Commitment - I think it is broken

[Berkut]

Just returned to the game after the last patch, hoping that it was a bit more playable as a PBEM game between two good opponents, and in a manner that reflects the actual war well.

My opponent sends Lee and most of his troops West, and we have a great fight in Paducah - attacking Union forces are (barely) repulsed after a hard fight. But I think, hey, he sent a LOT of troops West, I shall attack in the East with my large numeric superiority!

I have about 80 total troops, almost all trained (maybe 4 militia or something), a decent chunk of arty, and since he sent all his cavalry East, Manassas is actually 100% scouted (this is in 1862, having forgone an 1861 Eastern offensive).

He has 21 scouted troops in Manassas, plus some number behind and within reaction range. I figure I can get 2-1 odds after his reaction, which may not get a win with my poor leaders (Sherman is AC, but only a couple good Corps commanders, of course - most of the rest are mediocre at best) but at least will get a bloody fight and start bleeding the ANV and force him to either shift forces east or risk ANV being overrun.

Battle result:

18k Union losses, 6k Confederate.

Holy crap! What happened?

Turns out that out of my 80 or so troops, less than 1/3rd actually fought, 24 or so. He was able to react about 20 troops in (of course), and out of his 35-40 troops, all of them fought. So it ended up a 35 against 26 fight, with predictable results (considering his Southern Uber leaders don't get me started on THAT gross myth!).

So why is this? Well, I think the problem is that commitment is over-modelled. You get penalized for not having many MPs left (Oops! Crossing a river! Oops! union generals have relatively poor movement anyway! Oops! Defender is moving close to home/in home territory! Oops! Union leaders have poor leadership, so have yet another chance to fail to commit!).

I went into an attack with about 4-1 odds. His PanzerCorps Southern troops all teleport from all around Virginia into the space, spending less MPs than I spent to attack (even though I am the attacker) and made the fight 2-1. Then when commitment time comes along, all his troops fight, and only 25-35% of mine, making the fight actually in his favor by about 3:2, even though he had to move to my point of attack!

I don't know how to describe this other than broken. The strategic defense, coupled with the southern mythological leaders that are so much better than the northern ones (note that he only needed 2-3 infantry leaders to do all this, and of course they are his excellent leaders, compared to me needing probably eight or so, since I have more troops and lower command rating leaders) means that you simply cannot have anything like historic campaigns in Virginia.

I think this could be fixed simply by reducing the influence MPs have on commitment, at least for the attacker. It makes sense that a reacting force from outside the space attempting to react to an enemy attack they did not know was coming (or where it was coming) would be very limited by MPs, but the attacker should not be, since they know where they are attacking. It should matter somewhat of course, there were meeting engagements at times - but not all the time, and not to the degree we see here. Maybe have a penalty for getting to the fight with no MPs left, and a bonus forbeing there with more than half, but otherwise pretty much no effect?

We re-ran it again to see if it was just bad luck, but had the same result.

[Erik Rutins]

Who was the CSA general? What was Sherman's effective AC Attack rating and his CPs compared to the CSA Commander? Did your Theater Commander also have initiative on both sides?

In any case, I don't think it qualifies as broken at all. Whenever the CSA was able to focus on a single Union thrust, they generally came out on top. Also, in 1862 their leadership was generally better and the Union attacking across a river or in bad terrain certainly did have problems getting all its overwhelming forces into battle (Fredericksburg, Antietam, Chancellorsville, etc.).

As the Union, when you attack into Virginia, you need to attack on more than one axis. Make the CSA choose where to send their reacting forces rather than allowing them to concentrate them on one attack. Attack down the Shenandoah while you're attacking towards Richmond and consider an amphibious strike or an attack out of Fort Monroe at the same time too. Sure, this requires preparation and waiting for initiative, but it works and historically things started to fall apart for the CSA when they had too many threats to deal with at once and Union leadership had risen to the task.

I don't think that a Union player should be at all surprised at being defeated in 1862 by a smaller CSA force. With that said, I have won victories in 1862 in Virginia as the Union against a CSA human player and the AI, it just takes a good bit of effort and a plan that involves diversionary attacks. On a strategic scale, most of the first two years of battles in the East were effectively attritional for the Union rather than making any real progress and many of the later ones were the same. The CSA was historically outnumbered quite badly on paper in many of these battles too. Nothing mythological about that.

Regards,

- Erik

[Berkut]

18k to 6k losses is not attritional though - at all. In fact, those kinds of casualties were almost unheard of in the Civil War. And while the Union did have their setbacks (and Frederiskcburg had nothing to do with the river, but rather with attacking into prepared defenses), they almost never saw those kinds of losses, and in fact had their wins and draws as well.

I was not surprised I lost - I was surprised that the outcome was so lopsided, which suggests that there is not really any chance of a win there, or even a draw, or even a marginal loss that inflicts decent casualties. Rather the attack is literally hopeless, even with a LOT more troops.

I would argue the claim that overall the South had better commanders from top to bottom as well, but understand that is a likely necessary fudge for gameplay balance purposes.

Attacking on more than one axis sounds great, but doesn't actually work against a competent opponent who knows he can win fights even if he is outnumbered 2-1. He can just accept those odds at multiple points, and stomp the attack out, again with huge losses. So I send some troops off to attack on another axis - then there is that many fewer attacking on the main axis, and both forces still commit some paltry percentage of their troops to battle, while the South and their Uber Generals commit all of theirs, and you get the same outcome, or even worse.

In this game, a "On to Richmond" kind of attack is basically not possible. You cannot take on the ANV, even if you have triple their numbers, and even if by "taking them on" you are just hoping for a losing fight that you inflict somewhat historical casualty ratios in - that is simply not accurate. It should at least be *possible*, the South should at least be forced to deal with that possibility. Right now, they can denude Virginia with impunity, and focus on the West.

And this is largely because of the commitment rules, which essentially amount to the north cannot use more than 25-25% of their troops when attacking, while the South uses 100% of theirs.

Again, please do not respond with "You should not be surprised you lose..." because I am not. I am surprised that the South can move the bulk of their troops west, and STILL not just win, but crush a Union attack into Manassas. I am even more surprised at the arguemnt that this is how it actually was - that the South did not even need Lee to win, because the Northern defeat was built into the very fabric of the world.

[Capt Cliff]

I tend to agree with Erik on this. I too have won at Manassas in 62' but with Halleck as AC, but there I stopped. Marse Robert is a fierce defender. Now in the west Grant is running wild! He wins with 1:1.5 odds!! All his boy's fight. Also just because your guy's show as regular infantry don't mean they are vet's, they just can march in a straight line. Many of the southern units maybe one star or two star units and fight better, shoot first and do more damage. Also remember your crossing a river into Manassas and from Manassas into Fredricksburg which makes attacking there tough. Attacking into the Valley when your going for Manassas is good tactic's, an ampib into Norfolk as well even better.

Best strategy ... hit them where they ain't and get there firstest with the mostest.

[Berkut]

ORIGINAL: Erik Rutins

Who was the CSA general? What was Sherman's effective AC Attack rating and his CPs compared to the CSA Commander? Did your Theater Commander also have initiative on both sides?

Shermans had 11 CPs, IIRC. I would have to check at home. I would be happy to send you the save file before the battle.

CSA general was not Lee - JJ Johnston. If it had been Lee, I would not even bother to attack - the entire point of going on the offensive was that he had more troops in Kentucky than Virginia, AND Lee was sent west.

So this isn't even comparable to history, where the likes of McClellan and Pope are going up against Lee with the bulk of the Confederate forces. It is the second tier Confederate leadership with 1/3rd fewer troops going up against the AoP with as good leadership as I can get it, and a huge number of troops.

If even in the best possible circumstances the game system will not allow even a draw...I don't think that is right. Even with the South focuses on the East to the detriment of the West (pretty much the opposite of what we have here), the South still had their share of problems checking the Union, and certainly did not routinely just crush any offensive with trivial casualties.

[Berkut]

ORIGINAL: Capt Cliff

I tend to agree with Erik on this. I too have won at Manassas in 62' but with Halleck as AC, but there I stopped. Marse Robert is a fierce defender. Now in the west Grant is running wild! He wins with 1:1.5 odds!! All his boy's fight. Also just because your guy's show as regular infantry don't mean they are vet's, they just can march in a straight line. Many of the southern units maybe one star or two star units and fight better, shoot first and do more damage. Also remember your crossing a river into Manassas and from Manassas into Fredricksburg which makes attacking there tough. Attacking into the Valley when your going for Manassas is good tactic's, an ampib into Norfolk as well even better.

Best strategy ... hit them where they ain't and get there firstest with the mostest.

It isn't possible to get their firstest (the defender can move further than the attacked, even when the attacker knows where he will attack and the defender has to react), and the mostest doesn't matter since they wont fight anyway.

Interesting enough, Fredericksburg is often cited as the example of what happens when crummy commanders attack across a river into Lee. but lets look at that battle a little more closely.

The river actually did not matter. The Union spent two days getting their entire army across the river and deployed for battle after securing the beachhead. The South could not contest the beachead because of the Union gunboats and artillery. By the time the battle opened at Marys heights, the river was irrelevant, or even a benefit to the Union, since it meant they could not be flanked and always had support there.

Further, the Union, at this scale, certainly got their entire force into the fight - 70% did not decide not to involve themselves. The north had about a 1.5:1 force advantage in troops engaged, nowhere near the 3-1 I had.

Fredericksburg was a disaster because the Union tried to attack uphill into prepared defenses, and that never works unless you have very overwhelming numbers, as Lee himself would learn the next year.

I certainly hope the southern units do not fight better - the myth of "Ten rebs can take on one Yankee!" was widely regarded as myth even then. Neither army had been blooded yet, so presumably we were largely similar - if anything, I had more trained troops, since I think he sent a lot of his West.

[ckammp]

deleted

[Berkut]

The number of losses is not the issue, the 3:1 loss ratio is, and the fact that if we re-fight that battle 10 times, it will almost certainly be a crushing defeat every time. The problem is not the losing is possible, it is that losing, and losing extremely badly, is inevitable.

And again, this is against a non-historical West first southern strategy. No Lee. Half the troops.

Basically, what is being argued here is that there is no reason for the Southern player to put Lee and historical force levels in Virginia - he can defend it just fine with considerably less forces, and simply send them all west, because the North is Just That Bad, and cannot possibly attack successfully, even against a weakened Southern defense.

From a gameplay perspective, this is broken. The Southern player can strip the East with impunity, because the game system itself grossly punishes any attacker with mediocre leadership at the Corps level with compounded commitment checks meaning that even with MUCH greater numbers, it won't matter, since they aren't going to fight anyway.

Of course, if you have the Southern mythological uber leaders that can lead twice as many troops and certainly commit them all, this does not apply.

Anyway, if the consensus is that there is nothing wrong with these kinds of results (South strips the East, North reinforces the East, still gets utterly crushed with no chance of a draw even), then there really isn't any room to debate the issue anyway - nothing is going to change.

[Erik Rutins]

Berkut,

A lot of folks have played this game (including me) and gotten very realistic results. I don't think 3:1 losses are normal for ACW battles and they are certainly not normal in this game either. The rest however is pretty darn historical and it could just be that for some reason that one battle went very poorly for you. Try playing out the whole war rather than worrying about one fight and see if you still feel the same way. If you want, e-mail me your save file to erikr@matrixgames.com and I'll take a look.

Regards,

- Erik

[Joel Billings]

We should take a look at the save to see if we can explain the results. I don't think the commitment issue is the problem, although I'd have to see the save (you can send it to 2by3@2by3games.com as well as to Erik). Sounds like the Southern commander had many more Command Points and set up the battle in his favor. If you run it 10 times you will probably get a large spread of results. That said, I do agree that if it was easy for us to tweak the model, then one thing I would do is try to have the results come in only very rarely with the large differential in losses. The only way this can be explained is by the very rare Fredericksburg situation or by a surrender situation (like Harper's Ferry or Fort Donelson). I like to see these kinds of results as the result of a corps getting cut off and forced to surrender, and since it did happen in at least these two cases, it is a valid result. However, I'd say it happens a little more often in the game then I would prefer. The combat results system though does not allow us to do this kind of tweak, and overall I'd say the system works well. The east is a tough nut to crack because one good leader can use interior lines to keep a larger force from making headway. Multiple attacks on the same turn are the key, but as in the real war, victory is almost never found in the East, but rather in the west and via the coasts.

[Berkut]

ORIGINAL: Erik Rutins

Berkut,

A lot of folks have played this game (including me) and gotten very realistic results. I don't think 3:1 losses are normal for ACW battles and they are certainly not normal in this game either. The rest however is pretty darn historical and it could just be that for some reason that one battle went very poorly for you. Try playing out the whole war rather than worrying about one fight and see if you still feel the same way. If you want, e-mail me your save file to erikr@matrixgames.com and I'll take a look.

Regards,

- Erik

Erik, *I* have played this game and gotten very realistic results as well! Yeah, overall it probably works out, even with rather odd results in places, and I can understand the idea that fixing this (if in fact it needs fixing) could possible effect other things negatively.

But the Eastern Campaign, even if it was never as decisive as either side wanted it to be, is still crtically important. While it was not the decisive front the North wanted, the fact that it could have been, and the fact that both sides thought it was, throughout the war (and an argument can actually be made that it was, and that the South largely stymied the Northern efforts to win there) was critical to how the war played out.

What bothers me about this kind of result is that it is telling the Union player to not bother. There is no reason to attack here, because you won't achieve even attritional results - the South can block it off with a relatively small amount of resources (especially in leadership).

I guess to some extent this is the problem of hindsight.

I would like to note that I do NOT think that the commitment model in general is bad - quite the opposite in fact. My suspicion however is that there is kind of a statistical certainty to the failure to commit - that the variables that go into commitment are "checked" often enough that it becomes not very variable at all, since a combination of factors combine to weight the outcome very heavily, so that the Union troops coming across the river will have extremely low commitment rates due to the combination of movement points lost from the river, low command ratings with many leaders (which is a double whammy, since it means the check itself is modified AND then number of movement points, which is also part of the check, is also lower), and the ability of the the defender to move so easily in response. One excellent leader (Jackson, for example, or Longstreet) is really worth 3xtimes, since they can typically command 10 or 11 troops to the Union leaders only commanding 3-5. That means from a commitment perspective, the excellent Southern corps commanders become incredibly potent, literally several times the value of their northern counterparts.

I will send the save file when I get home tonight. Perhaps there is something obvious I am doing wrong. We did re-run the battle with the same results (I think the 2nd run was 5k to 15k casualties, with similar commitment levels - 100% Southern, 1/4th-1/3rd Northern).

I don't even mind that results like this are possible - again, it should be possible to just get your tail kicked. But that should be true for both sides (and I realize that a draw may very well constitute a tail kicking of the South). maybe part of the problem is that it is so hard to tell what happened from a strictly "game theory" perspective. Did I lose badly because I proverbially threw the 1,6 result on the CRT? Or was this a typical result, and it was a mistake to even engage?

[Erik Rutins]

Please send a save from both sides so that we can see the exact army setup for both sides as well. Thanks.

There are so many variables in each battle that it's almost impossible to respond in detail without seeing how you were both setup. My experience with this game is that lopsided casualties are rare, for the most part battles work out historically, even in the East. While I've seen some odd results, usually when lopsided casualties happen I've found it to be because one side or the other was doing something wrong, either in army setup, scouting, etc.

[Berkut]

Please don't tell me that the scouting level influences commitment as well!

[Joel Billings]

ORIGINAL: Berkut

Please don't tell me that the scouting level influences commitment as well!

I don't think it does, but not being spotted has a big impact on casualty ratios.