TOAW 3.5 approaches ?

[Curtis Lemay]

ORIGINAL: golden delicious

It's not a simple bug, no. It's a deliberate change which has broken the game, and needs to be put back.

No. It is a problem that has existed in TOAW from day one: Combat odds don't affect RFC chances. This was masked by another long-term issue that was fixed in 3.4: Terrain didn't affect RFC chances at all. Now they do. We're not going to change it back to where terrain doesn't affect RFC chances again. It's got to be changed to where combat odds do affect RFC chances. That's not a simple fix.

Readers can review my tests on it in this thread:

tm.asp?m=3056061

[governato]

ORIGINAL: Panama

ORIGINAL: governato
It should take weeks if not months to get to 'F' status, which is meant to represent permanent fortifications.

This is not true. These are field works. Not concrete bunkers. Things like tank ditches, mine fields, wire, earthen bunkers, trenches, etc.

12.4 Fortification (Advanced Rules)
All Land units may deploy into defensive positions. The
presence of units with an Engineering Capability increases the
chance of successfully Entrenching or Fortifying. Even if a unit
fails to Entrench or Fortify, it will increase the local Entrenchment
Level, making it easier for that unit and others to Entrench or
Fortify in the future. Once created, Field Fortifications are permanent.
As the Entrenchment Level of a location increases, units
in the location will find it easier to enter Entrenched or Fortified
Deployments. When a location changes hands, the Entrenchment
Level is automatically reduced by 25%. Once ordered to
dig in, units will continue to dig until their location is Fortified or
you give them other orders.

Fair enough, still it historically took months for the Germans to build fortified lines (so yes field works) like the Panther line and yet they were rather easily overcome. My main point is that it should take longer than a few turns to reach the maximum defensive strenght for a unit, especially when close to the enemy, irrespective of what it actually represents in game terms.

[Oberst_Klink]

Gentlemen please...

we all know what needs to be fixed. The important matter of the issue is, to convey and convince the head-honchos to work on it, no? Ain't help us and the cause if we have re-runs of different opinions we already addressed and 'fought' out

Klink, Oberst

[fogger]

Spot on Colonel Klink

[golden delicious]

ORIGINAL: Curtis Lemay
It's got to be changed to where combat odds do affect RFC chances. That's not a simple fix.

First- that's a useful set of tests. I appreciate it. I had picked up from somewhere that there'd been a straight doubling of the fortified bonus. God knows where that came from.

When you say combat odds, you're effectively talking about AR ratio, right?

If we're talking about a failed quality check (i.e. the unit retreating even though it's told not to) it's going to be because the unit has either taken a lot of losses or on a micro (sub-hex) level its position has been compromised and can't be restored.

Neither of these things necessarily correlate to the sheer weight of force applied to the defenders. If it did, human wave attacks would be a great idea. In fact I'd say they correlate most closely to losses- but of course losses are going to be pretty minimal because the unit has a 10x defence bonus or whatever.

I'd say that the problem is that by linking the RFC chance to fortified status, you've double counted the benefit, as the unit is already taking fewer losses and therefore less likely to even have to face a quality check. Moreover, if I read the "What's New" document correctly, the maximum chance of a fortified unit retreating on any one combat round is 16%, regardless of the nature of the attack or the condition of the defender. I gather from the wording that there was already such a limitation, but it doesn't state what the old figures were.

It seems to me that the thing to do is to ditch item IV.10 in the "What's New" document. That should put things to rights.

[Curtis Lemay]

ORIGINAL: golden delicious

ORIGINAL: Curtis Lemay
It's got to be changed to where combat odds do affect RFC chances. That's not a simple fix.

First- that's a useful set of tests. I appreciate it. I had picked up from somewhere that there'd been a straight doubling of the fortified bonus. God knows where that came from.

Well, there is some increase from terrain (squareroot of the sum of the squares). So, a fortified unit in good terrain is somewhat better off than in 3.2. But not doubled. I'm thinking that, if possible, fortified line terrain should not so sum with anything. Maybe that can be part of the fix.

When you say combat odds, you're effectively talking about AR ratio, right?

It would be post-combat odds, though.

If we're talking about a failed quality check (i.e. the unit retreating even though it's told not to) it's going to be because the unit has either taken a lot of losses or on a micro (sub-hex) level its position has been compromised and can't be restored.

Neither of these things necessarily correlate to the sheer weight of force applied to the defenders. If it did, human wave attacks would be a great idea. In fact I'd say they correlate most closely to losses- but of course losses are going to be pretty minimal because the unit has a 10x defence bonus or whatever.

Actually, human wave attacks are a great idea if the wave survives intact to reach the enemy (Japs vs. Chinese). If it's slaughtered or breaks off, though, it isn't such a good idea (Japs vs. US). That's why the test needs to be post-combat.

I'd say that the problem is that by linking the RFC chance to fortified status, you've double counted the benefit, as the unit is already taking fewer losses and therefore less likely to even have to face a quality check.

In Ignore Losses deployment, losses aren't considered. Only morale. Therein lies the problem. Prior to 3.4, terrain had no effect, and deployment much less effect on RFCs. It was as easy (or hard) to kick a unit out of clear terrain as fortified line.

Moreover, if I read the "What's New" document correctly, the maximum chance of a fortified unit retreating on any one combat round is 16%, regardless of the nature of the attack or the condition of the defender.

Yep. That's the problem. Because we hadn't discovered that combat odds didn't affect RFC chances. That needs to be added. Although, note that 9 straight 16% chances total to about 80% chance of success per turn - if you do it right.

I gather from the wording that there was already such a limitation, but it doesn't state what the old figures were.

I don't remember what the old level was - I think it was 50%. It was increased to match the fortified terrain value.

It seems to me that the thing to do is to ditch item IV.10 in the "What's New" document. That should put things to rights.

I can see some changes to the numbers, but the principle function should be retained. We certainly don't want to back out the terrain part, and I don't think the deployment numbers should be too different from similar terrain numbers. What we really need is to scale RFC chances by combat odds.

Regardless, it isn't a simple fix no matter what we do. It's going to require a lot of trial and error and lots of tests of each trial to check that it's working right.

In contrast, the AAA issue would be a simple fix. That's just a simple bug that, once fixed, we would know almost immediately that it was working right.

[sPzAbt653]

In Ignore Losses deployment, losses aren't considered.

In an Ignore Losses combat, are Losses increased due to the implied 'no retreat' order ?

It seems like the Combat Odds should figure into the equation, such as ... a 1-2 attack against a fortified Ignore Losses defender will net a 1% loss to the defender and no retreat, while a 5-1 attack against a fortified Ignore Losses defender will net a 16% loss to the defender and normally a retreat result but with Ignore Losses there will be no retreat with an additional 8% loss, or something similar.

[Curtis Lemay]

ORIGINAL: sPzAbt653

In Ignore Losses deployment, losses aren't considered.

In an Ignore Losses combat, are Losses increased due to the implied 'no retreat' order ?

It seems like the Combat Odds should figure into the equation, such as ... a 1-2 attack against a fortified Ignore Losses defender will net a 1% loss to the defender and no retreat, while a 5-1 attack against a fortified Ignore Losses defender will net a 16% loss to the defender and normally a retreat result but with Ignore Losses there will be no retreat with an additional 8% loss, or something similar.

The loss setting only affects the odds of the unit retreating/dropping out of an attack. Obviously, if it doesn't retreat/drop out, it stands to take further losses in subsequent rounds. And, combat odds do affect losses - via the increased firepower from the greater forces involved.

[golden delicious]

ORIGINAL: Curtis Lemay

Actually, human wave attacks are a great idea if the wave survives intact to reach the enemy (Japs vs. Chinese). If it's slaughtered or breaks off, though, it isn't such a good idea (Japs vs. US). That's why the test needs to be post-combat.

Still. In TOAW, massed infantry attacking aren't going to be slaughtered to a man. If you attack with a 10-1 AR ratio with light infantry, I'd bet it'll be at least 8-1 after one round.

When human waves work is when a) your opponent has very little firepower so you don't take losses anyway and b) the enemy has very low unit quality. After the first few months of the war, both would apply to the Chinese.

In Ignore Losses deployment, losses aren't considered. Only morale. Therein lies the problem.

I was thinking a unit doesn't automatically take a quality check just because the attack is still underway.

Thing would be to add the % losses to one of the die rolls. So multiply unit quality by (1-% losses) or something like that.

Although, note that 9 straight 16% chances total to about 80% chance of success per turn - if you do it right.

The difficulty is this is just the chance of even facing the quality check. If the unit is in reasonably good condition- even with a modest proficiency- you'll struggle to get to a 50% chance of the unit retreating. That's factoring in the round by round loss of supply and readiness.

I gather from the wording that there was already such a limitation, but it doesn't state what the old figures were.

I don't remember what the old level was - I think it was 50%. It was increased to match the fortified terrain value.

I can see some changes to the numbers, but the principle function should be retained. We certainly don't want to back out the terrain part,

Now it's in there, no- but I'd delete the inherent entrenchment levels of hexes. As it stands, one very rapidly gets that "F" in hill terrain because one starts off with 15% entrenchment in the hex. This is even more pronounced if one reduces the global % entrenchment rate (something I've experimented with and then reverted).

In contrast, the AAA issue would be a simple fix. That's just a simple bug that, once fixed, we would know almost immediately that it was working right.

Well, it's a bug, but that masks the broader problem. I'd like to see AAA reduce the effectiveness of attacking aircraft, especially low level aircraft. However, that's kind of a pipe dream until we get the ball rolling again on development.

[Curtis Lemay]

ORIGINAL: golden delicious

Still. In TOAW, massed infantry attacking aren't going to be slaughtered to a man. If you attack with a 10-1 AR ratio with light infantry, I'd bet it'll be at least 8-1 after one round.

The Jap human wave attacks were slaughtered to a man. So, if 8-1 remain, it wasn't really a human wave attack - it was something more conventional. Also, note that 8-1 is neutral against a fortified defender. You'll have to get odds beyond that to get any scaling.

I was thinking a unit doesn't automatically take a quality check just because the attack is still underway.

I think that's right. Some losses of some sort are necessary to get a check. But that's not a very high requirement.

Thing would be to add the % losses to one of the die rolls. So multiply unit quality by (1-% losses) or something like that.

That's wouldn't have much effect. And it would vary wildly with AD and such. Odds are a better factor.

Although, note that 9 straight 16% chances total to about 80% chance of success per turn - if you do it right.

The difficulty is this is just the chance of even facing the quality check. If the unit is in reasonably good condition- even with a modest proficiency- you'll struggle to get to a 50% chance of the unit retreating. That's factoring in the round by round loss of supply and readiness.

Oh sure - as I said, you've got to do it right. But even a 50% per turn chance seems pretty good against a fortified defender. I still think that some of the problem is that players had it so easy before - no AR check, no Cooperation Penalties, no terrain RFC effects, etc. You could just bombard your way to victory in most cases.

Now it's in there, no- but I'd delete the inherent entrenchment levels of hexes. As it stands, one very rapidly gets that "F" in hill terrain because one starts off with 15% entrenchment in the hex. This is even more pronounced if one reduces the global % entrenchment rate (something I've experimented with and then reverted).

Well, it's a bug, but that masks the broader problem. I'd like to see AAA reduce the effectiveness of attacking aircraft, especially low level aircraft. However, that's kind of a pipe dream until we get the ball rolling again on development.

Both of those are cans-of-worms that I don't think we can fit into 3.5's poor little code budget - however worthy they may be.

[golden delicious]

ORIGINAL: Curtis Lemay

The Jap human wave attacks were slaughtered to a man.

Right. Will that happen in TOAW, though? The scenario I'm designing is at divisional scale. Often the first wave gets slaughtered but there's a lot more division still to come.

Also, note that 8-1 is neutral against a fortified defender. You'll have to get odds beyond that to get any scaling.

That's just it. In reality, I'd rather attack with 4-1 and lots of artillery than 12-1 with none.

That's wouldn't have much effect. And it would vary wildly with AD and such. Odds are a better factor.

I don't think they are. Units don't retreat because they've been attacked by a strong unit, they retreat because they're getting mauled. Thinking further, I'd take the ratio of attacker to defender losses as an indication of whether a) people are going to start thinking they don't want to die and b) the commander is going to decide that the unit is no longer capable of holding its position.

Oh sure - as I said, you've got to do it right. But even a 50% per turn chance seems pretty good against a fortified defender.

The trouble is this is the maximum chance. Again, to take the scenario I'm working on, one can have a fortified division holding a 25km stretch of line. A strong force with plenty of support should be pretty much guaranteed to be able to root them out if they're prepared to take losses.

You could just bombard your way to victory in most cases.

You should see the quote that Jeremy just posted over at TDG (German experiences during Cobra). Bombardment can be truly devastating when used as part of a balanced diet. It doesn't do the job on it's own- hence the AR factor. For which we thank you.

Both of those are cans-of-worms that I don't think we can fit into 3.5's poor little code budget - however worthy they may be.

Agreed.

[Panama]

I just want to be able to finish my scenario and play the game without a plethora of work arounds and 'house' rules. That's all. And before summer would be great (or winter depending on which end of the orange you live on).

[Curtis Lemay]

ORIGINAL: golden delicious

Right. Will that happen in TOAW, though? The scenario I'm designing is at divisional scale. Often the first wave gets slaughtered but there's a lot more division still to come.

I can certainly imagine the net odds at the end of the combat being much worse than they were at the start - due to attacking units dropping out or evaporating. But actually modeling human waves with TOAW may be beyond its abilities - making the issue a red herring after all.

That's just it. In reality, I'd rather attack with 4-1 and lots of artillery than 12-1 with none.

Units don't retreat because they've been attacked by a strong unit, they retreat because they're getting mauled.

At least we're agreeing that the RFC chances need to be revised so that they depend upon the nature of the attack results in some fashion - whether it be ending odds or ending losses, etc. The way it is now (and always has been) is absurd.

I'm not adverse to incorporating some of the support into the odds (defender support too?), but certainly not more than the ground strength. Most losses are, in fact, caused by support - yet you can't surrender to an aircraft or artillery tube. It takes ground forces to gain ground (just as it takes ground forces to hold ground). That's why I think odds are a more realistic scalar. It's the threat of ground infiltration of your position that forces a retreat - not a bombardment.

Thinking further, I'd take the ratio of attacker to defender losses as an indication of whether a) people are going to start thinking they don't want to die and b) the commander is going to decide that the unit is no longer capable of holding its position.

But, even more important is that odds are a more workable solution. What you're describing just wouldn't work very well. Any loss ratio on a fortified defender would be heavily in his favor. Plus, losses are capricious (varying wildly with chance) - especially if the unit is down to only a few peices of equipment. It would be very difficult to keep track of losses over multiple combat rounds. Without that, it is very rare for a unit to take more than a few percent losses per round - the scalar effect would be insignificant.

The trouble is this is the maximum chance. Again, to take the scenario I'm working on, one can have a fortified division holding a 25km stretch of line. A strong force with plenty of support should be pretty much guaranteed to be able to root them out if they're prepared to take losses.

Of course. We were talking about the way it is now - and there is no argument that it needs to be changed. I was just pointing out that if you do it right (get all the rounds possible) even fortified defenses can have good chances to be RFC'd - even with the current problem.

Now, for the fix, I expect there will be a game scalar (like the Attrition Divider) that allows designers to vary the effect as desired - to accomodate different time scales or different periods (WWI vs. WWII, etc.).

You should see the quote that Jeremy just posted over at TDG (German experiences during Cobra). Bombardment can be truly devastating when used as part of a balanced diet. It doesn't do the job on it's own- hence the AR factor. For which we thank you.

Again, of course. But I was just pointing out that there had been a very lazy, gamey route to victory that we all knew about. That's gone now, and some players may have been too used to depending on it. I just can't help wondering if that might account for some fraction of this issue - not all of it, of course, since there is a recognized problem. But offensives under 3.4 are sure to be tougher than under 3.2 - just for ending the "bombard to victory" tactic. And that's a good thing.

[Curtis Lemay]

ORIGINAL: Panama

I just want to be able to finish my scenario and play the game without a plethora of work arounds and 'house' rules. That's all. And before summer would be great (or winter depending on which end of the orange you live on).

I've got five new scenarios being held hostage to 3.5. Nobody wants it finished more than me. But that's up to Ralph.

[Oberst_Klink]

ORIGINAL: Curtis Lemay

ORIGINAL: Panama

I just want to be able to finish my scenario and play the game without a plethora of work arounds and 'house' rules. That's all. And before summer would be great (or winter depending on which end of the orange you live on).

I've got five new scenarios being held hostage to 3.5. Nobody wants it finished more than me. But that's up to Ralph.

Touche! I like the expression 'held hostages' regarding the scenarios In my case it's soon due... 70th anniversary of the 3rd battle of Kharkov; perhaps I release it, despite no-3.5.

Klink, Oberst

[golden delicious]

ORIGINAL: Curtis Lemay

At least we're agreeing that the RFC chances need to be revised so that they depend upon the nature of the attack results in some fashion - whether it be ending odds or ending losses, etc. The way it is now (and always has been) is absurd.

I'm not adverse to incorporating some of the support into the odds (defender support too?), but certainly not more than the ground strength. Most losses are, in fact, caused by support - yet you can't surrender to an aircraft or artillery tube. It takes ground forces to gain ground (just as it takes ground forces to hold ground). That's why I think odds are a more realistic scalar. It's the threat of ground infiltration of your position that forces a retreat - not a bombardment.

Well, there's two scenarios here. Either the commander has decided that he can't hold his position, or he loses control over his subordinates and is unable to force them to stay in their positions.

The latter's primarily going to be a function of losses. Further, a high portion of losses compared to the attacker suggests the attacker is winning on the tactical scale and forcing units out of their positions. That's why I'm in favour of using that ratio. A large attacking force which makes no serious impact on the defender is not going to cause them to retreat.

But, even more important is that odds are a more workable solution. What you're describing just wouldn't work very well. Any loss ratio on a fortified defender would be heavily in his favor.

Not if you use the %. Typically, the defender is taking significantly less losses in absolute terms, but may well be taking 20-50% losses and still holding the hex under the current situation. The attacker will generally only take 5% losses per round.

Now, for the fix, I expect there will be a game scalar (like the Attrition Divider) that allows designers to vary the effect as desired - to accomodate different time scales or different periods (WWI vs. WWII, etc.).

Excellent. These global or force specific variables are gold dust for designers.

But offensives under 3.4 are sure to be tougher than under 3.2 - just for ending the "bombard to victory" tactic. And that's a good thing.

Mm. But were offensives in TOAW generally too easy in 3.2? I think you're agreeing that they weren't, because of two problems that cancelled each other out.

[golden delicious]

ORIGINAL: Panama

I just want to be able to finish my scenario and play the game without a plethora of work arounds and 'house' rules. That's all. And before summer would be great (or winter depending on which end of the orange you live on).

What I'm trying to avoid is making extensive changes to the detail of my scenario to account for the problems. Actually, things like house rules and global variables are a good way to deal with this problem. If we get a new version, you can switch them back in half an hour and then get testing.

[Curtis Lemay]

ORIGINAL: golden delicious

Well, there's two scenarios here. Either the commander has decided that he can't hold his position, or he loses control over his subordinates and is unable to force them to stay in their positions.

The latter's primarily going to be a function of losses. Further, a high portion of losses compared to the attacker suggests the attacker is winning on the tactical scale and forcing units out of their positions. That's why I'm in favour of using that ratio.

A morale check failure is an involuntary result - there's no "cost vs. benefit analysis" involved (that would be a voluntary retreat made during the friendly movement phase). So, while friendly losses may figure into it, I don't see enemy losses doing so. Worse, as I've said, most losses are caused by support. So, high friendly losses will not usually represent success by the attacker on the ground, but rather just that the attack was heavily supported. Again, bombardments don't gain ground - that is done by ground forces. I still say that odds are a more realistic factor.

A large attacking force which makes no serious impact on the defender is not going to cause them to retreat.

Maybe not, but it's usually a safe assumption that it is making serious impact. It has to inflict some losses to get any morale check to begin with - and the game doesn't remember what losses it may have inflicted in previous rounds. For certain, facing heavy odds would be a factor in that involuntary morale check result.

Not if you use the %. Typically, the defender is taking significantly less losses in absolute terms, but may well be taking 20-50% losses and still holding the hex under the current situation. The attacker will generally only take 5% losses per round.

If that were true we wouldn't even be having this discussion. A unit that is taking those kinds of losses per round would quickly be evaporated or reduced to RBC size in a few rounds.* This problem would never have been noticed. Support skews losses badly enough to ruin your theory. Defender support will decimate the attackers regardless of the size of the defender. Attacker support will cause most of the defender losses. And there just isn't going to be that great a differential between the two on a per-round basis.

*(Note that RBC ability has been enhanced under 3.4 - a factor that was expected to counter some of the defender enhancements. Players need to make sure that they don't overlook that).

And think what that means: It's "bombard to victory" all over again (almost). Attackers will risk no more ground force than necessary to pass the first AR check (1:1 at most). They will then pile on the support to ramp up the losses. In fact, the round may end with all attackers broken off, yet still force a retreat due to the losses incurred.

For certain, odds are a more workable solution. And think how much more interesting the game will be if players have to pay attention to combat odds for once.

Excellent. These global or force specific variables are gold dust for designers.

It's something TOAW has needed even if there hadn't been this problem. WWI was just different from WWII. It wasn't just because both sides were heavily entrenched that there was a stalemate. Even during the mobile phase it was much harder to gain ground than in other wars. Part of the reason the Schlieffen Plan failed was that the German tactical commanders found that they had to keep increasing the odds to keep to the advance schedule. This forced them to contract their lines so much that they had to pass under Paris instead of around it.

Mm. But were offensives in TOAW generally too easy in 3.2? I think you're agreeing that they weren't, because of two problems that cancelled each other out.

I think there probably were designs out there that depended upon that gamey tactic for their scenarios to work, just as there were players that depended on it. For sure, ditching that tactic will make things harder for those designers & players. They're going to have to use the entire TOAW tool box. And, again, that's a good thing.

[golden delicious]

ORIGINAL: Curtis Lemay

A morale check failure is an involuntary result - there's no "cost vs. benefit analysis" involved (that would be a voluntary retreat made during the friendly movement phase).

It's involuntary on the part of the player. However it may be a decision made by the commander of the unit: he sees his men getting killed to no purpose and decides to take a risk on his career to get the unit out.

Thinking further about the micro scale, if one unit in a defensive posture retreats (i.e. ignoring anyone who's just retreating through the defences), that would seriously undermine the defence of the whole hex. I wonder if that could be modelled as well?

So, high friendly losses will not usually represent success by the attacker on the ground, but rather just that the attack was heavily supported. Again, bombardments don't gain ground - that is done by ground forces. I still say that odds are a more realistic factor.

But isn't this covered by the AR factor? No forces on the ground will mean the bombardment has little effect.

If that were true we wouldn't even be having this discussion. A unit that is taking those kinds of losses per round would quickly be evaporated or reduced to RBC size in a few rounds.*

Whatever the absolute numbers, I find that a well-supported attack generally deals 2-3 times the percentage losses than it receives. Just that there are normally a lot more than 2-3 times as many attackers.

*(Note that RBC ability has been enhanced under 3.4 - a factor that was expected to counter some of the defender enhancements. Players need to make sure that they don't overlook that).

This does help- especially with that one 1-3 unit that just won't budge. Normally it will on the second round.

In fact, the round may end with all attackers broken off, yet still force a retreat due to the losses incurred.

So make the game test AR again at the end of the attack. If AR is below 100, no quality check for the defender.

It wasn't just because both sides were heavily entrenched that there was a stalemate. Even during the mobile phase it was much harder to gain ground than in other wars.

Well. In 1914, no-one really understood the new world of battlefield tactics. So comparatively small forces were able to hold positions that were otherwise untenable.

Then in 1918, the terrain was simultaneously incredibly difficult to move through (due to the churning up of the soil caused by four years of shelling and deforestation) and easy to defend (due to four years of building line upon line of reserve trenches).

[Curtis Lemay]

ORIGINAL: golden delicious

It's involuntary on the part of the player. However it may be a decision made by the commander of the unit: he sees his men getting killed to no purpose and decides to take a risk on his career to get the unit out.

In game terms, that is either a "minimize losses" or "limit losses" setting. "Ignore losses", on the other hand, does not give the local commander that option. The unit itself is under orders to, well, "ignore losses". Only the unit (including the commander) breaking allows for a retreat. And since the chances are tied to the unit proficiency, the units most able to make cost/benefit choices (the best ones) are the ones least likely to do so. Again, clearly, no cost/benefit decisions are involved in RFCs by "ignore losses" units. They stand until they break.

Thinking further about the micro scale, if one unit in a defensive posture retreats (i.e. ignoring anyone who's just retreating through the defences), that would seriously undermine the defence of the whole hex. I wonder if that could be modelled as well?

Yes!! Base the RFC on ending Combat Odds! [:D]

But isn't this covered by the AR factor? No forces on the ground will mean the bombardment has little effect.

The AR only needs a max of 1:1 odds to guarantee full benefits. From that point, support may be piled on in unlimited amounts.

Whatever the absolute numbers, I find that a well-supported attack generally deals 2-3 times the percentage losses than it receives. Just that there are normally a lot more than 2-3 times as many attackers.

I don't think those numbers are reliable. Especially since there would actually be no need to commit 2-3 times as many ground attackers - in fact, no more than 1:1.

So make the game test AR again at the end of the attack. If AR is below 100, no quality check for the defender.

So if the AR is less than 100, there is no chance of the defender retreating? It would need to be even more sophisticated than that. In fact, the combat odds would be the ideal final check. [:D]

Well. In 1914, no-one really understood the new world of battlefield tactics. So comparatively small forces were able to hold positions that were otherwise untenable.

Then in 1918, the terrain was simultaneously incredibly difficult to move through (due to the churning up of the soil caused by four years of shelling and deforestation) and easy to defend (due to four years of building line upon line of reserve trenches).

Doesn't matter what the reasons were, WWI was just different from WWII in terms of ability to force units to retreat. TOAW has long needed a means to model that. This fix could do it.