American Civil War Game Club (ACWGC)

ACWGC Forums

* ACWGC    * Dpt. of Records (DoR)    *Club Recruiting Office     ACWGC Memorial

* CSA HQ    * VMI   * Join CSA    

* Union HQ   * UMA   * Join Union    

CSA Armies:   ANV   AoT

Union Armies:   AotP    AotT

Link Express

Club Forums:     NWC    CCC     Home Pages:     NWC    CCC    ACWGC
It is currently Thu Mar 28, 2024 12:36 pm

All times are UTC - 5 hours




Post new topic Reply to topic  [ 15 posts ] 
Author Message
PostPosted: Sun Oct 10, 2021 5:37 pm 
Offline

Joined: Mon Aug 30, 2021 11:28 am
Posts: 71
Up front: I hope my math is correct. I don't do probability often.

I noticed while playing JTS that not everyone seems aware that the user manual promotes the benefits of firing a whole stack of units at a target rather than firing the units individually. To quote the latest version of the JTS manual (version Oct. 2020):

Quote:
You may fire several units in a stack at once rather than each one individually.
Note: Firing multiple units at once does not increase the effectiveness of fire. Each unit's fire result is handled separately then the total casualties and fatigue are added together and the losses displayed. However, there are some advantages to firing stacked units.
● Firing multiple units at once in Turn Mode does reduce the chance for return fire as Opportunity Fire by the AI can be triggered each time you fire. So firing multiple units at once provides only one chance for return fire by the AI.
● Firing stacked units also results in all the casualties being added together before checks are made for whether or not a unit will check morale (see Morale) and whether or not guns are destroyed (see Combat Results) or crews are killed (see Uncrewed Artillery and Recrewing).

The benefits of the first bullet are clear. But I wondered whether the second bullet was actually a net benefit or not. So I decided to crunch some numbers. My conclusions follow:

Shooting at artillery: Shoot the whole stack at them! This increases the chance of causing either a crew kill from infantry fire or a gun loss from artillery fire. Shooting at artillery units individually with your stacked units will decrease your chance of causing a crew kill or gun loss. It does however increase your chance of causing a morale check on artillery. Gun loss and morale are checked by the program separately it seems. I'm not sure whether artillery is subject to morale checks due to fire combat (they can take morale checks due to a nearby unit routing though). If artillery is not subject to morale checks from fire combat, then that is one more reason to shoot the whole stack. Edit: I tested and artillery is subject to morale checks from fire combat.

Shooting at infantry or cavalry: Shoot at them with each unit in the stack individually! This has no effect on the losses the target takes compared to stack fire. But because of the way the probability of morale checks is calculated, firing stacked units individually does increase chance of the target unit making a morale check. Causing casualties to infantry is good, but causing morale checks is great. Morale checks provide immediate benefits. A passed morale check disrupts the unit. A failed morale check routs the unit and causes all units in the same or adjacent hexes to make morale checks.

These conclusions ignore other tactical considerations that you must consider as they arise. But if you were only going to fire at a single target with a stack anyway, keep those two points in mind.

The differences in probability of causing a gun loss, crew kill or morale check between firing a stack vs individual units can be pronounced, especially when a large amount of casualties are possible. When fewer casualties are possible, the effect can be negligible. Here are some examples.

Assume we have a stack of 8 artillery counters, all of the same type of gun so that firepower values are the same. Assume that if we used the stack to fire at a target (either individually or at once) that 8 casualties are possible on average and that each counter would cause on average 1 casualty. The probabilities of the various outcomes are (Note 1: the percentages are rounded. Note 2: artillery cannot cause crew kills, but the percentages are included here for convenience. Just pretend the artillery stack is now an infantry stack and ignore the gun loss probability.):

Casualties: 8 total (1 each)
Stack fire:
Gun Loss (GL) 32%, Crew Kill (CK) 8%, Morale Check (MC) 24%
Individual fire:
GL 28%, CK 8%, MC 27%

Even with the small amount of expected average casualties, there are differences in the probability of outcomes between stack and individual fire. I will list out a few more instances with increasingly higher casualties and working up in multiples of 8 for convenience.

For a total of 16 expected casualties, the probabilities are:
Casualties: 16 total (2 each)
Stack fire:
GL 64%, CK 16%, MC 39%
Individual fire:
GL 49%, CK 15%, MC 46%

Casualties: 24 total (3 each)
Stack fire:
GL 96%, CK 24%, MC 49%
Individual fire:
GL 64%, CK 22%, MC 60%

Finally assume an enemy unit gets close enough to our big artillery stack that we can cause some serious casualties.
Casualties: 64 total (8 each)
Stack fire:
GL 256%, CK 64%, MC 72%
Individual fire:
GL 95%, CK 49%, MC 89%

Notice how different the probabilities are between the two cases. An artillery stack firing at an artillery target for a 256% gun loss gives the same target a chance to survive by firing at it individually. Knowing this fact also opens up an even smarter play for us: we waste the firepower of four units to kill one gun. We can instead split the stack into two smaller stacks of four guns each to kill two guns for a 100% chance each; or we can split it into two stacks of three counters and one stack of two counters to have good odds of killing 2-3 guns. And so on and so forth. Correction: Artillery overkill can kill more than one gun from the same battery in the same fire result!

To conclude: if my math is correct, the implication in the JTS User manual that stack firing is better is not always true. It depends on the target. Again this doesn't cover every tactical situation. But if you are looking to maximize your chance of causing gun kills and morale checks, then then this advice may help.

_________________
WDS: Antietam, Chickamauga, Gettysburg, Overland


Last edited by Logrus Pattern on Tue Oct 12, 2021 7:08 am, edited 2 times in total.

Top
 Profile Send private message  
Reply with quote  
PostPosted: Mon Oct 11, 2021 4:22 am 
Offline
User avatar

Joined: Mon Jul 19, 2021 4:14 pm
Posts: 75
Good job, sir. I agree your conclusion in Gun Loss and Crew Kill. But I try to argue on the morale check.

---

I check some of all and believe your calculation is right. But the problem is they are based on an assumption, that all checks are applying for every firing loss instead of adding up all loss. It is correct for gun loss check and crew kill check. But is it also correct for morale check? It would be appreciated if someone can tell me whether the morale check before the movement phase is based on the every loss they suffered in the last enemy turn or total loss. (eg. Do 8 morale check for 8 times loss, or only do 1 morale check no matter how many times loss?) I assume only doing 1 morale check now because it may be more simple for game program. And, here comes another related question. As we know, your unit may be disrupted by (multiple) opportunity fire when moving. So, is the morale check when moving based on the every loss or total loss? I assume doing morale check for every opportunity fire. But I have no real answers for those questions.

---

Even if morale check for every loss in the last enemy turn, I may still fire with whole stack at most conditions. Because:

  • It may be no more than 10% difference in morale check in most occasion. (Lower loss. Less amount of firing units.) But,
  • Opportunity Fire from enemy may bring more loss as well as more opportunity to disrupt your firing units. Should not be ignored.
  • Less mouse clicks. My right wrist will be aching when replying too many game files and clicking too many in hours! Not a joke. It did happen twice.

---

That's my opinion. Your comments are welcome.

_________________
BG Ashdoll Ren
3rd Division
II Corps / Army of Northern Virginia


"Days and weeks of sheer boredom, interspersed with times of stark terror!"

Image


Top
 Profile Send private message  
Reply with quote  
PostPosted: Mon Oct 11, 2021 8:16 am 
Offline

Joined: Mon Aug 30, 2021 11:28 am
Posts: 71
Hi Ashdoll,

I agree that ambiguity remains as to how the loss, kill and morale checks work. Based on the portion of the manual I quoted though, I understand it clearly that every fire result that results in a loss causes a gun, kill or possible morale check, regardless if it occurs during offensive or defensive fire. The behavior I observe in the game supports this conclusion. When you split up offensive fire into individual units, the program obviously makes immediate loss checks against the target units. We know this because we can immediately see the results. What we don't see though is whether the loss causes a morale check. What we also don't know is whether a unit can be subjected to more than one morale check, or whether, as I think you were getting at, whether a unit can be subject to only one morale check regardless of how many fire combat results would normally cause it to check morale. IE if I fire on a unit three times and three of those fires cause losses but only two of those losses cause morale checks, does the program make two morale checks for the target unit or only one? If a unit can be forced to make multiple morale checks per turn, this makes splitting fire up individually even more attractive.

In answer to your other question, it seems that morale checks during defensive fire are based on each individual fire combat loss. I tested this by shuffling units around in front of artillery. You can observe how a unit that fails a morale check becomes disrupted immediately during the movement phase / offensive player's turn. This raises another question however: though the manual doesn't state it, we know now that units that take a morale check from offensive fire and pass it still become disrupted. Do units that pass a morale check during defensive fire also become disrupted?

I agree with you completely that it makes poor sense to provoke more opportunity fire from bigger units in the hopes of getting a few percentage points more chance for a morale check. It makes most sense to split up fire against infantry when using artillery safe in the rear. However these tactics would shine even more using manual defensive fire; you could fire your units as stacks instead of the AI firing individually, and there is no risk of return fire. You also know immediately if you disrupted an attacking unit. It's worth remembering that a disrupted unit fires at 50% strength, is more susceptible to failing future morale checks, and cannot assault. This can put a crimp on an opponent's attack plan, especially if their big stacks are caught in a bad spot and can't melee their way out. Not only do they take full strength defensive fire, but they also get to immediately suffer a round of full strength offensive fire before getting to return fire at 50% strength.

_________________
WDS: Antietam, Chickamauga, Gettysburg, Overland


Top
 Profile Send private message  
Reply with quote  
PostPosted: Mon Oct 11, 2021 11:18 am 
Offline
User avatar

Joined: Mon Jul 19, 2021 4:14 pm
Posts: 75
For who care about how Morale Check happen,

there is an error in this post, please check my newer reply post.


---

Just for clarification.

---

Morale Check 1 & 2


There are two checks I ​may call both of them morale check. (Different from the rules. Just personal custom.)

  • Morale Check 1 : Determine if the morale check 2 happen?

    Random (0-1) Against Loss/(Loss+25)

    Or, it can be directly regarded the Loss/(Loss+25) as the probability of Morale Check 2 happening.

  • Morale Check 2 : Determine disrupted or routed.

    Random (1-6) Against the Quality value

---

Morale Check during the turn & at the end


I believe there is two time points for Morale check 1 and 2 and they are different in results when passing or not.

  • Every time immediately for playing players' units being fired at (Defensive Fire Loss). I call it 'during the turn'.
  • In the end of playing players' turn for un-playing players' units (All loss they suffered in the turn.). I call it 'at the end'.

    It can be seen in the report dialog when un-playing player begin his turn, like '4 units routed out of 20 units checked'.


Notice that both sides, playing players' units and un-playing players' units, are checked in different times.


---
Q1

Do the Morale Check 1 during the turn happen multiple times? Yes. Every time for every loss.

Do the Morale Check 1 at the end happen multiple times? No answer. But I believe only once at the end for every unit and is based on the whole loss.Yes, it is totally same as 'during the turn'. Every time for every loss. And, in fact, it happens when loss instead of at the end. There is actually NO the kind of Morale Check 1 at the end.



The calculation and discussion in the first and second post is completely just about Morale Check 1. It is about if the multiple times losses cause multiple Morale Check 1.

The manual wrote:

Quote:
When a unit suffers casualties due to combat, it will take a Morale Check to see if it Routs or becomes Disrupted at the end of the Phase.


Sadly, I just found the manual last updated in December 15, 2012. Hope it didn't change in the related parts.

What is modified by the words 'at the end'? It is subtle whether Morale Check 1 and 2 at the end happen immediately or at the end? I believe at the end.

I didn't find an answer if the multiple times losses cause multiple Morale Check 1. You wrote:

Quote:
Based on the portion of the manual I quoted though, I understand it clearly that every fire result that results in a loss causes a gun, kill or possible morale check, regardless if it occurs during offensive or defensive fire.


Could you quote the portion directly? I may miss something. And,

Quote:
When you split up offensive fire into individual units, the program obviously makes immediate loss checks against the target units. We know this because we can immediately see the results. What we don't see though is whether the loss causes a morale check.


Yes, the loss is solved immediately. But does the Morale Check 1 happen immediately or later? Given my quote 'at the end of the Phase', I believe it is solved later. But, we still don't have an answer in our question if the multiple times losses cause multiple Morale Check 1. I still believe only once.

---
Q2

(If Morale Check 1 fails...)

Do the Morale Check 2 during the turn happen multiple times? Yes. Every time for every loss.

Do the Morale Check 2 at the end happen multiple times? I believe no.No. Tested.


You came up with a new and different question in the third post.

Quote:
What we also don't know is whether a unit can be subjected to more than one morale check, or whether, as I think you were getting at, whether a unit can be subject to only one morale check regardless of how many fire combat results would normally cause it to check morale.


It is first based on assumption that the Morale Check 1 (at the end) can happen multiple times. Then, do the morale Check 2 (at the end) also happen multiple times? If the Morale 1 and 2 only happen 'at the end of the Phase', instead of immediately, I think it will be strange if Morale Check 2 happened multiple times. I believe the answer is NO.


---
Q3

What happened when Morale Check 2 happen during the turn? Pass(No Effect) Fail(Disrupted)

What happened when Morale Check 2 happen at the end? Pass(Disrupted) Fail(Routed)


You wrote:
Quote:
...we know now that units that take a morale check from offensive fire and pass it still become disrupted. Do units that pass a morale check during defensive fire also become disrupted?


I believe there is a clear answer in manual.

Manual:
Quote:
A unit that fails the Morale Check during a Defensive Fire Phase becomes Disrupted, while a unit that fails the Morale Check during any other Phase becomes Routed.


In Defensive Fire (Opportunity Fire), a firing&moving unit being fired at becomes disrupted because it fails in both Morale Check 1 and 2 continuously. So, units passing a Morale Check 2 in Defensive Fire won't be disrupted.


----------------

That's all my opinion, including a lot of guesses. Maybe I am totally wrong. All morale check happens immediately after every loss. LOL. Discussion is enough. It's time for game testing.

Or did someone else have a final answer?

Your comments are welcome.

_________________
BG Ashdoll Ren
3rd Division
II Corps / Army of Northern Virginia


"Days and weeks of sheer boredom, interspersed with times of stark terror!"

Image


Last edited by Ashdoll Ren on Tue Oct 12, 2021 3:20 am, edited 3 times in total.

Top
 Profile Send private message  
Reply with quote  
PostPosted: Mon Oct 11, 2021 1:02 pm 
Offline

Joined: Mon Aug 30, 2021 11:28 am
Posts: 71
Ashdoll,

Keep in mind that parts of the manual are incorrect as far as how the game works. For example, I only figured out that morale checks (MC2 in your parlance) from offensive fire automatically cause disruption if passed from observing the game. The manual is incorrect or incomplete on this point.

Q1:
The latest manuals are found on this page: https://www.johntillersoftware.com/Updates.html

Quote:
When a unit suffers casualties due to combat, it will take a Morale Check to see if it Routs or becomes Disrupted at the end of the Phase.

This isn't quite correct. I used to take this literally and assume that a morale check occurred every time a loss occurs. What actually happens is an MC1 happens every time a loss occurs. The second half of the sentence then actually describes an MC2.

Quote:
Could you quote the portion directly? I may miss something.

Yes, I'm referring to this portion of the manual that I quoted in my original post:
Quote:
Firing stacked units also results in all the casualties being added together before checks are made for whether or not a unit will check morale (see Morale) and whether or not guns are destroyed (see Combat Results) or crews are killed (see Uncrewed Artillery and Recrewing).

I agree that the MC1 is probably done at the end of the turn for offensive fire, but that's just a guess. It doesn't seem to have a practical game impact though. I have no answer to this question and unfortunately the program's workings are not visible to the player.
I disagree though that MC1 is conducted only once for the total number of losses. I think each loss incurs an MC1 during offensive fire just as it does for defensive fire.

Q2:
Quote:
Do the Morale Check 2 during the turn happen multiple times? Yes. Every time for every loss.

I think you mixed up MC1 and MC2 here or are taking the manual at face value. The manual says at one point that MC2 happens every time a loss occurs, but later the manual also says it only happens if MC1 determines it does. MC1 happens every time a loss occurs but not MC2.

Again, unfortunately the player cannot tell from the program if MC2 can happen more than once. I would not be surprised if your assumption is correct though that MC2 can only happen to a unit once.

Q3:
Again, the portion of the manual you are quoting here is incomplete or incorrect. The manual states that an MC2 at the end results in pass/rout, but in reality it results in disrupted/rout. That is why I feel it is valid to question whether an MC2 during the turn (defensive fire) results in pass/disrupted or disrupted/disrupted. Unfortunately there is no message when a unit is subjected to a morale check during defensive fire, not even in the combat results dialogue.

There may however be a way to test this relatively quickly. According to the rules, 'A' quality units can't fail morale checks without some other factor that degrades morale (disrupted, flanking, low ammo, etc.). To fail a morale check, the unit has to roll higher than its morale value. 'A' units have a morale of 6 so you can't roll higher than a 6 on d6. So if you shoot at an 'A' unit long enough in defensive fire to see whether you can get it to disrupt, then we know that MC1 from defensive fire automatically causes disruption as well.

This test should also work with 'B' units with a leader (5+1 morale).

It would make sense to me that a defensive MC1 automatically causes disruption. Otherwise there is no way to disrupt an 'A' unit with defensive fire and keep them from melee because they can't fail morale checks.

I'll try it out.

_________________
WDS: Antietam, Chickamauga, Gettysburg, Overland


Last edited by Logrus Pattern on Mon Oct 11, 2021 1:32 pm, edited 1 time in total.

Top
 Profile Send private message  
Reply with quote  
PostPosted: Mon Oct 11, 2021 1:23 pm 
Offline

Joined: Mon Aug 30, 2021 11:28 am
Posts: 71
Morale checks (MC2) from defensive fire (ie during the turn) have an outcome of pass(undisrupted)/fail(disrupted).

It was an easy test. I set up a game with phases and fired Napoleons point blank at 'B' units with leaders. Another thing I learned: when playing in phases (manual defensive fire optional rule on), the player is informed when a unit makes a morale check as a result of defensive fire. This occurs in the command report when the offensive fire phase begins immediately after the defensive fire phase.

This means then that 'A' units and 'B' units with leaders and 'C' units with leaders and flank protection cannot be disrupted with defensive fire.

_________________
WDS: Antietam, Chickamauga, Gettysburg, Overland


Top
 Profile Send private message  
Reply with quote  
PostPosted: Mon Oct 11, 2021 2:30 pm 
Offline

Joined: Mon Aug 30, 2021 11:28 am
Posts: 71
I solved another question. The program checks immediately after a loss whether a unit must make a morale check and saves the result. In other words, MC1 happens after each loss and the program saves it. MC2 then happens at the beginning of the next player's turn.

I found this out by shooting at a unit and immediately saving the file then copying the save file to another folder so the game could not overwrite it. Then I advanced the phase and checked if a morale check (MC2) had occurred. When it did, I reloaded the save game file from the other folder. Every single time it does, all the same units are subject to morale checks (they are all disrupted) when the phase is advanced. However, different results occurred each time I loaded it; sometimes no units routed, sometimes multiple units routed. So, at least during phase gameplay, MC1 is conducted and the result is saved right away after each shot resulting in a loss. MC2 is checked at the beginning of the next player's turn or phase independently of MC1.

So the last question remains open: can a unit be forced to check morale twice in the same turn or phase?


I think you might be able to prove this using the method above and by shooting at an 'A' unit with offensive fire once until you have a save where the 'A' unit makes a morale check. Then reload that save and plink at the unit with as many units as you can without raising fatigue above 300. If the 'A' unit routs, then that would indicate that units can be forced to make multiple morale checks.

My reasoning: the first morale check, since it is from offensive fire, will disrupt the 'A' unit regardless if it passes or fails. If the program checks again and takes into account the 'A' units disrupted status (its morale is now 5 instead of 6), and if the 'A' unit fails a second check, then it will rout.

This is a painful test though that might not reveal a conclusive result. Hopefully JTS can answer this question instead. I think I'm going to let this one sit.

_________________
WDS: Antietam, Chickamauga, Gettysburg, Overland


Top
 Profile Send private message  
Reply with quote  
PostPosted: Mon Oct 11, 2021 9:51 pm 
Offline
User avatar

Joined: Tue May 22, 2001 10:10 pm
Posts: 1035
Location: USA
[quote="Ashdoll Ren"]Just for clarification.
...
Sadly, I just found the manual last updated in December 15, 2012. Hope it didn't change in the related parts.
...

Manuals were updated in October 2020, see
https://johntillersoftware.com/Updates.html

You can download the manuals and replace those in your games.
I understand the updated manuals will be part of the next set of updates.

_________________
Gen. Ken Miller

Image

The McKeesport Union Guard

3/1/II
AotP


Top
 Profile Send private message  
Reply with quote  
PostPosted: Tue Oct 12, 2021 12:16 am 
Offline
User avatar

Joined: Mon Jul 19, 2021 4:14 pm
Posts: 75
Thank Logrus and krmiller for the link, I will check it later.

Thank Logrus for his good job.


---

I test as Logrus said and I believe Logrus are right.

Logrus Pattern wrote:
I solved another question. The program checks immediately after a loss whether a unit must make a morale check and saves the result. In other words, MC1 happens after each loss and the program saves it. MC2 then happens at the beginning of the next player's turn.


Based on the conclusion, I believe I solve the last question. MC2 causing by others (not defensive fire ) happens only once at the end.

Testing is basically firing at a unit as many as we can.

---

Definition of MC1 and MC2 wrote:
There are two checks I ​may call both of them morale check. (Different from the rules. Personal custom, but more clear.)

Morale Check 1 : Determine if the morale check 2 happen?

Random (0-1) Against Loss/(Loss+25)

Or, it can be directly regarded the Loss/(Loss+25) as the probability of Morale Check 2 happening.

Morale Check 2 : Determine disrupted or routed.

Random (1-6) Against the Quality value


---

Test 1 Does the fatigue from the firing have an effect on the MC2 at the end immediately?

TEST 1 wrote:
Fire at a unit once.

The unit is a 990 men 'A' unit stacking with leaders.

The one shot casued a '339' men loss and Max Fatigue. And it is obviously there are only 1 MC2.

The Quality may be

Have an effect: 6(A)+1(with Leader)-2(Max Fatigue) = 5

Have no effect: 6(A)+1(with Leader) = 7


Reload the file multiple times, the unit may be routed. Conclusion is that the fatigue from the firing have an effect on the MC2 at the end immediately. It may be obvious for most players even without testing, just for preciseness.

---

Test 2 If there are multiple MC2, does the Disrupted from the former MC2 have an effect on the later MC2 at the end immediately?


Logrus Pattern wrote:
My reasoning: the first morale check, since it is from offensive fire, will disrupt the 'A' unit regardless if it passes or fails. If the program checks again and takes into account the 'A' units disrupted status (its morale is now 5 instead of 6), and if the 'A' unit fails a second check, then it will rout.


(It is obviously no need to discuss Disrupted if there is only single MC2. )

TEST 2 wrote:
Fire at a unit once as many as we can.

The unit is a 990 men 'A' unit stacking with leaders and units in both side.

There were 22 times shot, with an average of 42.7 men, to cause 940 men loss and Max Fatigue.

It can be believed that multiple MC1 fails, which may cause multiple MC2.

Quality may be

Have an effect: 6(A)+1(with Leader)+1(flank)-2(Max Fatigue)-1(Disrupted) = 5

Have no effect: 6(A)+1(with Leader)+1(flank)-2(Max Fatigue) = 6


There is no routed at least in 14 times I tested. If the answer was yes, the probability of routed would be further more than 16.7% because of disrupted and multiple MC2.

It can be believed the 'disrupted having an effect' and 'multiple MC2' CAN NOT be both ture. But 'disrupted having an effect' is meaningless for 'single MC2'. In order words, 'disrupted having an effect' is wrong.

(
If 'single MC2' is ture, 'disrupted having an effect' is meaningless.
If 'multiple MC2' is ture, 'disrupted having an effect' is wrong because of testing results.
'Single MC2' or 'multiple MC2' must be ture.
So, the 'disrupted having an effect' must be meaningless or wrong.

I believe it is a simple inference, but not sure how to express in English. Non-native. I tried my best...
)

---

Test 3 Mutiple MC2 or Single MC2?

TEST 3 wrote:
Fire at a unit once as many as we can.

The unit is a 990 men 'A' unit.

There were 24 times shot, with an average of 39.6 men, to cause 950 men loss and Max Fatigue.

In fact the unit get High Fatigue fast, we can regard the Qualify directly as: 6(A)-2(Max Fatigue) = 4.

Single MC2: The probability of routed should approach 33.3%.

Multiple MC2: The probability of routed should approach

[Tab]1-(66.7%)^n

n is the times of MC2, it should approach 15 and at least further more 5.

If n = 14, the probability is about 100%.

If n =5, the probability is about 85%.[/Tab]


Result: 3 times routed of 10. It is clear enough.

So, it can be believed there is only single MC2 at the end.

In fact, it may also be like that multiple MC2 is allowed, but no more than 2 times. But I think it is odd and unlikely. :)

---

Summary

How do MC1 and MC2 happen?

MC1 happens after every loss no matter from defensive fire or others immediately. It can happen more times and is based on every time loss.

If MC1 fails,

MC2 caused by defensive fire happens immediately. It can happen more times for one unit in one turn.
If pass, no effect. If fail, disrupted.

MC2 caused by others happens at the end of the turn and only happens once for one unit.
If pass, disrupted. If fail, routed.



* By the way, it seems no additional effect when a Disrupted unit get another Disrupted result. (So a disrupted unit can wander boldly, with limited punishment to enter 'Disrupted' Terrain or to approach enemy. ) I blieve it should have.
If the new Disrupted caused by defensive fire or terrain, the unit should use up all left movement points and stop moving immediately.
If the new Disrupted caused by others (MC2 at the end), the unit should become routed.
It seems common in boardgames, but maybe JTS rules have different considerations.

---

The Same Example:

Casualties: 16 total
1 time in 16 each, MC1: 39.0%
2 times in 8 each, MC1: 42.6%
4 times in 4 each, MC1: 44.8%
8 times in 2 each, MC1: 45.9%
16 times in 1 each, MC1: 46.6%

Casualties: 32 total
1 time in 32 each, MC1: 56.1%
2 times in 16 each, MC1: 62.8%
4 times in 8 each, MC1: 67.1%
8 times in 4 each, MC1: 69.5%
16 times in 2 each, MC1: 70.8%
32 times in 1 each, MC1: 71.4%


* It seems meaningless for over 8 times because it seems unable to stack over 8 combat unit in one stack. Also, you can seldomly have over 4 combat unit in one stack.

** It seems dividing fire to 2 times get the single increase most.

---

Tips:

Logrus Pattern is right. Because of multiple immediate MC1, firing individually can cause a higher probability of doing MC2. In other words, target may be disrupted more probably. It is basically indifferent or a slight increase (<10%) in most conditions.

Because of single MC2, target won't directly be routed more probably because of individual fire, but may be routed more probably because higher probability of doing MC2. The increase is further lower than the increase of MC1, based on units' quality.

The risk from opportunity fire increasing should not be ignored. But it may be very useful for a great amount of small units in one stack in a safe place, like small long range guns. (eg. 8 batteries of 2 guns stacking together) The more the amount (times of firing) is, the more the Morale Check increase.

No comments on the conditions of Manual Defensive Fire. Double emails...

Personally, I believe it is a bug unreal that players can get benefit in causing Morale Check from individual fire than intensive fire. It should be revised. I believe heavy loss in volleying in short time should cause panic more easily.

---

That's all. The conclusion may still have loopholes and be based on Probability & Statistics. And I still prefer intensive fire, but may try to divide some.

It is a great talk to solve one of my biggest problems about JTS ACW, how the Morale Check (1&2) happened!! I will save these posts permanently. LOL

Your comments are welcome.

_________________
BG Ashdoll Ren
3rd Division
II Corps / Army of Northern Virginia


"Days and weeks of sheer boredom, interspersed with times of stark terror!"

Image


Last edited by Ashdoll Ren on Tue Oct 12, 2021 4:42 am, edited 6 times in total.

Top
 Profile Send private message  
Reply with quote  
PostPosted: Tue Oct 12, 2021 4:13 am 
Offline

Joined: Mon Aug 30, 2021 11:28 am
Posts: 71
Hi Ashdoll,
I concur with your tests! Thank you for finding a way to test for multiple morale checks (MC2) that did not require hundreds of tests.

A last note on firing individually to try and increase your odds of causing a morale check: if a unit would only take about 15 or fewer total casualties from a stack firing at it, there is no benefit to firing the units individually. In most cases it is probably not worth taking the time, aside from other considerations.

Thinking about what we've written about in this thread, if you are willing to really micromanage your units, a player greatly familiar with the rules can do a number of things to use the rules to increase their odds. Many are not worth doing all the time, but some are worth doing in critical situations. For example, in turn play, if you really need to have your units undisrupted at the end of a move in order to do a melee, then you can move the counters individually to try to give the units the flank morale bonus as much as possible during the move. Similarly you could move a leader with as many units as possible to get that bonus. Another consideration is that each time you add a counter to melee from a new hex, it triggers final defensive fire. Having units in a stack reduces the opportunities for multiple such final defensive fires by potentially sacrificing the flank bonus.

I disagree that individual fire is a bug. I guess that it is an unintended consequence of a deliberate design decision, specifically the formula for how the probability of a morale check (MC1) is calculated versus the probability of gun losses and crew kills, as well as decisions about when to resolve the effects of fire combat in the program and some differences between phase play and turn play. The conclusions we came to require some spreadsheet math and are not apparent at first glance, at least not to me. It would not surprise me if the designer(s) had other aspects in mind when deciding to use this formula.

We could think out some solutions together but we would still have unintended consequences. Also the program unfortunately doesn't support modding it in such a way, so until we can modify the game to taste it's easiest to just understand the rules of the game as they are so that everyone is on a level playing field. I also don't think the small fan base has the numbers to support splintering up into smaller groups because people want to play mods that change the basic game behavior.

That said, I think my solution would be to check morale (MC1) at the end of a turn or phase only once based on the total casualties a unit took from all fire combat. I haven't thought the consequences through all the way.

What I like about the formula for MC1 is that unit size is not a factor. It may seem odd that an 800 strength unit is as affected by a loss of 10 men as a 100 strength unit is. But then consider eight 100 strength units in a hex constantly failing morale checks against an 800 strength unit that is near unbreakable. The rule is an abstraction that I think works well in the service of making a good game.

On the other hand, that same aspect about unit size could be used to criticize the formulas for gun loss and crew kill. Gun loss is not greatly problematic, since the size of an ACW gun crew seems fairly standard and the only oddity is that regardless of how many casualties caused by one hit, a counter can only lose one gun at a time. So extra firepower will go wasted if the player is not paying attention and overkilling. Edit: It is possible to kill more than one gun in a battery from a single fire combat result! Crew kill however seems to work the same regardless of counter strength: a one gun counter has the same chance of being killed from a loss of 1 man as a 20 gun counter does. That is more difficult to justify but that would be an extreme and unrealistic case. Since artillery losses are not tracked like infantry losses, the design shortcut seems to be to let probability do its thing and that things will shake out more or less right in the end. And I think it, like the other rules in the game, more or less works.

_________________
WDS: Antietam, Chickamauga, Gettysburg, Overland


Top
 Profile Send private message  
Reply with quote  
PostPosted: Tue Oct 12, 2021 7:19 am 
Offline

Joined: Mon Aug 30, 2021 11:28 am
Posts: 71
I guess I'm not done with this topic yet. I read through old posts and learned that it is possible to kill more than one gun in an artillery battery from a single fire result. I had not played enough to see it happen yet. This means that artillery overkill only happens if the stack would cause losses greater than the target unit's strength so as to eliminate it.

Even if overkill occurs, there is still a consolation prize. When I tried multi-gun kills to see for myself, I learned this as well: if a unit is eliminated, losses in excess of the eliminated unit's strength are converted into fatigue and an additional unit in the target hex is given that fatigue (but no losses). This makes stack fire more attractive.

One last consideration: when using stack fire against an overstacked target hex, every single firing unit gets the same firing bonus from the target being overstacked. If you were instead to fire on the overstacked hex individually, each successive firing unit would get an ever decreasing overstack bonus for its fire. Where an overstacked target hex is concerned, firing units as a stack will on average result in more casualties than firing the units individually would.

_________________
WDS: Antietam, Chickamauga, Gettysburg, Overland


Top
 Profile Send private message  
Reply with quote  
PostPosted: Sat Oct 16, 2021 4:54 pm 
Offline

Joined: Tue Nov 11, 2003 9:52 am
Posts: 1324
Wow! Don't think I have seen an analysis this comprehensive since the Tavern was opened!

_________________
MG Mike Mihalik
Forrest's Cavalry Corps
AoWest/CSA


Top
 Profile Send private message  
Reply with quote  
PostPosted: Sun Oct 17, 2021 8:25 am 
Offline
User avatar

Joined: Tue May 22, 2001 10:10 pm
Posts: 1035
Location: USA
mihalik wrote:
Wow! Don't think I have seen an analysis this comprehensive since the Tavern was opened!

Mike,

I recall seeing similar articles on the old board games in the AH General, so called best set ups and first moves.

There wasn't much talk about fire in the BG games because it was all phase mode and Morale Checks were made anytime you took a SP loss while they became random in the HSP/JTS games and Turn Mode was added putting in new variables into the game. The biggest concern was the AI taking low odds long range shots, especially with artillery.

I have a paper from the Union War College, a very good study of Melee in the BG system by Bob Breen who retired a long time ago that I shared with my cadets when I was a training officer at the Academy and several others on the shorter scenarios. So there were some studies on those games.

I could rant about how the bean counters have ruined everything but that discussion belongs in the Smoking Room. :mrgreen:
Lets just say they call it the Art of War and let it go at that.

_________________
Gen. Ken Miller

Image

The McKeesport Union Guard

3/1/II
AotP


Top
 Profile Send private message  
Reply with quote  
PostPosted: Mon Oct 18, 2021 2:03 pm 
Offline

Joined: Mon Aug 30, 2021 11:28 am
Posts: 71
I was thinking about posting something similar on melee, but then saw to my embarrassment that along with melee, much of what has been covered here on fire combat has also in one form or another been covered either in the forum or the Union War College. I didn't intend to reinvent the wheel.

The melee articles are good and current and already cover the essentials as well as the details. So long as someone can roughly judge the basic probability of melee success, which has been given elsewhere, I would only address what my biggest misconceptions as an absolute beginner were. I didn't realize that hex terrain does not affect melee combat (hexside terrain does) and that neither disruption nor fatigue affect the defender in melee.

I've got a spreadsheet that I can type some numbers into and it spits out expected combat results. As I was programming it I too had my reservations about whether it reduced the game to a deterministic game of numbers. But as I did, I learned so much about the game rules that it became apparent to me that I had been playing with a false understanding of the rules in many respects. I realized I wasn't able to tell how big the hill of beans was until I had counted all the beans at least once.

I think the game is complex enough, especially against a human opponent, that it can't be reduced to a mere game of numbers. I don't calculate every fire combat in advance to determine the best shot. Instead I use a couple of rules of thumb to judge whether a particular engagement is worth it versus letting a unit run low on ammo for the sake of a few kills. I think even for someone playing for pure fun such rules of thumb are necessary in order for an otherwise tactically skilled player not to be punished by game rules that are sometimes confusing and not intuitive. Likewise being able to calculate the melee probability down to a tenth of a percent hasn't helped me when I've played poorly otherwise.

The extreme fog of war rule greatly cuts down on bean counting. It seems it is not so popular, but I really like it for exactly that reason. The only real way to judge the melee odds is to see how badly your troops get chewed up by the enemy's defensive fire. I think it has made me play much more conservatively when I've used it.

_________________
WDS: Antietam, Chickamauga, Gettysburg, Overland


Top
 Profile Send private message  
Reply with quote  
PostPosted: Mon Oct 18, 2021 4:53 pm 
Offline

Joined: Tue Nov 11, 2003 9:52 am
Posts: 1324
krmiller_usa wrote:
Mike,

I recall seeing similar articles on the old board games in the AH General, so called best set ups and first moves.

There wasn't much talk about fire in the BG games because it was all phase mode and Morale Checks were made anytime you took a SP loss while they became random in the HSP/JTS games and Turn Mode was added putting in new variables into the game. The biggest concern was the AI taking low odds long range shots, especially with artillery.

I have a paper from the Union War College, a very good study of Melee in the BG system by Bob Breen who retired a long time ago that I shared with my cadets when I was a training officer at the Academy and several others on the shorter scenarios. So there were some studies on those games.

I could rant about how the bean counters have ruined everything but that discussion belongs in the Smoking Room. :mrgreen:
Lets just say they call it the Art of War and let it go at that.


Hi, Ken,

I do remember the analyses in the General, but the variables were much fewer in those old board games.

We gave up playing turn a long time ago. It is one thing to have a system that is unpredictable, quite another where things frequently happen that defy logic.

I do think the turn concept is superior, but its implementation in the game needs some work. My humble opinion.

_________________
MG Mike Mihalik
Forrest's Cavalry Corps
AoWest/CSA


Top
 Profile Send private message  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 15 posts ] 

All times are UTC - 5 hours


Who is online

Users browsing this forum: No registered users and 118 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group