[Best approach to a hedgehog defense?]

All right. I think some of this is starting to sink in. Here are the three biggest points (IMHO) about this type of defensive situation that I've managed to grasp so far:

1. Reserves. In a fortified line, it is relatively difficult to pull units off of the line to respond to a breach or threatened breach elsewhere. For one thing, it's hard to do this when the units are facing in the direction that the enemy is generally advancing. And even if you do it, such units have a long and potentially dangerous distance to cover to get to the breach. So, in that type of defense, you have to keep more of your initial force in a mobile reserve -- or respond to any breach by falling back.

In the case of a hedgehog or circle defense, it is easier to pull units off the line when they are completely on the other side from the breach, and it is a shorter route they have to travel. This allows the defense to get away with a smaller initial mobile reserve.

The attacker of course knows this about the defense and may attempt appropriate countermeasures -- for example, by slowly dropping HE into the center of the defense to disrupt movement

2. Withdrawing to an inner line. In a hedgehog or circle defense, the idea that one can plan on an orderly withdrawal to a new strong point (i.e., the "castle keep") is probably a mirage. If you have a properly strong initial defense scheme (see point #3 below) and the attacker still gets a firm fire base established on your perimeter (also see point #3), you will be prevented from making such a withdrawal. OTOH, if you are able to make such a withdrawal, it probably means that your initial defensive scheme was weak and you are only delaying the inevitable.

3. Fire plan. Perhaps the hardest point to sink in with me is this: that in setting up your initial defensive scheme, you need to ensure you have at least as much firepower directed at your own perimeter as out beyond the perimeter. You have to keep the attacker from holding a firm position on your perimeter, because he will be able to keep feeding new troops and ammo into that position, and eventually overwhelm you with protected firepower. The significant of the "castle keep" is *not* that it is a place of refuge and a new defense line, but that it is a high point from which fire can be directed onto an attacker that has breached the outer walls. And the amount of firepower you need is large, because the attacker is in a more protected zone once he gets into your perimeter.

On the attacker's side, he has to assume that there will be a lot of firepower hitting him as soon as he manages to break into the perimeter. For that reason, he has to be careful that he doesn't leave himself vulnerable to a heavy HE barrage or the like.

OK, how am I doing so far? Anything major I've missed?

[Best approach to a hedgehog defense?]

Originally posted by Andreas:

Ah, but in TDR you have a much much better chance, and better options than in CH. I think it is by far the better scenario.

[/QB]

Probably so. In truth, I've only played CH against the AI, and I figured out how to beat it by taking advantage of a stupid AI trick. OTOH I've only played TDR hot seat. It was while noodling around with the defensive setup that I started to think about the problem in a more general way. And the fact that you didn't give the attacker any obvious tools for breaking open the defense (like heavy HE for example ) got me started thinking about this type of problem from the attacker perspective. It is, of course, not the sort of heavily fortified position that Jason has described in his posts (except for those two dug-in T34s, of course ). But perhaps the terrain features make up for that in a way, given the scale of the battle.

I'm glad that you think this is turning out to be a worthwhile thread.

[Best approach to a hedgehog defense?]

Jason: Everything you say makes a lot of sense. But let me probe a little further on this specific point:

Do you see an essential difference (from either the defender or attacker POV) between a hedgehog defensive position and a fortified line (i.e., flanks on opposite map edges rather than a circle or near circle)? My quick take on your last post suggests to me that you'd have written exactly the same thing if I'd asked about the problem of a fortified line.

It strikes me that a hedgehog defense is something like the medieval castle defense. That was how I got to where I did in my last post. It seemed to me that the defense would be weighing the decision of counterattacking to retake the ramparts, or withdrawing to the keep. And the attacker would be weighing whether to focus on holding his position on the castle walls or throwing everything into a thrust at the keep before the defender could get established there. In both cases, it has to do with how you employ your reserves.

[Best approach to a hedgehog defense?]

Andreas: Let me quickly say that I never assumed The Dirt Road was intended to be an example of a hedgehog defense. That was why I only named it when you posted. It was just that playing it (and studying it a little) started a train of thought which led to this thread. I do admit though that the "parameters" I listed were perhaps too much influenced by that scenario. I think Jason's first post was particularly helpful in redefining the question into something more general.

We have had other threads discussing other forms of defense and related attack strategies: e.g., fortified line, pak fronts, reverse slopes, echeloned defense, etc. I thought that perhaps it would be useful to single out this other type of formation and to see what knowledgeable people had to say about it.

IIRC, you are also the author of the infamous Cemetery Hill scenario. There are some similarities between the two scenarios it seems to me -- notably in how meager the offerings you give the poor German attackers.

[Best approach to a hedgehog defense?]

Thinking about this some more, I guess that the key tactical problem in a hedgehog defense (or what I am calling a hedgehog defense -- which may in some cases be little more than a "circle the wagons" situation) is that it is particularly vulnerable to exploitation through a breach in the main defensive line. If an attacker can penetrate far enough into the center, there's no place for the defender to withdraw and the whole defensive position turns into a rout. On the other hand, the defender has a relatively concentrated formation which *ought* to be make it easier to get reserves to the breach.

On the assumption that a competent attacker is going to make a breach somewhere in the line, my thinking is that the defender should have decided in advance whether he will most likely use the reserve (1) to contain the breach while the main line withdraws to an interior defensive position, or (2) to stage a counterattack to close the breach and reestablish the original defense line. This decision would have to factor in the size and type of resources available to the defense, the terrain, how much the position has been fortified with trenches, wire, mines, etc.

Likewise, the attacker has to anticipate whether the defense is likely to try (or have the capability to try) such a counterattack. Part of this would include looking at the attacker's ability to disrupt the formation and positioning of a counterattacking force. Depending on this assessment, the attacker might organize to facilitate exploitation as soon as a breach is made, or to maximize resistance to a counterattack.

Either side could guess wrong, of course, and things might play out differently, but I'm mostly interested right now in how one would construct a basic game plan for this type of battle. Just thinking out loud at this point ...

[Best approach to a hedgehog defense?]

Jason: I was hoping you would weigh in on this one.

The idea of coming at the defense from two adjacent vectors seems right from the experimenting I've been doing. Shifting a main body from one vector to the other is not always that easy, but that may be more reflective of my skill than anything else. Of course, a good defensive system can also make that more difficult with ranged fire (especially since the CM model makes it more difficult to move laterally under any kind of fire). But you've often said that it's usually a mistake to worry about picking the *right* place to attack, as much as to just pick a place and hit it hard.

Your answer does suggest to me that there may be two levels to this question. One level (which is what I've been working with) is smaller in scale. That is, the defender isn't going to have to move troops around all that much to backstop a threatened sector. Rather, the defender will be trying to conduct an orderly fallback toward the interior while giving the attacker a severe bloody nose in the process.

The other level (which is what I think you are mostly addressing) is larger, where the defender has to worry about a sector collapsing and having the whole position blow up as a result. If I am right about this distinction, then (like Sergei's point about salients) it does give both defender and attacker some logical framework for thinking about the problem in setting up for a given battle. That's exactly the sort of thing I was hoping to get.

[Best approach to a hedgehog defense?]

Bigduke: In this case, vodka might work better, but I drank the bottle on the way to the battlefield, so there you go ...

[Best approach to a hedgehog defense?]

Andreas: You may be right that my parameters suggest an unreasonably difficult problem for the attacker. That is partly why I wanted to start this thread. In other types of battles, I find it fairly easy to tell if the nut can be cracked (or, alternatively, if the defense can be held), but I realized I didn't have that level of certainty about a hedgehog situation.

The principles you suggest make a lot of sense, but can be hard to do. I *think* the reason is that the defense is so concentrated it can be hard to suppress all of the defender's ranged firepower (mostly MGs) in order to close on one segment -- if you don't have heavy FOs to make the defender pay for being so concentrated. My best thinking so far is to use smoke to screen off some of that firepower. But that only gives you so much time to break into the defensive line.

By the way, this thread was stimulated by my spending time with one of your excellent scenarios -- The Dirt Road. I'm sure that the scenario is not nearly as "unrealistic" as may be suggested by the parameters I described. Because I am playing it hot seat, I got to thinking about both questions of "best defense" and "best attack."

[Best approach to a hedgehog defense?]

Sergei: Sorry if the question seemed vague. It was stimulated by my studying a particular scenario, but I did not want to start a "how do I solve this scenario" thread. I was more interested in general principles and theory.

I like your point about salients. I'm not sure it's helpful to think of the whole defense as a salient, but it's reasonable to assume there would be one or more salient positions in a hedgehog defense (a perfect circle being highly improbable given the randomness of terrain). Given that, I assume a competent defender would want to organize so as to give strong protection to the flanks of such salients (which the scenario designer did in the default defense by the way). Of course, that might weaken other sectors. That suggests an attacker might want to probe both a salient and a non-salient to see which seems most vulnerable.

Also, the best use of smoke is a good topic. The scenario in question has only light FOs. I came to the conclusion that they were only useful for smoke in this situation, so I have been experimenting with different ways of getting the most from that resource. It does seem that smoke is best used to screen off one segment of the defense from the rest but that's just my best guess.

[Best approach to a hedgehog defense?]

I don't think this has been a topic before, and I think it might be interesting to get some views from people. I'm actually interested in both sides of the question: that is, what is the best way to organize a hedgehog defense, and what is the best way to attack it.

A few parameters: I am talking about a defense on the scale of a medium-sized CM game, not something representative of, say, the large hedgehog defenses on the Eastern Front. The point is that the defense has no flanks and is at least mostly surrounded (or could be). I am assuming the defense is organized around suitable terrain (one or more hills, good cover, lots of clear terrain around the position). I am assuming the attacker has limited FO or direct HE capability, and therefore must rely on some sort of attack. Similarly, I'm assuming the defense is relying heavily on MGs and infantry with some limited HE.

Some sample defense questions: Is it best to guess where the attack is coming from and concentrate firepower there, or spread the firepower evenly around the position? Do you like to have the infantry on the edges with MGs, etc, to the rear, or keep the infantry in the middle of the defense? Large or small reserve force? What else should one do?

Some sample attack questions: Is it best to concentrate everything against one segment of the defense, or come at it from two directions (perhaps with one being a feint)? Is there a way to get leverage on a hedgehog position similar to a flanking attack, or is it just a matter of trying to bludgeon your way in? Any other preferred approaches?

[Real-time, turn based engines and maximum game size.]

Originally posted by kipanderson:

Do remember that even with the WEGO option on the underlying engine will still be real-time in CMX2. All that is happening is that every minute the real-time game pauses to take on new orders. But even with the WEGO option the turn is not calculated first as in CMX1. In CMX2 when both players have hit the Go button the real-time engine just kicks in and off it goes based on the orders given during the pause but processing as it travels through the minute to the next enforced pause.

OK, so with the real-time option, the graphics data is pumped immediately to the computer display rather than being stored and then shown as a "movie." I don't see that as a big change, except that the real-time option wouldn't allow battles so big that the calculations take longer than a minute (wouldn't be a problem if you choose the WEGO option). The other change (at least I assume this) is that in the real-time option you have to give your orders while the game flow continues, rather than freezing everything at sixty second intervals or at your own "pauses". Again, I don't see that as a big change to the engine. The fact that the graphic display of the calculations is presented after the calculations are complete does not, to my mind, make CM a turn-based game. A turn-based game would be one in which we were taking turns. I still think that CM is fundamentally a real-time engine now. However, it's not important that we agree about that. I'm just saying that I think you should not be concerned as long as you are still willing to play in the WEGO style.

[Real-time, turn based engines and maximum game size.]

Still think there's some confusion here. BFC isn't crippling the engine in order to have a real-time option. The game already mostly runs in real time. Yes, in a real large battle, the calculations can take more than a minute to complete, but that doesn't matter because you could never play a real large battle in real-time anyway. Too many units to keep track of. Maybe there will be some sort of limit on the size of the battle if you choose the real-time option. I also don't think that BFC is punting on significant StratAI improvements because of the real-time option. I think they are punting on those because they don't want to invest their time there.

[Real-time, turn based engines and maximum game size.]

Are we perhaps confusing the nature of the game engine with the style of play. Most of what goes on in the game engine is done in real time now. The TacAI, which is a huge component of the engine is all real time. Spotting is real time. Combat results are calculated in real time. I've seen pretty compelling evidence that the AI even gives movement orders in real time now (not just in the so-called "orders phase"). If you can think of something that will change in the game that conforms to "the accepted wisdom on this forum that one of the hits you take when moving from a turn based engine to real-time is a “massive” decrease in game size," I'd be interested to know what it is.

I'm still not exactly sure what BFC is contemplating for the "Real Time option" but I have to assume it is mostly a style of play question -- that is, you don't get to halt the action while you give orders. And yes, I assume if you prefer to play that way, you won't be commanding a battalion when you do it.

[We need to have something like a "Retreat" command]

Originally posted by C'Rogers:

But as Juan pointed out, with the ability to specify movement order priorities (and perhaps how the movement is carried out) this whole discussion may be moot concerning CM:SF.

I guess I am missing just what the thinking here is. As I understand it, the major change is that we will be able to specify various actions for each waypoint, but that nothing will change as to what happens between waypoints (I think there may have been some players asking for BFC to add the ability to micromanage what happens between waypoints, but I'm not aware that BFC ever concurred).

Even now you can give a unit a covered arc facing the enemy which has an effect between waypoints. Giving a unit a covered arc when it is moving laterally or away from enemy fire does not appear to have much benefit, except when the unit stops (as ordered or when pinned). Then the covered arc helps to keep the morale problem from getting worse during subsequent fire. This, I'm guessing, will be the same in CM2. The only difference in CM2 will be that you can change this covered arc at each waypoint. However, this appears to be the only change -- and even then we're talking about orders that affect what happens at a waypoint, not between waypoints. For that reason, I have to assume that BFC is going to carry forward the idea of flank and rear fire having a very adverse affect on morale, including during movement

Side note: I've started thinking that this is a contributing factor -- a relatively small one maybe, but a factor nevertheless -- to the relatively high casualty rates in a CM battle. As it stands now, in many cases the best tactic is to leave forward units to their fates and hope they take a reasonable proportion of the enemy with them.

[We need to have something like a "Retreat" command]

Here's the problem in a nutshell: AFVs have a reverse command, and infantry units do not. But the need is the same, because the way in which CM works (and this is still likely to the case in CM2), infantry units are more resilient to fire from the front than from the side, just like with AFVs.

[A Dumb Question]

There is another technique for learning which doesn't get mentioned much, perhaps because it doesn't occur to everyone to try it: to play both sides of a battle using Hot Seat. I do this quite a bit (particularly as I find playing the AI to be almost worthless, and I don't have much opportunity to play H2H). Playing the game this way isn't to everyone's taste, of course. But in the more specific context of learning, there are two benefits you get from this approach.

First, it makes it easier to learn exactly how CM works -- for example, things like spotting and how to avoid being spotted, the effect of different amounts of FP from different ranges on different types of targets, how the TacAI deals with different types of threats and opportunities, etc. You don't have to guess about these things, because you can actually see the results from both sides.

Second, having figured out the mechanics of CM, you can more accurately test the effect of different tactics, since you will be able to see precisely what the impact is on the other side, compared to what you might have expected.

There is a third benefit, of course, which probably goes without saying. You never lose.

[We need to have something like a "Retreat" command]

Not surprised that the way I titled this thread is causing some problems with understanding the issue I'm raising. That's OK. We'll work our way thru it.

I chose to focus on contrasting the ability to Advance directly forward versus Advancing directly backward, because I think this makes the issue clear. However, keep in mind that this problem exists for any movement that is anywhere from roughly 90 to 180 degrees from a firing enemy (this is why re-jiggering the Withdraw command isn't relevant).

It may be the real source of the problem is that the model makes fire from the flank or rear too devastating on morale in all cases, not just movement. This may have worked OK in CMBO, when infantry morale wasn't very brittle to begin with. With the change in behavior modeling in CMBB, the effect from flank or rear fire became too pronounced.

On the other hand, if the way the CM models the effect of flank or rear fire is reasonable in most cases, but not when the target is a unit that is moving cautiously (i.e., MTC) or moving while expecting fire (e.g. Advance), then maybe the problem can be addressed by modifying the model for those commands.

I thought a different command might be necessary because of the kind of considerations that Cassh describes -- that is, to capture the fact that there is some negative to moving laterally or to the rear when under fire. That is why I titled the thread that way.

[We need to have something like a "Retreat" command]

Originally posted by MikeyD:

The CMx1 'Withdraw' command works as long as you don't withdraw under fire and don't go far. In other words, as the tank swings its turret towards the building your troops take the opportunity to dash out the back. The same movement in the middle of an open field being raked my mg fire doesn't exactly work.

We could make a laundry list of troop movements that aren't handled particularly well by the CMx1 AI. The ability to withdraw in a coordinated manner was not much worse than the ability to attack in a coordinated manner! But CM was always more of a 'big picture' company+ scale engagement game than squad-scale. That's why we were willing to put up with those abstracted squads. Now CMx2's got 1-1 representation and is geared towards a smaller scale (no more picking your forces by the Battalion). Less wiggle room for BFC on 'abstracted' unit reactions.

Nope, not addressing the specific point I am trying to make. I contend that using the Advance command to move back from a covered position to another covered position should work about the same as using the Advance command to move in the opposite direction. I'm looking for people to either agree with that, or to tell me that IRL a squad moving to the rear under fire was much more likely to be badly messed up than if moving in the opposite direction.

The performance of the AI is not relevant. I mentioned "abstraction" to add color to my understanding that Advancing to the rear would involve the same sort of leap-frogging movement as Advancing to the front.

[AI]

Originally posted by mav1:

SteveP, what do you think of BfC's approach to CMx2. What I was hopeing for was an improve CM on a grander scale, with a campaign {to replace operations}. But they have gone for smaller action, based on urban warfare {I think}.

Also what are your ideas on how to make a campaign for combat mission. I would like to compare your ideas on how to do a campaign with my ideas.

If you did a search you'd find that this is a very old debate in this forum. There are a wide variety of personal preferences among the people who play CM, and BFC can't please them all. My own view is that I'm happy to see BFC working toward getting a game that produces very good results when played at a relative small scale (also it's a lot easier to play). Once they do that, then I think it makes sense to look at how best to scale it up.

[AI]

Originally posted by mav1:

I agree with you SteveP, that how well the AI work's is down to the scenario designer. A few of the designers in the scenerio's included in the game got it wrong. But I wonder if you give more time to the AI to think, could you improve the AI. If during the time the player is thinking of his/her move. The AI uses the time to think of its move (dont know if this is done already). This approach is done on a game called Galactic Civilation, its supposedly got good AI because of it. Dont know if its relevent, but its a suggestion.

I don't think it helps much to compare CM's AI to other games, because different game designers approach the problem with different design strategies. My understanding of CM's AI is that it tries to get an optimal result (using "fuzzy logic") from an accumulation of values that are either set in the code or are situational. BFC has to make improvements to that incrementally, in order to know if any one change is working successfully within a reasonable range of acceptable outcomes. IMHO, the performance of the AI could be significantly better just by making two improvements: recognizing command relationships and keeping units in command radius to the extent possible (which may be in the works based on things that Steve has said in other threads), and getting the AI to use other movements commands besides move and run (or fast for AFVs). Beyond that, I'm not sure what is realistic to hope for.

[We need to have something like a "Retreat" command]

*Bump*

I suppose everyone on this board plays only meeting engagements.

However, I think this is a significant problem in the CM design since CMBB came out, and nothing I read in this forum convinces me it won't be a problem in CM2.

[We need to have something like a "Retreat" command]

Having given a lengthy description of the problem I see, let me add to it a more concise one:

Either a unit should be able to Advance laterally or to the rear with the same, or nearly same, level of protection from fire as when Advancing toward the enemy (because the movement drill is basically the same) -- in which case, the model needs to be changed.

Or, it is more realistic that a unit moving laterally or to the rear is going to be hosed regardless of what movement order has been given -- in which case, the model is working just fine.

[We need to have something like a "Retreat" command]

Or "Advance to the rear" if you prefer. Allowing for the fact that it's hard to know how the new modeling for infantry will affect things in CM2, I want to point out a problem in CM now that I think might be carried over if not addressed.

It is clear that an infantry unit's morale is adversely affected when it is fired upon from its flank or rear. I cannot tell if the effect is increased as the fire moves toward the rear, but it may be as simple as any fire from 90 to 180 degrees from the target's facing will increase the morale impact by a set factor. I recently determined through testing that this also occurs when the target is moving. For example, a unit advancing laterally across the front of a firing enemy (say crossing over from one building to another directly opposite) is much more likely to panic, break or rout than a unit advancing directly toward the enemy. The same thing happens, of course, if the unit tries to move back to a rear position. Basically, by turning its back to the enemy, it has a very low probability of making it to rear cover in good order.

AFAIK, the Advance command is intended to be an abstraction of a leap-frogging movement that reduces the unit's average overall vulnerability to fire. The unit can fire while advancing, but fire isn't essential to using the Advance command successfully. If a squad is using this same leap-frogging movement to move toward the rear, it should be receiving the same, or nearly the same benefit. I would, of course, argue the same for lateral movement as well.

The problem now is that without the ability to make an orderly retreat from an exposed position, units that are out in front of a MLR have a high likelihood of never getting back, if doing so involves being under fire for any length of time.

[AI]

Originally posted by Cpl Steiner:

Having said that, I would have thought you could still set up a viable CM:SF scenario in which the Syrians ambush a US column, with the player taking the US side and the computer taking the ambushing Syrian side. The Syrians would have fixed setup zoned dictated by the scenario designer, and would just have to wipe out the US column from their starting positions or die trying. As the US side, the player would have to get out of the ambush or destroy the ambushers. I can imagine this would make for quite an exciting scenario. [/QB]

Clever scenario designers who understand the limitations of the AI have been able to create situations that resemble ambushes. Moreover, a lot of players (and I suspect a lot of scenario designers) don't realize that the AI automatically hides all its units at the beginning of a battle when it is on defense. It also apparently has the ability to unhide them in the middle of a turn (something a player can't do). It does this on a semi-random basis. It's a kind of pseudo-ambush effect that may have been intended for this purpose by BFC. I don't like it, because it ends up being a stupid AI trick most of the time, but we'll see what they decide to do in CM2.

Clever scenario designers have also been able to create situations that help the AI to attack or do meeting engagements. The best trick is to reduce the amount of movement the AI needs to get into battle. It's the movement toward contact phase that screws things up. That may be why in QBs the map is relatively shallow (more so than one would like if playing the defensive side). With a shallow map, the AI doesn't have to move as far.

[Camouflage And Hiding]

Originally posted by Peter Cairns:

SteveP,

Just because thats how they are treated in CM1 doesn't mean that CM:SF needs to follow suit, I'd be happier with a system based on size rather than class, and one that treated a mortar like an HMG.

Peter.

All that remains to be seen, of course. My guess is that the disparity between MGs and small mortars will be corrected, at least to some extent. My real point was the difficulty of treating small profile or small size AFVs the same as large guns (which do not have a profile rating) for ease of being spotted. For all I know, it may be easy to give guns a profile rating that is consistent with AFV ratings. Or it may be a similar problem to trying to equate the effects of the BV of a small mortar shell with the effects of the FP of a high caliber HMG, even though IRL the effects were probably very similar. I don't know how that will play out in CM2.