retreat direction

[bevilacqua]

I'm playing my first PBEM and something that happened a couple of times is frustrating me. My units are retreating in the opposite direction they should, placing themselves in a trap, like is shown on the picture.

As far as I know, units will retreat in the direction of HQs or the nearest supply point. My HQs were in the hexes just behind (for those divisions that had HQs - some didn't). I had only reorganizing units on the lower hex and I knew that the low proficiency and low supply units I had with them and near them wouldn't hold position, so I retreated all those that could receive order and reinforced the two adjacent hexes which were attacked in this picture, knowing they wouldn't hold also, but hoping to make my opponent lose some time. But, for my surprise, when they were attacked, most of the units (some of them good, fresh units) jumped back into the lower hex, placing themselves into a trap and evaporated after the next 180 degree flanking attack.

This was the most unnatural move, since HQs were to the back of the line, supply points also, and they entered a river hex, with higher move cost. It seemed really random and made me think that it is prohibitive to place good units near such an hex, but it shouldn't be. Those units were rounded by friendly, good order, units and should retreat in their direction in this case.

Is there a way to avoid that or do I have to pray and wait for the game to shuffle my units around setting my chances? Maybe I'm really missing something here. If so, please let me know.

[bevilacqua]

Here is it. I forgot to place it in the original message.

[Panama]

Yeah, in one of the North African scenarios the Allies at Tobruk retreat TOWARDS the Axis units instead of towards the Med. It's really not much fun to play that one.

[Oberst_Klink]

Manual - 13.11 retreats
If a defending unit attempts to break off, it will look for a safe location in the direction of the nearest friendly cooperative Headquarters or Supply Source. The unit will attempt to disengage and retreat into that safe (well, what's safe?!) location. If such a location is not available, the unit will instead have its Readiness reduced to 33% and it will refuse further orders until it Reorganizes. In practical terms,this reduces the unit to a milling mob of uncoordinated troops, which will offer little resistance if attacked again.

Isn't a retreat also affected by the terrain, proficiency (what's left...) of the unit, movement points left,stack value, combat loss setting, etc.? You moved the NKPA lads northwards. The left retreating unit was on a river hex... terrain I think might be also a factor (Ralph, where are you?!)In general, yes - those "stupid" retreats happen. I can't say (maybe I never gave a d*mm) that it happened alot to me. If the squaddies get routed, they panic, ... I am sure the last thing they try to figure out is where the supply point or HQ is. Found this, so I'll quote:

Though force preservation is a basic element of warfare and wargames, it is critically important. The one thing that will save you is defense in depth. When one unit is forced to retreat, it is likely retreating into a hex where another unit is already entrenched.

So, which hex was the one with the best entrenchment value in yur case?!

Again, just guessing, only RT himself might shed some light on it.

[Curtis Lemay]

ORIGINAL: bevilacqua

Here is it. I forgot to place it in the original message.

I can't tell from that shot but you might have blocked them in yourself via stacking limits. Once a hex has nine friendlies in it nothing more can retreat there.

Regardless, a retreat is an adverse combat result and therefore retreat paths may be thought of as blocked internally to the hex by the attackers - forcing the retreaters into undesirable directions. In other words, the attackers can be thought to penetrate the defenders, spread out behind them within the hex, and roll them up into a trapped location.

[ralphtricky]

If someone can post a save where I can reproduce the problem I'll see if it's fixed.

Another possible cause is that it does try to retreat towards the point which has the closest distance to the nearest supply + the distiance to the nearest COMPATIBLE HQ. Note the word COMPATIBLE there.

I think I see the problem, there is a slight bias towards keeping NEXT TO the enemy units instead of away from if there is a owned supply point on the map, and AWAY FROM if there is no owned supply point on the map. That's probalby what is happening.

I can't think of why it would treat those two cases differently. I'll see if I can change that to AWAY FROM in all cases instead.

I'll try to get it into 3.4 if there is time.

Thanks,
Ralph

[Curtis Lemay]

ORIGINAL: ralphtrick

If someone can post a save where I can reproduce the problem I'll see if it's fixed.

Another possible cause is that it does try to retreat towards the point which has the closest distance to the nearest supply + the distiance to the nearest COMPATIBLE HQ. Note the word COMPATIBLE there.

I think I see the problem, there is a slight bias towards keeping NEXT TO the enemy units instead of away from if there is a owned supply point on the map, and AWAY FROM if there is no owned supply point on the map. That's probalby what is happening.

I can't think of why it would treat those two cases differently. I'll see if I can change that to AWAY FROM in all cases instead.

I'll try to get it into 3.4 if there is time.

Let's discuss this before you make a snap reaction. If you change this it will become harder to achieve a breakthrough.

[ralphtricky]

OK. I'll hold off on changing this until 3.5 so we can discuss further.We may also want to look at how strong the effect of the supply source should be in determining the direction of retreat. One possibility we might look into as well would be whether the stance should effect the retreat direction, and make it so that limit losses are more likely to retreat in a safe direction while ignore losses is more likely to do retreat along the lfront line or something like that.

[bevilacqua]

Stacking limits were far from being reached on all hexes around. Even the units without HQ should retreat back in the direction of friendly good order units and supply points, I think.

Enemies blocking the way could be considered to happen sometimes. It should be more probable if the units are flanked, which wasn't the case. And it wasn't one or other units, but a whole lot of them from both sides, converging into the advanced hex and being completely eliminated later. Looking the configuration one notice that this kind of encirclement by the enemy, blocking retreat, happening simultaneously on both sides would be most unprobable. I could accept one or two units, from a group of ten, going into this hex, but not more than half of them. Additionally, terrain to the north of them required less movement points, which means it has an easier access... it would really be the natural path in this case.

This wasn't an isolated case. It happens frequently. Not always was it to my disadvantage (I'm not whinning here), but having it a predominantly random nature, players are helpless in a situation in which they should, at least, be able to estimate outcome in order to plan... if not, what's the point?

The fact is that placing units that way (not that I have much choice in such a situation) will make games with few units dicey if I have no idea of what is the most probable way of retreat. Place the units an pray. One bad dice roll and a carefully placed defense breaks.


Regarding the greater difficulty it would generate (the change Ralphtrick proposes) to attain a breakthrough, couldn't it be based on deployment, so that ignore losses deployment would keep the present behaviour, representing the unit intention to keep its elan? Limit losses would make them more prone to retreat in the direction of their HQs and supply points. So players would get a better control of it behaviour, if not absolute.

[PRUSSIAN TOM]

My hat goes off to Ralph & Gen. LeMay. One reason this is my favorite game (by the company who I bought the momst games from) is that YOU HARD WORKING GRUNTS (sorry General, but I know they can not be paying you what you and Ralph are worth by the hour (and the other people who work on older games. Matrix supports their product! [:)] Sometimes we get a little antsy when you take your time to get it right before releasing a patch, but questions on the formus seem to get answers by either GROGNARDS who know their stuff, or MATRIX EMPLOYEES (who are [paid tom know ther SH**, and DO) REPLYS ASAP. Myb grateful thanks.[:D]. I get my problems solved faster than the company that made my (new) PC caN SOLVE warranty issues. Keep up the great work....it is appreciated by us dum-dums.

[Curtis Lemay]

ORIGINAL: bevilacqua

Stacking limits were far from being reached on all hexes around.

But could they have been reached during the retreats? Just because they were clear at the end of your player turn doesn't mean they remained clear throughout the enemy combat phase. Units tend to sub-divide during combat, and that fills up the stacking limits in a hurry.

Even the units without HQ should retreat back in the direction of friendly good order units and supply points, I think.

If it were an entirely voluntary action I would agree with you. But it isn't. It's an adverse combat result forced upon the defenders by the victorious attackers - who may have penetrated the defenses sufficiently to critically compromise the defenders' retreat options.

Don't think of the defender's hex as a hex. Think of it as a giant tactical battlefield. Imagine the attackers driving through the defense and spreading out in their rear - blocking the most desireable paths out, and rolling the rest into a trap.

And it wasn't one or other units, but a whole lot of them from both sides, converging into the advanced hex and being completely eliminated later.

See my stacking comments above. If not that, it was bad luck. I'm pretty sure the direction is random if there is are choices.

This wasn't an isolated case. It happens frequently. Not always was it to my disadvantage (I'm not whinning here), but having it a predominantly random nature, players are helpless in a situation in which they should, at least, be able to estimate outcome in order to plan... if not, what's the point?

"No plan survives contact with the enemy." Again, these are adverse combat results. If you're planning, you have to include these sorts of results in those plans.

Regarding the greater difficulty it would generate (the change Ralphtrick proposes) to attain a breakthrough, couldn't it be based on deployment, so that ignore losses deployment would keep the present behaviour, representing the unit intention to keep its elan? Limit losses would make them more prone to retreat in the direction of their HQs and supply points. So players would get a better control of it behaviour, if not absolute.

Possibly there should be some impact of that. But, let's recognize that breakthroughs are especially hard in TOAW. This will make them harder.

[Curtis Lemay]

ORIGINAL: PRUSSIAN TOM

My hat goes off to Ralph & Gen. LeMay. One reason this is my favorite game (by the company who I bought the momst games from) is that YOU HARD WORKING GRUNTS (sorry General, but I know they can not be paying you what you and Ralph are worth by the hour (and the other people who work on older games. Matrix supports their product! [:)] Sometimes we get a little antsy when you take your time to get it right before releasing a patch, but questions on the formus seem to get answers by either GROGNARDS who know their stuff, or MATRIX EMPLOYEES (who are [paid tom know ther SH**, and DO) REPLYS ASAP. Myb grateful thanks.[:D]. I get my problems solved faster than the company that made my (new) PC caN SOLVE warranty issues. Keep up the great work....it is appreciated by us dum-dums.

Thanks. Just for the record, I'm not paid anything. I think Ralph gets less than minimum wage. It's a labor of love.

[bevilacqua]

ORIGINAL: Curtis Lemay

ORIGINAL: bevilacqua

Stacking limits were far from being reached on all hexes around.

But could they have been reached during the retreats? Just because they were clear at the end of your player turn doesn't mean they remained clear throughout the enemy combat phase. Units tend to sub-divide during combat, and that fills up the stacking limits in a hurry.

I can count, but thank you for the concern.

Even the units without HQ should retreat back in the direction of friendly good order units and supply points, I think.

If it were an entirely voluntary action I would agree with you. But it isn't. It's an adverse combat result forced upon the defenders by the victorious attackers - who may have penetrated the defenses sufficiently to critically compromise the defenders' retreat options.

Don't think of the defender's hex as a hex. Think of it as a giant tactical battlefield. Imagine the attackers driving through the defense and spreading out in their rear - blocking the most desireable paths out, and rolling the rest into a trap.

Oh, I thought that 15 km was the hex size. I always found it strange that they looked so tiny on my computer screen.

I think it has something to do, perhaps, with the variable Ralphtrickey refered to. If it isn't to be voluntary, there shouldn't be a variable attracting units in the direction of an enemy one. The move you describe is possible, of course, but wouldn't be predominant.

And it wasn't one or other units, but a whole lot of them from both sides, converging into the advanced hex and being completely eliminated later.

See my stacking comments above. If not that, it was bad luck. I'm pretty sure the direction is random if there is are choices.

This happens with some frequency.

This wasn't an isolated case. It happens frequently. Not always was it to my disadvantage (I'm not whinning here), but having it a predominantly random nature, players are helpless in a situation in which they should, at least, be able to estimate outcome in order to plan... if not, what's the point?

"No plan survives contact with the enemy." Again, these are adverse combat results. If you're planning, you have to include these sorts of results in those plans.

[/quote]

When planning, one must be able to estimate results. Using this authority based quote as an absolute truth is an extreme simplification. Let's just throw dices in the beginning of each turn and let a complete random event decide the outcome of a scenario.

Regarding the greater difficulty it would generate (the change Ralphtrick proposes) to attain a breakthrough, couldn't it be based on deployment, so that ignore losses deployment would keep the present behaviour, representing the unit intention to keep its elan? Limit losses would make them more prone to retreat in the direction of their HQs and supply points. So players would get a better control of it behaviour, if not absolute.

Possibly there should be some impact of that. But, let's recognize that breakthroughs are especially hard in TOAW. This will make them harder.

[/quote]

Here I agree with you. Breakthroughs are hard, but using an artificial engine functionality to prevent this isn't the best idea. IMHO, I think that one cause for this is the absurd cost of movement near all kind of units (large ones or small ones), so that Toaw becomes more of a game based on counters than one based on real forces. It's possible to place some very small units just to delay enemy by making movement costs very high. In fact, terrain effect on movement is insignificant compared to the effect a COUNTER can have, no matter the size of the unit contained. But that's just my opinion and I don't want to begin a flame war.

[Oberst_Klink]

See attached... It can happen and like some of us pointed out, depends on various factors...

[bevilacqua]

If you read what I said you'll see that in no moment I deny it can happen, I just said it wasn't predominant.

If you read what Ralphtrickey said, you'll see that there is a variable that favors it to happen, when there is no reason to. Why the hell would a unit be more prone to retreat towards the enemy, in panic or not.

If someone would have the time and will (I don't, for sure) to raise the numbers and see how many times units retreated in the direction of their own forces and how many times they retreated in the direction of the enemy, when not flanked and when near to good order friendly units, I can bet that the numbers would clearly show that the first option is the one that predominates. Ok, in most cases it wouldn't be a zero or one result, but I can bet that in most cases, most of the troops were able to go in the direction of their own forces. The manual itself states this tendency, by pointing that troops will usually retreat in the direction of friendly HQ or supply points, but, it seems, there is a variable preventing them from doing it and no reason for this. That's all. But I'm just expressing my opinion here, not speaking against the game.

ORIGINAL: Oberst_Klink

See attached... It can happen and like some of us pointed out, depends on various factors...

[Oberst_Klink]

ORIGINAL: bevilacqua

If you read what I said you'll see that in no moment I deny it can happen, I just said it wasn't predominant.

If you read what Ralphtrickey said, you'll see that there is a variable that favors it to happen, when there is no reason to. Why the hell would a unit be more prone to retreat towards the enemy, in panic or not.

If someone would have the time and will (I don't, for sure) to raise the numbers and see how many times units retreated in the direction of their own forces and how many times they retreated in the direction of the enemy, when not flanked and when near to good order friendly units, I can bet that the numbers would clearly show that the first option is the one that predominates. Ok, in most cases it wouldn't be a zero or one result, but I can bet that in most cases, most of the troops were able to go in the direction of their own forces. The manual itself states this tendency, by pointing that troops will usually retreat in the direction of friendly HQ or supply points, but, it seems, there is a variable preventing them from doing it and no reason for this. That's all. But I'm just expressing my opinion here, not speaking against the game.

I know that you're not speaking against the game. And yes, I read what the other lads said/posted. BTW, have you got the sitrep log or the TOAW.log for the scenario you are playing? That might shed some light on it, too.

Thanks mate!

Klink, Oberst

[PRUSSIAN TOM]

Gen. Curtis LeMay & Ralph:

I understand you are in it for the sake of doing it right, but that's why it's my favorite game (And the big N. Afrika Scenario is my favorite scenario). You spent the time to do a great job[&o]

Not all software simulations are created equal, and because of your (all of you, collectively) commitmitment, I buy and enjoy a LOT of Matrix games. (Sorry they don't give you a kickback, but a good job usually isn't about the money anyway, like you said, it's a labor of love.
THANKS AGAIN. [&o]

[bevilacqua]

Here is what I meant when I said toaw is a game based on counters rather than one based on forces and that's what makes breaking through a line so difficult.

I have setup a very quick test and you can see by the picture that the distance which took one unit 2 turns to cross through a massive mountain chain, took 7 turns for another unit to cross through a corridor of open ground sided by enemy units. Ok, it had to deal with the enemy, which makes movement slow and careful, isn't it? Well, all those counters contained only a single truck each, which means that if the enemy places single trucks alond the way while retreating, what would be a nice walk through open contry, makes unit stuck by their own paranoia.

Conclusion: better than having a whole mountain chain between you and your enemy is having a couple of old trucks placed kilometers away one from the other.

It's obvious that movement penalties from enemy proximity should be proportional to units size. The way it is, what counts, when trying to delay an enemy is the number of counters and not what they contain. I'm not talking about combat, but about using small units to make the enemy stuck in the middle of counters and slowing it MUCH MORE than the worst kind of terrain.

[bevilacqua]

Once more I forgot the picture. Sorry

[Curtis Lemay]

I've been doing some rigorous tests on retreats (using test scenarios), and I've been unable to cause any retreat direction other than towards the nearest supply point. HQs seem to be ignored. This has surprised me because I can clearly recall lots of game cases of units scattering in multiple directions. But I've tried the test scenarios all the way back to TOAW I (not ACOW - TOAW I).

One issue - the supply point chosen is the nearest one, even if it is in the enemy rear. This causes units to retreat towards the enemy. Even surrounding the combat with friendly supply points still has the units retreating in one direction.