Combat Results Wearing me out

[Adam the VIth]

Okay -- look, I know I am going to get hammered here, but remember, I do love this game, I've invested my time and money in it -- I've tried to make suggestions and help out......but.....

I'm starting to have my doubts about the combat results.

I'll start posting weird results here.....I can make saves available to Pocus to be looked at.

1:
CSA 95 elements (including 20 10-20lb parrott)
USA 43 elements

Three different battles (in the same 15 day turn).
Battle 1: 402/828 casualty ratio (CSA/USA) 47/40 luck roll
CSA had 27 elements crossing minor river, USA was 392 dug in.
CSA was in offensive mode, USA defensive,
AJ Johnston commanding CSA. Halleck, USA.

Okay, result made sense, CSA inflicted most losses in range fire (lots of arty), took a bunch of casualties in assault. Halleck sucks, etc.

Battle 2:
10/681 casualty ratio, all casualties in ranged fire, now Reynolds commanding USA (much better commander for them) luck roll 50/0 (you can get a zero?), same off/def. Now 76 CSA elements crossing river.

Battle 3: (now CSA is trying to retreat, despite 2:1 odds and defensive orders)
8193/1612 casualty ratio, lots in both phases.
now USA is off, CSA def. CSA now dug in to 32 (not much), USA 0.
Luck roll 45/49 in favor of USA.
now 49 CSA units crossing river.

Okay, so in essence, over the 15 days, one can envision the CSA force surrounding the USA, begining to dig in and fighting some battles, duelling arty, etc. Then, that last result makes no sense. Then the Union comes out of its position and attacks at odds of worse than 1:2 and whips the CSA's ass to a 4:1 ratio in casualties?

I don't know how one can explain that. Sure, I can think of ways the defenders can pull off a great attack and all, but also inflicting 4:1 in casualties?

Perhaps this river crossing thing is causing problems. Look, the CSA crossed the river once. Not three times. THe first battle could potentially have that as a factor, but somehow the engine has to adjust and realize they crossed 8 days ago and should no longer be penalized.

I love the game, but almost every turn, I get a "bad" battle result. I know about the Civil War....there were not battles with casualty ratios like this....Even Fredericksburg, casualties were 3:1, and that was with idiots charging dug in troops, uphill and out of supporting artillery range.

Comments, similar results? I'll admit, I'm getting frustrated......I'd like to see if perhaps the rivers, or luck, or something is getting too strong a factor in battle results.

AC

[runyan99]

Not sure about your specific example, but generally I find the battle results to be quite good. In fact of all the aspects of the game that I have been scrutinizing, I think the battle results are one area where I have almost always found the results reasonable. I certainly don't see a 'bad' result every turn. Generally, all of my battle results seem plausible.

[bloodybucket]

If there was a turn log that listed battle results, then they could be easily evaluated and compared. Right now, my impressions are strictly anecdotal, but the battles seem OK to me.

[McNaughton]

I think the last battle, where for whatever reason, the CSA forces retreated and were caught completely off guard by a Union counter attack. For whatever reason the commander pulled out. The Union forces attacked because the Confederates were in retreat. It could be quite possible that the large numbers of artillery actually proved a negative factor (20% fo the entire force), since their evasion and movement factor is low, slowing down the force significantly. While I don't have a complete understanding of battles, I think that these factors, based on initiative, speed, evasion, etc., have intense ability to affect the situation. It isn't a matter of forces providing a point value for attack or defense, but, cavalry, infantry and artillery all have their strengths and weaknesses.

What was the composition of the Union force? How much cavalry did they have? Cavalry is the opposite of artillery, effective in close, high initiative and evasion. Cavalry can savage artillery.

What forces did you find suffered the greatest casualties? Was it your infantry, artillery or cavalry?

To me, it seems like you had an over-proportionate amount of artillery, which actually created a weakness in your position (makes your force slow to react on the battlefield, giving initiative to the Union, hence the successful counterattack).

Also, taking a look at the generals I see great disparity.

Reynolds has stats of 4-4-4 (very good stats actually!)
A.S. Johnston has stats of 4-2-1 (average, but poor for Confederates)

You can see that Johnston is extremely outmatched when it comes to his ability to defend against Reynolds (4:1). Johnston's impact on the defending force is minimal, while Reynold's impact on his attacking force is phenominal. The 2:1 odds were probably rated as significantly closer to 1:1 primarily because of leader ability (don't know if this is calculated into odds).

The high amount of artillery (low initiative and evasion), a river the CS had to cross to attack and retreat, great disparity in leader quality when Reynolds takes command, lack of CS entrenchment all conspired to doom the Confederate attack, even with outstanding numbers.

[aristoteles]

You may check your [color="RoyalBlue"]Battle log.txt[/color] in your [color="SandyBrown"]LOGS[/color] folder and see why that happened.

[Queeg]

Those results don't look odd to me.

CSA attacks across a river, inflicts some casualties but doesn't capture the USA entrenchments. USA counterattacks with a better general, one with especially good attack abilities. CSA has to retreat back across the river and takes considerable losses in the process.

Frankly, I'd be concerned if the results did NOT turn out like that.

[mike1962]

Yeah, when I encounter I lopsided battle, there is usually a good cause.

[Pocus]

the battle log (in full debug) is your friend here, but beware, in full debug (named error reporting in the options window), you can get several Mb of actions!

About river crossing, the rule is this one: you won't get the penalty if you have a force already in the region and you have 11% military control (meaning you have a small bridgehead and the force already there allow you to deploy somehow correctly). Now, thinking about it, perhaps you should not get the penalty either if you have say 51% military control in the region, meaning you achieved a sufficient battle result to have crossed the river (in a previous fight) and are now correctly deployed on the other side.

And in a fight, you get military control (representing gaining ground) when you fight on offensive and inflicts more losses to the enemy than you suffer (you are pushing him).

[mikee64]

OK, here's one I'd like some feedback on; this is using 1.05g.

First battle of Manassas in a pbem:



PGT was defending with army HQ and some infantry (no CP penalties) from inside the town; Bonham and Holmes each had a corp under their command with no CP penalties outside of town. All were defending except for Holmes who was set to attack posture.
As you can see the initial global combat values were 1341:846 in favor of the CS, who also got a 181 entrench bonus. Dice roll was 49-48 in favor of CS; the CS failed 3 morale checks as opposed to 27 for the US. Both sides opened fire at range 5; the artillery was primarily 12 pounders (CS) vs. 10 lb. Parrotts.

The CS did have a large number of sub-units not commanded, due to the presence of only the 3 high level leaders. Other than that, I'm not sure what happened here.

In general I think the combat engine has done a pretty amazing job given all the variables involved; I've certainly had some good and bad luck in battles with an occasional outlier that still seem rational. This one stood out as a bit odd nonetheless.

Any ideas?

Edit: Bonham and Holmes retreated, but PGT held the town. The battle happened day 5. On day 6 a large bgde moved into the region, remaining outside the town. On day 6 I also got a message "We failed to retreat before battle" in the region, but there were no follow up assaults by McDowell.

I'll post the next turns follow up battles next.

[Bodders]

Hehe, I had a 14 day battle to take Chancellorsville in a 1.05g game against the ai - in total we both lost about 60 000 troops and I took it eventually! The results do seem to get odd when any of the defenders happen to be inside the town and the opponent has 'assault' posture. Maybe they should be allowed to retreat to another area and only get stuck inside forts or maybe for battle resolution, assault should be modified to 'attack' in town spaces?

[mikee64]

Reorganizing from Fredericksburg, Bonham and Holmes are sent back into Manassas with totally fresh commands. Johnston is also sent in via rail from Harpers Ferry. They arrive in a coordinated attack on Day 7, bringing a GCV of 1546 to McDowell's 977:



The battle continues 5 days later, GCV is now 825:321 CS troops continue to melt away:



And finally, despite being down by a GCV of 643:86, McDowell still inflicts 5000 casualties while suffering only 1000:



By now I'm just confused, and I swear the only thing I have had to drink today is coffee!

[mikee64]

The results do seem to get odd when any of the defenders happen to be inside the town and the opponent has 'assault' posture.

Could be something like this, even after all the losses PGT and his small force of attached infantry remained inside the town, unhurt.

[Bodders]

It's true, that one makes no sense at all

Supposedly none of your units are inside the structure, according to the tooltip, which was my only thought that somehow you were getting a limited frontage.

It's that Union assault posture I tell 'ya, mucks things up

[Bodders]

Could be something like this, even after all the losses PGT and his small force of attached infantry remained inside the town, unhurt.

Yes, in my battle, which has sadly been consigned to the permanently deleted bin now, PGT kept retreating in to the town - day after day for 14 days of assaults on him

[Pocus]

First battle don't make much sense, I agree.
Second one, you are crossing the river.

I'm requesting your save by mail.

[mikee64]

save files sent... does the river crossing affect all three of the battles on the second turn? Because the symbol is only shown for the first one.

[Pocus]

got your mail (but I'm unsure I will check the battle before I return from vacations, scrutinizing the battle log is a rather time expensive thing!).
About the battles, if the symbol is only shown in the first battle, then the others are not using river crossing. Sorry to be more precise

[Zoetermeer]

I have to agree with Adam. I don't know much about the inner workings of the combat system and I don't have any proof either, but I do see some wild combat loss ratios in this game. I think lowering the hit coefficient alleviated the problem of losing massive numbers of forces in a lopsided defeat, but it didn't really change the fact that those defeats are still excessively lopsided.

I get the feeling that the overall numbers of casualties in battles are generally pretty good (about what you'd expect from a garden-variety Civil War battle given the number of men involved). I'm not an expert on the Napoleonic wars, but the casualty ratios seem a little better suited to that era than the Civil War.

[Jacek]

Well, casualty ratio is fine with me. In every battle 33% of losses goes back into your pool. So dead, wounded and priosoners make the rest. Now, when you keep that in mind, then the battle results in terms of losses make sense. The different story is the unit destruction thing - it happens too often sometimes. I think it is driven by the NM system - in AACW most NM points are collected by players by destructing opponents' units, so the combat system "helps" both parties to achieve that. Imagine playing AACW with very loooow hit coefficient - this game could go for ages because none of the parties could earn enough NM points in battles to win the war !