armies because of this tactic. Until I get catanks, I am using armies composed entirely of Horse Archers, Scythian Mercs or Persian Cav.

This sounds like a weak combination as there are no troops to support the missiles, but trust me, it works. To kill the Seleucids, whose armies at this stage usually compose of 70%+ phalanx units, I can simply ride over, pump their phalanxes full of arrows and wait until they get pissed off and lift their sarissas. Then, I charge straight at them causing an almost instant rout to them.

Out of the countless battles I have fought with the Seleucids, I can only remember actually losing 1 of them. I know from experience, phalanx units are pathetically weak when they aren't in that formation, so always keep those pikes ready. I think the time I did lose was because my horse archers ran out of arrows and charged into a phalanx out of impetuousity.

To summarise, when fighting eastern armies with phalanx armies, don't dare take them out of formation. The phalanx requires patience (which is why I hate using it lol) to be used successfully.

And no, lowering the sarissas in the middle of a melee does NOT work. Trust me, I know from plenty of Seleucid arse-whooping experience.

[A.Saturnus]

Generally, the behaviour of the AI isn`t a very good cue how to do something. Whether lifting the sarissa can work has to be found out in multiplayer.

[Puzz3D]

My campaign battles have played well with the 0.75 lethality to non-spears because I control the phalanx and don't take them out of phalanx formation, and the AI factions I've gone up against don't have phalanx units. Breaking up its battleline and lifting the pikes means the AI can't use the phalanx unit effectively.

Multiplayers repeatedly hit the "form phalanx" button because it forces the the men in the unit to switch from their sword back to their pikes momentarily.