Current division Meta?

[Cavalry]

I do wonder what value "spread the incoming attack" has now that we have the coordination stat to focus fire onto one division at a time.

Whatever, a division now can only do 50-70% damage to an enemy division vs 100% on previous DLC; and spread the rest to all other divisions.

Now for example on 95 width battle, the maximum width that can join is +22%: 95*1.22 = 115 width

With 27w, you can join as much as 4 division, total 108 widht
With 21w template, you can join as much as 5 division, total 110w, that is 25% more total Org, more Org regain rate... 22 21 w is the highest width that can easy join battles.

The trinity support: support Art, support Rocket Art, support AA get discounted in cost; so using smaller division now will have both cost effective and total firepower per width effective. More and more players using 10w in attack now. Bigger width can have some advantage in defense and in logistic/maintainance support, but I think not go over 22w.

 

[Corpse Fool]

Hey man, I'm fine with advice. I'm not fine with being patronised. Please actually tell me what flaws it has, instead.

Well, now that you've actually asked me what those flaws were instead of expressing complete disinterest, I'll tell you. I'm basically just going to copy my reddit comment.

• Misnames over width as over stacking
• Ignores actual over stacking
• Ignores tactics
• Ignores certain flank counts and weights them in a particular way not all would agree with
• Portions of it still refer to max 20% over width for max of 30% penalty
• The problem with taking average or weighted averages of the penalties, is that we aren't fighting in average or weighted average terrain. The same idea behind the statement "The average person does not exist." The devil is in the details and focusing on averages too much loses the context. In game, we'll be operating under specific conditions, not averaged ones.
• It only goes from widths 6-50, by 1's and there are more possible widths inside and outside that range.
• Squaring the penalty from going over width is extremely strange. Somehow, having 3x26 for 78/80 is better than having 3x27 for 81/80, that actually has a penalty that reduces them to ~79.48 fighting width and having more stats than the 26's, but squaring the penalty leaves them with 77.99 total width.

I do wonder what value "spread the incoming attack" has now that we have the coordination stat to focus fire onto one division at a time.

I have made some graphs approaching that sort of topic, here. Probably not immediately what you're looking for, but it is something I guess.

 

[Corpse Fool]

Now for example on 95 width battle, the maximum width that can join is +22%: 95*1.22 = 115 width

With 27w, you can join as much as 4 division, total 108 widht
With 22w template, you can join as much as 5 division, total 110w, that is 25% more total Org, more Org regain rate... 22 w is the highest width that can easy join battles.

I'm still not really understanding why you want to go so far over width.

 

[Cavalry]

I'm still not really understanding why you want to go so far over width.

Let's see an attack scenario. The total Org of overwidth, more division is of course superior + 25% or +50% more total ORG. Now look at how much attack

3 division 27w, 500 attack, 81w at 90w front (18.5 attack per width): Overwidth penalty: 0%, total attack: 1500
4 division 27w, as above: 108w/90w Overwidth penalty: 30%, total attack: 1400 (less than 7% above)
5 division 22w, 18.5 attack per width like above (actually can have more), 110w/90w, penalty 33%, total attack 1358

So the total attack is roughly the same, overwidth or not, probably lower 7%. But I will attack with even more divisions, to have reserve to trigger the Encirclement tactics, this way I get +25% attack and INSTANTLY remove the penalty because the width goto +50% to 135 width, and most of my division are already join, instead of waiting, and we have only 12 hours for the tactics to last.

Well, the 22w max is probably a littel wrong calculation. Better go 21W or lower width to be safe.

 

[aletoledo]

Let's see an attack scenario. The total Org of overwidth, more division is of course superior + 25% or +50% more total ORG. Now look at how much attack

3 division 27w, 500 attack, 81w at 90w front (18.5 attack per width): Overwidth penalty: 0%, total attack: 1500
4 division 27w, as above: 108w/90w Overwidth penalty: 30%, total attack: 1400 (less than 7% above)
5 division 22w, 18.5 attack per width like above (actually can have more), 110w/90w, penalty 33%, total attack 1358

So the total attack is roughly the same, overwidth or not, probably lower 7%. But I will attack with even more divisions, to have reserve to trigger the Encirclement tactics, this way I get +25% attack and INSTANTLY remove the penalty because the width goto +50% to 135 width, and most of my division are already join, instead of waiting, and we have only 12 hours for the tactics to last.

Well, the 22w max is probably a littel wrong calculation. Better go 21W or lower width to be safe.

I will add to this, in that if the attacking divisions are roach infantry, while the damage is coming from CAS, then it doesn't matter what the over-width penalty is. The goal is to just bring enough ORG for the CAS to do it's job. Since there is not an ORD penalty for going over-width, then squeezing as many divisions into the battle is the goal.

If you're using armored divisions to be your damage dealer though, I would think being as close to width, while not going over, is the ideal. You're not worrying about ORG at that point, since you have the armor and hardness bonuses. So a 30w division would be best, since you could still trigger the enciclement tactic with a minimum of 4 divisions, but with 0% penalty.

 

[Corpse Fool]

3 division 27w, 500 attack, 81w at 90w front (18.5 attack per width): Overwidth penalty: 0%, total attack: 1500
4 division 27w, as above: 108w/90w Overwidth penalty: 30%, total attack: 1400 (less than 7% above)
5 division 22w, 18.5 attack per width like above (actually can have more), 110w/90w, penalty 33%, total attack 1358

My idea here is that, if we know what size combat we're going to be going into (90), and if we know how many templates we want going into that combat (5), why not use something more like 90/5=18w? Or if you're specifically arguing against going past 22w, we could use the /3 and /4 factors being 30w and 23.

3x30w, 90/90, no penalty, 1665 attacks.
4x23w, 92/90, -3.3%, 1609.5 attacks.
5x18w, 90/90, no penalty, 1665 attacks.

All of those still fill the width and nothing stops you from having reserves to still trigger your encirclement tactic. At 3/4/5 templates for total org, we're getting more attacks out the other end

I think that you're actually betting on the encirclement proccing, which changes the width from 90 to 135, so lets look at that.

4x30w, 120/135, no penalty, 2220 attacks.
5x30w, 150/135, -16.67%, 2312.5 attacks.
5x27w, 135/135 no penalty, 2500 attacks.
5x23w 115/135, no penalty, 2127.5 attacks.
6x23w 138/135, -3.3%, 2467.9 attacks.
6x22w, 132/135, no penalty, 2442 attacks.
7x22w, 154/135, -21% penalty, 2247.5 attacks.
7x18w, 126/135, no penalty, 2331 attacks.
8x18w, 144/135, -10% penalty, 2397.6 attacks.

It still seems like in that sort of situation (which doesn't account at all for other situations, which may be more of the point), that 30, 27, 23 or 18 have some fairly notable advantages over 22, though I suppose there are also some disadvantages. At the 6 or 7 templates for org, 23 or 18 get more attacks out.

 

[Cavalry]

My idea here is that, if we know what size combat we're going to be going into (90), and if we know how many templates we want going into that combat (5), why not use something more like 90/5=18w? Or if you're specifically arguing against going past 22w, we could use the /3 and /4 factors being 30w and 23.

Yeah, for this 90w, you may squeeze something like 7% more attack. But at the cost of lower Org and lower Encirclement tactics benefit (need time to join up to 135w). But the problems is many times it is not exactly 90w, terrain and tactics can vary it. 45w and 67.5w is ugly for 30w divisions. And terrain varies too, in practice we will enjoy attack everywhere. The best terrain for tank is the terrain without AT. A divisions over 21w can have the risk not fill the front and not trigger the Encirclement tactics.

I am not saying 21w is the best, we often start at 18-20w tank division and may keep that or go lower a bit at a low xp cost.

Another thing about micro control, if you want not go overwidth with smaller divisions, says you start with 9 x 10w division in 90w, no reserve then it does not trigger Enciclement, you need to click 1 more time for a 10w division. But that division can join the battle and now you don't have reserve again to trigger Encirclement. So why bother with too much micro, now click all the 20x 10w tank you want, and throw them to the battle just one time, ignoring the width. You get the benefit that these go to target provinces faster. In the past we use 40w to limit joining battle but now higher width is no longer good enough to try.

 

[Corpse Fool]

Yeah, for this 90w, you may squeeze something like 7% more attack. But at the cost of lower Org and lower Encirclement tactics benefit (need time to join up to 135w).

I'm looking at 18.5-22.6% more attack, which is much larger than 7% in the 90w case.

I'm not really sure the amount of org you'd be missing from using 5x18w v 5x22w is really all that big of a deal. Using pure infantry with your support trinity mentioned earlier, the 22w has 47.14 org, the 18w has 45 (before doctrines). Which is less, but not by much. Besides, raw org matters less than e-org, and in the 90w case, even having a capped out penalty at -33% reduces the stats your 22w template offers to be as if it was only 14.74w, which is reducing its breakthrough/defense which means you may or may not be suffering increased damage which would drain your raw org faster, resulting in less e-org.

As for the encirclement tactic benefit, I just realize that 5x22w in a 90w doesn't actually allow the 5th template to fight, it is 0.2 width too big. So you're stuck with 4 templates which aren't filling up the width, and so encirclement won't proc. I suppose this is why you mentioned 21w instead in other comments scattered around, or perhaps why you used a 95w combat as the initial example rather than 90w.

With regards to org, 21w requires some sort of artillery in either a 10/1 or 9/1 format, which accounting for the trinity means 42.86 or 41.54 org, which would be less than what the 18w offers. Using 10/1 and shifting the AA from the trinity into the line would be 46.15, which is slightly higher than the 18w. Of course, this is ignoring doctrines which might wiggle things around a bit.

But the problems is many times it is not exactly 90w, terrain and tactics can vary it. 45w and 67.5w is ugly for 30w divisions. And terrain varies too, in practice we will enjoy attack everywhere.

I agree, and this sort of thing is more of what I was hoping to get from you. That you had already dug into this and could show that choosing particular widths that would intentionally make you so close to the maximum over width penalty, would actually reliably be presenting you with some benefit. And I certainly can see the angle you're approaching it from now, where you are banking on an early width expansion to bypass the penalty without allowing the enemy to drag more width into the combat at the start. I certainly can see some benefits to that sort of manipulation, thank you for helping me realize that.

Another thing about micro control, if you want not go overwidth with smaller divisions, says you start with 9 x 10w division in 90w, then it is not trigger Enciclement, you need to click 1 more time for a 10w division. But that division can join the battle and now you don't have reserve again to trigger Encirclement. So why bother with too much micro, now click all the 20x 10w tank you want, and throw them to the battle just one time, ignoring the width. You get the benefit that these go to target provinces faster. In the past we use 40w to limit joining battle but now higher width is no longer good enough to try.

This is something I was going to bring up myself against the 22w. 6x22w at 135 performed a lot better than 7x because of the penalty (though, that org...), which meant that you'd have to be throwing specifically 6 formations rather than just grabbing a stack and right clicking.

I'd also like to suggest that rather than encirclement which you can't prefer and can be countered by a preferable tactic, as well as having some officer restrictions on it, using MA doctrine and its mass charge tactic has the same sort of +50% width, can't be countered and can be preferred, with an extra bonus from a spirit that gives it something like a 58% proc chance. Of course, this makes you use MA-R and also choosing some of your spirits/preferred tactics for you, which is something you might not be interested in.

 

[Cavalry]

I'm looking at 18.5-22.6% more attack, which is much larger than 7% in the 90w case.

Good enough, if there is an example for that

I'm not really sure the amount of org you'd be missing from using 5x18w v 5x22w is really all that big of a deal.

Of course not much, I mostly look at 18-21w vs 27w.

Using pure infantry with your support trinity mentioned earlier, the 22w has 47.14 org, the 18w has 45 (before doctrines). Which is less, but not by much. Besides, raw org matters less than e-org, and in the 90w case, even having a capped out penalty at -33% reduces the stats your 22w template offers to be as if it was only 14.74w, which is reducing its breakthrough/defense which means you may or may not be suffering increased damage which would drain your raw org faster, resulting in less e-org.

Thanks for the mentioned defense penalty. Now I just +33% width to the defense template to cover it, made it more toward 21w. For attack 12-15w will do. Note that in defense we cannot choose overwidth or not. If the template is small then they will join overwidth automatically.

 

[Corpse Fool]

Good enough, if there is an example for that

Your 5x22 total was 1358 attacks, wasn't it? 5x18 was 1665, which is 22.6% higher.

Rather than assuming 18.5 per width, lets try using actual numbers, pure infantry with just a single support artillery for simplicity, 6 attacks per infantry 15 per support artillery. 22w is therefore 81 attacks per template, but the capped penalty reduces it to 54.27. 18w is only 69 attacks which is less than 81, but it doesn't get reduced so it stays at 69, which is 27% more than 54.27. That number would jump around a lot depending on what your balance of attacks is between supports and line, which is template design and tech level and doctrines and stuff, which I didn't account for here.

The operative mechanism here is that it is only the line infantry battalions that add to the width, but the resulting width penalty lowers the attacks the supports offer. Since the lions share of the attacks for these templates are going to be coming from the support companies instead of the width (especially with the trinity), that penalty is going to be hurting your damage output. Using the same sort of pure inf templates in the 90w scenario, even 30w ends up with more attacks in total than the 22w 30w are 105 each 315 total, 22w being 54.27 each, 271.35 total. For comparison, 18w would be 345.

27+ width is definitely where you'd start seriously considering adding line artillery to the template, if you haven't already with something like a 9/1 or whatever. Even then, with the trinity having 2 support arties, you are still getting the lions share of the attacks from the support companies.

 

[Cavalry]

If you're using armored divisions to be your damage dealer though, I would think being as close to width, while not going over, is the ideal. You're not worrying about ORG at that point, since you have the armor and hardness bonuses. So a 30w division would be best, since you could still trigger the enciclement tactic with a minimum of 4 divisions, but with 0% penalty.

Most of armored divisions still need Org because they start with lower Org and low number vs big number of infantry Org wall. One attack, you often attack again at only half Org. Redundant Org let them do constant attack after break a hole, which is quite powerful because you don't need extra armor division and extra supply to exploit the hole.

 

[ltccone]

I use 9-1 (21w) infantry divisions and 4 MARM, 6 MOT, 1 SPA, 1 SPAA, and 1 TD panzer divisions (27w). That works for me.

 

[Corpse Fool]

4 MARM, 6 MOT, 1 SPA, 1 SPAA, and 1 TD panzer divisions (27w).

Why the SPG? I'm assuming the SPAA is for %highest armour and the TD is for %highest piercing and some hard attack, but I'm not really sure what the SPG is doing.

 

[ltccone]

Why the SPG? I'm assuming the SPAA is for %highest armour and the TD is for %highest piercing and some hard attack, but I'm not really sure what the SPG is doing.

You're correct about the why for the TDs and SPAA, for the SPA is for the extra soft attack, but mostly for the role play.

 

[Maxwell Tornado]

Well, now that you've actually asked me what those flaws were instead of expressing complete disinterest, I'll tell you. I'm basically just going to copy my reddit comment.

• Misnames over width as over stacking
• Ignores actual over stacking
• Ignores tactics
• Ignores certain flank counts and weights them in a particular way not all would agree with
• Portions of it still refer to max 20% over width for max of 30% penalty
• The problem with taking average or weighted averages of the penalties, is that we aren't fighting in average or weighted average terrain. The same idea behind the statement "The average person does not exist." The devil is in the details and focusing on averages too much loses the context. In game, we'll be operating under specific conditions, not averaged ones.
• It only goes from widths 6-50, by 1's and there are more possible widths inside and outside that range.
• Squaring the penalty from going over width is extremely strange. Somehow, having 3x26 for 78/80 is better than having 3x27 for 81/80, that actually has a penalty that reduces them to ~79.48 fighting width and having more stats than the 26's, but squaring the penalty leaves them with 77.99 total width.

I have made some graphs approaching that sort of topic, here. Probably not immediately what you're looking for, but it is something I guess.

... There's a flanking mechanic in this game? I thought that was only in EU4.

We need averages, because unless you're crazy, you're not gonna make different templates for different terrain, and actually micro them to always be in their terrain.

 

[Corpse Fool]

... There's a flanking mechanic in this game? I thought that was only in EU4.

Yes. Attacking from multiple provinces will increase the available width, reduce fort levels, and is one of the ways to gain XP for an officer trait, I forget which one.

We need averages, because unless you're crazy, you're not gonna make different templates for different terrain, and actually micro them to always be in their terrain.

Even then, only looking at the averages is going to be hiding certain tidbits. 27w templates are recommended in a fair number of places, but they most likely will suffer the -30% over width penalty in the basic 90w plains combat if you have 4+ in the stack. They do perfectly fit into the 135 if a flank opens or you roll a particular tactic, but you won't always be under those conditions.

Rather than looking at averages, I've made up graphs like my "catgraph" that show the 'lowest maximum penalty' that a particular width might suffer under any and all sets of circumstances, which is not an average and is helpful.

 

[Maxwell Tornado]

Yes. Attacking from multiple provinces will increase the available width, reduce fort levels, and is one of the ways to gain XP for an officer trait, I forget which one.

Even then, only looking at the averages is going to be hiding certain tidbits. 27w templates are recommended in a fair number of places, but they most likely will suffer the -30% over width penalty in the basic 90w plains combat if you have 4+ in the stack. They do perfectly fit into the 135 if a flank opens or you roll a particular tactic, but you won't always be under those conditions.

Rather than looking at averages, I've made up graphs like my "catgraph" that show the 'lowest maximum penalty' that a particular width might suffer under any and all sets of circumstances, which is not an average and is helpful.

Ah, that. Right. I didn't know it reduced fort levels.

So a range of about 19.6 to 22.8 is optimal in your opinion?

 

[Corpse Fool]

So a range of about 19.6 to 22.8 is optimal in your opinion?

In terms of universality, that particular range does have a lot of appeal. I wouldn't really go so far as to say it would be optimal, but it does save a lot of micromanagement.

 

[Maxwell Tornado]

In terms of universality, that particular range does have a lot of appeal. I wouldn't really go so far as to say it would be optimal, but it does save a lot of micromanagement.

Okay, let's try this again:

Me smart man. Me make güd fight team. Fight team WIDE. How wide?

 

[Corpse Fool]

Okay, let's try this again:

Me smart man. Me make güd fight team. Fight team WIDE. How wide?

It's been several hours of my trying to find a way to respond to this, and I still have no idea what to say.