Matrix Games Forums

At the request of Mziln, I have moved the discussion on the value of a hex to this new thread. Here are a posts (some are excerpts) that lead up to the discussion in AI for MWIF - USSR.
==========================

I am coming around to the view that the USSR should have a defensive line stretching from north to south with the northern end maybe floating in the air. The USSR would depend on terrain and scarcity of German units and/or supply to prevent flanking manuevers there. [There was one Germany vs Russia game I use to play a lot where the Russian routinely hinged his line on the Terrain Effects Chart that covered the map north of Moscow.] There might be isolated pockets of resistance (e.g., Leningrad, Sevastopol, Rostov, Kerch straits, Caucasus). These pockets would occur as the USSR line retreats east seeking better terrain, a shorter line, and simply distance from the German units that are in supply.

To remain viable, the USSR needs to have enough units to man the front line - at least so it isn't easy pickin's for Germany. He also needs to have enough production capacity to replenish losses. The last may be supplemented by lend lease. The usual state of affairs is that Germany attacks during the good weather turns and kills Russians faster the they can be replaced. The USSR retreats during good weather turns to prevent this. When the weather turns bad (for the Germans), the USSR stops giving up ground and can replenish his unit count because losses are now fewer than reinforcements.

Do you have a sense of what the burn rate for USSR infantry corps is during good weather impulses (or turns)? This would answer the question of how large a production capacity the USSR needs for sustaining his viability.

_____________________________

Steve
=======================
in my experience it varies from game to game, heavily depending on the year barbarossa is done and where the germans happen to be pushing. as a rough estimate id say 8-12 INF/MIL/GARR corps in summer, mabye a few DIVs, and planes if the Russian air force decides to fight. usually it is more losses for the russian in a 1941 and 42 barbarossa than a 1940, although russia is so weak in early turns that those few losses sometimes are much more valuable units like HQ, ARM, MECH, and guns. if going north the germans will likely inflict less losses because the russians will be in cities, bad terrain, and bad weather. on the other hand, going south results in lots of clear terrain that the ARM and MECH can blitz to all hell. in bad winter turns it is sometimes better for the german not to attack, because he cant destroy more units than the extra BPs the russians would get for the PM bonus, so the Russian AI should expect lower production during winter.

_____________________________

Member #3 of the EBEA
Comrade #4 of the e-Socialist Liberation Army
=======================

Your concept of lines is correct too. They tend to be ZOC lines very soon after Barbarossa starts. There is a game dynamic where counters are compressed very tight at the border and become much thinner as Germany pushes into the USSR. In recent history I've always managed to hold the Rostov city line so I'm not familiar with what happens after Rostov goes.

=======================
In general I think that this front line approach is good, but many times you see a USSR that is not able to hold a whole frontline and then you will neet to retreat back with several subfronts and employ a ZOC defense. or maybe hope that bad weather/poor supply or sheer distance will stop the axis from advancing to far. It is very far from moscow to Sverdlovsk and your reinforcements might be able to make it before the axis reach it.

After losing Moscow & Stalingrad which you should count on doing against a competent human player I would try to defend the areas described in previous post(Leningrad, Rostov, Sevastopol, Caucasus and Ural)

When reinforcements arrive you can build on thoose fronts and eventually link them up and advance against the axis.
=======================
This is not totally correct as the game have a fixed end-date, the axis can also trade space against time in the latter part of the game, and it is hard to advance quickly against a solid axis line.
To advance from the polish border to Berlin between '43 and '45 is much easier than to advance from the asian map to berlin in the same timeframe.
=======================
Ah, you're coming around to the problem that I believe is at the heart of the matter: what is the value of a hex. Here we are concerned with the value of a north-south hex line separating Berlin from Sverdlovsk. Is it better to sacrifice 6 garrison and militia units and only retreat one hex eastwards or to lose 3 and retreat 2 hexes eastwards? The same principles apply for the Germans once the USSR starts counterattacking and pushing them westwards.

Here is how I think the AIO can solve the problem. If we know the acceptable burn rate per turn (losses in corps sized units) and can estimate the number of impulses in the turn, then we can adjust how many hexes we give up (east to west hex rows actually) to keep the burn rate per impulse to an acceptable level. I view this decision as being at the operational level.

At the strategic level, the USSR AIO is in desperate defense because of his numerical inferiority (both total land points and total air points). Operationally, he withdraws his line to minimize losses - trading space for units. Tactically, he needs to build a good defensive line. There is also the other consideration about not giving up so much space that he loses too much production capacity and/or the ability to push the German all the way back to Berlin before the game is over. I think the burn rate can be used to control the speed of the retreat so it is neither precipitous nor fatally stubborn.

_____________________________

Steve
=======================
Working in terms of BP per side and per major power is easy for the AIO to do. When playing over the board, I always track that and the total strength points on board and in production by branch of service. Simplistic predictions for all these numbers are also easy to do and pretty accurate to boot. It's the number of hex rows (stretching north to south in this instance) that are difficult to include in the decision logic. In France every hex is only given up after a fight and retaking it is hopeless. That is clearly not the case in Russia because of the luxury of trading space for units and/or time.

_____________________________

Steve
=======================
OK, I've been trying to get my mind around how to determine the value of a hex. Honestly, I'm having a tough time with the concept. Can you give me a little more information about how it will be used?

When I think about defending and attacking in Russia it depends on whether its the early or late game.

Early game (as has already been noted) is about preserving the Russian ARM units. There is no hex for which I would sacrifice an ARM class corp. Otherwise, I think about establishing 'hard points'. A hard point is a hex where it is very difficult to attack. In general they are hexes that eliminate offense ARM bonus and/or blitz attacks. Sometimes they may be a clear hex protected by a river where I could put an ARM class corp and two AT guns. (Using the 2D10 table) they are characterized by ...

=======================
Here are some excepts from the current AIO design document about the value of hexes. Because they are pretty obvious in Russia, I have omitted related sections about which types of units to produce and how to deploy them into the frontlines. Also missing is my preliminary analysis about cutting supply lines, since that is not in a coherent form yet. Note this is all a work in progress and comments, suggestions, and criticisms are always read seriously.

P.S. I know this is long but in my opinion, handling the tactics of land combat is the most essential element of getting the AIO to play well.

======================

2.9 Hexes
The ultimate goal is to be able to evaluate the risk of losing units in combat versus the gain of capturing hexes close to enemy capitals, resources and factories. There has to be very strong motivation in terms of combat values (CVs) for the AIO, playing Germany, to drive on Paris and Moscow and to contemplate invading Britain. Victory cities only play a role in the Grand Strategist decision making.

All of this leads to the value of land hexes. Now in and of themselves, hexes usually have no value. The exceptions are some specific hexes (capitals, resources, factories, rail lines, and victory hexes) and to that list we can add sources of supply when supply is hard to come by. But even the most mundane of hexes can rise in importance when the battle lines draw near. It is as part of a front line that hexes achieve their primary importance. Looking at Germany versus France in 1940, Germany versus the USSR in 1941, and China versus Japan in any year, the outstanding characteristic of the land combat is where the front lines are. What exactly is the value of a hex in the front line?

When on defense, we can count hexes defended by rivers as worth double “the average CV for the front line” since they effectively double defensive strength. We can count forest hexes as half the enemy’s average tactical air CV since they cut ground strikes and ground support in half. We give cities a bonus for forcing the enemy to use the assault table. We penalize clear hexes if the enemy has armor capable of overrunning our weaker units. And there are similar adjustments for other types of hex and hexside terrain. But when on the attack, all of these change in importance. It is better to attack weak hexes and force the enemy to abandon strong ones because of the threat to his lines of communications.

For both attackers and defenders, maintaining a continuous line is very important. Only under unusual circumstances can the goal of maintaining a continuous line be put aside.

8.25 Estimate land combat odds and losses
Weather, action choice/activity limits, and initiative affect likely CV losses. Those calculations can be done under the various conditions and probable losses (weighted by the probabilities for the weather et al) figured out. There are two ways to go here: (1) take the straight probable losses, or (2) take the minimal losses under the worse conditions.

8.26 Choose attack hexes - normal; s.11.16 (4)
Primary tactical stance
The FM receives information from the JCS which includes whether he should be on the offensive or defensive against the countries opposing him.

Front lines
Assuming that the primary tactical stance is offensive, the FM determines where his front line is (front lines are) for each enemy country. For example, at the start of the Global War scenario, Germany has one front line facing France and Great Britain and a second facing Poland. Against France and England, Germany is strongly defensive and against Poland, going in for the kill. Note that the front line includes coastal hexes that can be invaded and hexes behind the front line proper that can be attacked using paradrop and glider units.

Attacking units
Focusing on a single front line, the FM determines the land units he has available with which to attack. That is, the units are face up, in supply, and can reach an enemy zone of control this impulse. Sorting these units by combat strength and separating the corps from the divisions, he creates hypothetical stacks of 3 units per hex, again arranged from the best through to the worst. This is just a first pass for estimation purposes, since the units may or may not be able to actually form up in those groups because of their positions in the line. What he now knows is the highest number of combat factors that he can bring to bear on 1, 2, 3, or more hexes.

Defending units
Switching over to the enemy front line, he examines each hex that is in the ZOC of an enemy unit and sees if he can move units into it. Since being able to move into an enemy unit’s ZOC means you can attack the enemy unit, this determines which enemy hexes he can attack. Paradrops and invasions are slightly different, but a comparable procedure is used. For each attackable enemy hex, he calculates the defensive strength of the hex. As part of this calculation, he includes in the defensive strength the number of shifts (because enemy units are disrupted or there are forts present), the number of units in the hex, and the hex’s susceptibility to the use of any special units the FM has available. For instance, whether using armor or an engineer might have an effect.. Lastly, this calculation looks at the effect of terrain on the combat (e.g., assault table mandated) and counts the number of hexes from which the FM can attack this impulse (an attack across a river would count as half a hex).

Weak and strong points
For each hex that might be attacked, the FM makes a crude estimate of the odds he can achieve if the best possible units were to reach ideal positions for the attack. These estimates let the FM rank the hexes according to their vulnerability. Basically, the enemy weak points and strong points have been determined.

Disruption status
One of the subgoals of the FM during a turn is to turn all of the enemy units face down. Though this achieves only a temporal advantage, it can be decisive during long summer turns. The following items relate directly towards achieving that subgoal.

At the beginning of each impulse, the FM makes assessment A1 for each of his frontlines and records for each side the number of: (1) face up/down corps, (2) face up/down divisions, (3) face up/down tactical bombers, (4) face up/down strategic bombers (re: carpet bombing), (5) face up/down fighters, (6) face up/down ATRs, (7) units capable of immediate paradrops, (8) units capable of immediate invasions, and (9) offensive chits available. In addition, the FM identifies which units are out of supply, and the supply lines for units that are in supply. Other needed information is the likely number of impulses remaining i the turn, likely weather for the rest of the turn, and reinforcements available to both sides.

Tactical mode
What the FM needs to determine is which of 4 modes of attack to use: (1) destroy enemy units this impulse, (2) disrupt enemy units this impulse so they can be destroyed in a later impulse (or turn), (3) maneuver so better attacks can be made in the future, or (4) push the enemy back. Maneuvering can change supply status for friendly and enemy units, increase the number of hexes from which to attack an enemy hex, and improve the selection of hexes from which to attack. In situations where the enemy has the ability to counterattack, maneuvering can improve the FM’s resulting defensive line at the end of his impulse.

The calculations of estimated results provide the expected changes (for both sides) to the front line. This includes kills and disruptions. In addition to disruptions, units can also become face down because they were committed during the impulse (e.g., air units). And face down units might be reorganized. When the impulse is over, the FM makes a new assessment, A2, for each front line., and compares it to A1.

By extrapolating the change from A1 to A2 over the remaining impulses in the turn, the FM judges whether disruption is a viable tactic for the current turn. It works if the enemy units are mostly face down with no reorganization capability left while the FM still has a viable attacking force face up.

Direct attacks to kill enemy units is usually the best tactic but it might cause too many friendly casualties or disruptions. Or, if the FM limits attacks to only those with excellent odds, it might be too slow. Yet again, maneuvering is rarely fast and there might not be enough impulses in the turn to use the disruption tactic. The FM must be willing to accept that none of the modes will work as well as he would like and simply go with half measures, or no attacks at all.

4.3.8 Field Marshals
Based on the assessments and estimates of land units only, decide on your situation: (1) desperate defense, (2) strongly defensive, (3) somewhat defensive, (4) balanced, (5) somewhat aggressive, (6) attacking, or (7) going in for the kill. To some degree the 1st and 7th will depend on the proximity of objective hexes to the front line.

I will ignore paradrops and invasions for the moment. They need to be analyzed separately and from two perspectives: (1) opportunities to attack and (2) need to defend against possible attacks. For now, I will assume that the only units involved in an offensive are coming by land.

The value of a hex, and of a front line in general, depends on the tactical balance in the theater of operations. This is true both for attacking and defending.

...

1. Hexes that have direct value in CVs define what we are trying to defend (or capture from the German point of view).
2. The relative strength of land forces in the theater of operations tells us whether we are attacking or defending.
3. When defending we look for other units that might come to our aid, be they our own units from other theaters or from our allies. Actually, we might look for other units to help even when on the attack.
4. We calculate the length of the frontline we need to hold, and the average number and defensive strength of units we can put in each hex.
5. We decide whether to hold every hex, every other hex, or every third hex depending on the strength we can muster per hex for each possibility.
6. We worry about overruns.
7. We choose hexes with good defensive terrain.
8. We try different placements protecting the more valuable hexes first.
9. We perform post evaluations of placements critically, looking for weaknesses that the attacker might exploit.
10. We reduce the number of valuable hexes we are trying to defend if the post evaluations find too much fault with the placements.

...

Here is how I think the AIO should analyze the position for France in 1940.

(1) France is in a Somewhat Defensive position numerically. It would be worse except that the front line is short and every hex is behind the Maginot line. France can get about 14 strength points in each hex, which are tripled behind the fortified line to 42 points per hex. The weakest part of the line is Strasbourg because it can be attacked from 4 hexes but the Germans probably can’t get more that 18 points per hex which only gives them odds of 2:1 on the assault table at the very best. That assumes that air units are contributing a lot to the attack odds. The French units might be disrupted, but still, with extra units in the rear to replace lost units and participate in counter attacks, the French defensive line looks very solid.
(2) The major risk is having to defend the borders with the neutrals to both the left and right of the Maginot line.
(3) Switzerland looks like a tough nut to crack. It only has 2 hexes on the German border, both are mountainous, and the Swiss have 6 corps with which to defend (average strength of 4.5). Even if Italy gets involved, the Swiss only have to add one hex to their defensive line. The Swiss defensive strength is 10 per hex doubled for mountain, with the mountain unit tripled. That comes out to about 22 strength points per hex. The Germans are only attacking from 3 hexes at best, for 2:1 odds. It looks very similar to the Maginot line. In a worse case scenario, Germany would get an open hexside on the end of the Maginot line which isn’t very much gain for a lot potential casualties.
(4) Belgium is vastly weaker. Their 4 corps average 3.5 strength points each and they have 4 border hexes to defend against a German invasion. Two of those hexes are clear and susceptible to being overrun. With one unit in each frontline hex, the Belgian defensive strength is only 3.5 per hex and the Germans can get excellent odds against any hex they choose. The German air force can really help out here because the French and British can’t assist Belgium during the impulse that war is declared and the German Stukas can increase the odds with a couple of shifts all by themselves. Even more distressing is that once Belgium falls, the French frontline becomes extended by 6 hexes, 5 of which are clear hexes with virtually no good defensive terrain between Belgium and Paris.
(5) Therefore, France should ignore its border with Switzerland and set up units to defend against a German declaration of war on Belgium. That means France needs to defend 10 hexes of frontline from Calais to the border with Switzerland.
(6) The British can help out here by defending Calais and maybe even Lille if they are being generous with land units in continental Europe. That leaves 8 hexes for France to defend. 1231 and 1232 are the weak points because both of them are clear terrain and the Germans can attack from two hexes. France is hard pressed to get two units in every hex with divisional units on top to take losses. Even optimistically, they can only expect to get 12 strength points per hex.
(7) Given that Germany will be able to put 18 points per hex, that gives them 3:1 odds on the Blitzkrieg table. Since there is clear terrain in almost all the hexes behind the frontline, advance after combat looks like a serious threat.
(8) As soon as Germany declares war on Belgium, France goes over to a Strongly Defensive posture. Once Germany gets within 2 hexes of Paris, France goes to Desperate Defense.
(9) If possible, France should move into Belgium because that shortens the frontline by a hex. In general, France should press their frontline right up against the Germans because that increases the number of hexes between the Germans and almost all the valuable hexes France is trying to defend.
(10) We are getting closer to where the value of hexes comes into play. Let’s start by examining the 2 die 10 land combat table. The important numbers here are expected attacker losses, expected extra attacker losses due to terrain or weather, expected defender losses (excluding shattered units), and the probabilities that the defensive hex will be left empty if it holds 1, 2, or 3 units.

Odds A losses A+ losses D losses Empty 1 Empty 2 Empty 3
1:2 Assault 1.32 .38 .31 25% 3% 1%
2:1 Assault .93 .40 .81 67% 21% 15%
5:1 Assault .42 .25 1.59 97% 72% 64%



Increasing the odds from 1:2 to 2:1 reduces the attacker losses by .39 and increases the defender losses by .50. The chances of taking out a 3 hex stack so the attacker can advance into an empty hex increases by 14%. Similar calculations can be done for other changes in the odds on the assault and blitzkrieg tables.

(11) In combination with the CV of individual units, we can now calculate the expected CV loss of a hex at various odds levels, for both the attacker and the defender. Putting in our knowledge of the terrain, we can see the effect terrain has on expected CV losses inflicted and received. We can also compare the Assault table to the Blitzkrieg table for expected losses and probability of having the defended hex vacated.
(12) All together, we now can compared the defensive value of 1132 (a clear hex) with 1131 (a woods hex half defended by a river). Eh, viola! We have a relative value for a hex in the frontline measured in terms of expected CV losses.
(13) We can examine each hex in the front line, assuming the attacker focuses maximum pressure on it, to determine expected CV losses. When we have done them all, we can then assign a value to each hex in either absolute CVs or as a ratio when compared to the best defensive hex in the line. Let’s go with the absolute value by simply subtracting the change in attacker CV losses from those of the defender. One of the CV loss values will be negative so you can think about this number as the net change to the two engaged armies. To summarize, hex 1131 is X number of CVs better than 1132. Alternatively, we could do this as a ratio and say that hex 1132 has only Y% of the defensive value of 1131. This loses the CV measure but it provides a direct measure of the relative worth of the hexes.
(14) One of the neat applications of this analysis is measuring the value of a hex adjacent to an intrinsically valuable hex. For example, going back to Poland, we can measure the improved odds Germany achieves by having an extra hex from which to attack Warsaw. That gives us the change in expected CVs and the value of the additional adjacent hex. In this discussion about France in 1940, we can apply it to evaluating the advantage of getting an extra hex from which to attack Lille or even the resource hexes 1232 and 1030. The value of hexes adjacent to valuable hexes can now be calculated in CVs.
(15) Indeed, if we have to give up one of two hexes, both of which are adjacent to a valuable hex, then we could use the ratio measure to determine which is better/easier to hold.
(16) A frontline is only as good as its weakest hex. Therefore we can use the hex CV measure to try out different placements of units to maximize the strength of the weakest hex.
(17) We can also examine different possible frontlines and perform the same assessment. This will let us decide whether to pull out of Belgium (1234 and 1233) if we lose 1232 and defend in France proper instead. The advantage is going to be that Lille is a city so the assault table has to be used. Both Lille and Calais are attackable from fewer hexes (3 to 2 and 2 to 1) than the hexes in Belgium. It should also let us measure the benefit of a shorter frontline since the average strength point per hex is likely to increase.
(18) However, pulling back means we are letting the Germans get closer to valuable hexes, Paris in particular.
(19) Time is another factor that needs to be considered when choosing a defensive line. If we can cause the attacking units to be disrupted, or at least increase the probability that they will, then that reduces the number of units the attacker can use to attack in the next impulse. Of course we should always be measuring the enemy’s ability to reorganize units with the goal of reducing that number to zero by either using ground strikes, keeping fighters available to intercept ATRs, or disrupting so many enemy units that the opponent depletes his reorganization capability.
(20) While on the subject of time, we should look ahead 2 or 3 turns to predict what the future has in store in the way of reinforcements for both sides. This is perhaps most crucial as we approach the end of a turn. Typically by then, during a major offensive, many units will be disrupted and the enemy’s ability to attack severely impaired. Most of his air units will have flown missions too, so he is unlikely to be able to put together more than one solid attack per impulse. Our evaluation of the defensive prowess of our frontline might be overly optimistic. We evaluate how feeble our opponent is with all those disrupted units and think about resting on our laurels. That is very dangerous, for if the opponent ends the turn (has the last impulse in the turn) and then gets the initiative for the next turn, we can be looking at all face up enemy units with his full complement of air units while our reinforcements have yet to arrive and our frontline irregularly formed after the opponent’s last attack. This possibility has to be considered very seriously and contingency plans made. It only takes one bad turn to transform a good defense position into a lost cause.
(21) It should be fairly easy to calculate how many attacks the enemy can make in the next impulse. There might be several numbers, such as 2 at very good odds, plus 1 at ok odds, plus 1 other at poor odds. From the defensive point of view we want to know probable CV losses for each side, the probable number of disrupted units for each side, likely defensive hexes left vacate, likely advance after combat hexes (including blitzkrieg advances), and likely retreats. We would do this for 2, 3, and 4 attacks (using the immediately previous hypothetical).
(22) Post analysis (look ahead) after pessimistic, likely, and optimistic results should be done. That lets us evaluate how bad things might be when we next get to move. Here is where we should worry about the turn ending and the opponent gaining the initiative for the next turn - and having to face the dreaded two impulses in a row.
(23) Comparative post analysis of different frontlines would let us choose between them. What we want to take into consideration is the attrition that inevitably occurs during combat. If we start out at a slight numerical disadvantage and equal casualties/disruptions are taken for a serious of impulses, the opponent usually ends up with units left to move while we have none. How many impulses remain in the turn becomes a crucial question. How many reinforcements both sides are going to receive also becomes crucial information for making good decisions.
(24) It would be nice to continue doing look ahead for several turns but that is a red herring. The probabilistic outcomes make looking ahead more than 1 impulse silly. We should rely on first principles of what constitutes a good defensive line instead. The incremental gain in information versus the cost in CPU time (and programming time) makes looking ahead more than 1 impulse not worth the effort. What we might do as a half measure is extrapolate from the previous impulse through to the expected end of turn. For example, we might have lost one hex of the frontline (Germans occupy what had previously been held by French units) and had so many casualties and disruptions on each side. If that continues for 3 more enemy impulses, then we will be back an entire hex row (closer to Paris).
(25) Looking back at the Poland example, we could expect to lose so many units/hexes during the first German impulse. A post analysis after pessimistic, likely, and optimistic results will let us predict if the second impulse will be about the same, better, or worse. Figuring out how long it will be until Warsaw falls should be doable. Applying a similar analytical process to the French frontline, the eventual fall of Paris can be crudely estimated.

_____________________________

Steve

=======================
Do the calculations also allow/defenda against throwing in amphib landings or naval bombardment in examining potential combat? Not much of a threat for AI playing Germany v. France but could be a defensive factor if the AI was the Allies.

_____________________________

/Greyshaft
=======================
The Field Marshal (FM) decision maker is aware of the possibility of support from the Admiralty and the Air Marshal and performs attack calculations both with and without their support. He then reports the end result of his calculations to the Joint Chiefs of Staff (JCS) decision maker, who directly decides which Action to take for each impulse, and indirectly decides on joint operations. If the JCS, through the choice of Action, enables the joint operations, then the FM chooses which hexes to attack and requests support from the other direct reports to the JCS (Admiralty & AM) - both of whom should be able to provide it from their available naval moves and air missions. This is a little hairy, just as it is when playing over the board. There are a lot of interdependencies and conflicting objectives between the branches of service (and amongst allies for that matter).

So, yes, the calculations do include all the odd bits of land combat that make up the rules for same in WIF.

Bt the way, the creation of individual decision makers facilitates writing the AIO. Designing it as one large monolithic logic structure would be a nightmare.

_____________________________

Steve
=======================
I would like to add here that even if anonther frontline seems better using all the calculations you exposed, the game also have to assess 2 things :
- Will his army be able to shift from the current frontline to the better one in good order ?
- Will keeping the less good, but currently occupied, frontline make you buy some time ? Sometime it may be better to hold a defensive line, even if it is not the best, just long enough and retreat when it is very threatened. I mean, it would be bad if the AIO didn't held some ground just to achive better defensive positions.

I'm not sure if you talked about this, but the AIO should also be able to foresee eventual outflankings, either because the flanks are not solidely anchored (Russia, China with the Europe scale), or because of possible invasions (northern Europe, Italy), and deal with these threat.
=======================

I would like to add here that even if anonther frontline seems better using all the calculations you exposed, the game also have to assess 2 things :
- Will his army be able to shift from the current frontline to the better one in good order ?
- Will keeping the less good, but currently occupied, frontline make you buy some time ? Sometime it may be better to hold a defensive line, even if it is not the best, just long enough and retreat when it is very threatened. I mean, it would be bad if the AIO didn't held some ground just to achive better defensive positions.

I'm not sure if you talked about this, but the AIO should also be able to foresee eventual outflankings, either because the flanks are not solidely anchored (Russia, China with the Europe scale), or because of possible invasions (northern Europe, Italy), and deal with these threat.

Yes, I agree.

Each proposed frontline can be examined by the AIO, obviously starting with the current one (don't move any units). Typically other frontlines to examine will be to the rear, but at times, the enemy might leave hexes open towards the front that need to be checked out too. It would be very embarrassing if the enemy could simply vacate a large section of the front line and the AIO wouldn't advance to recapture good defensive hexes simply because it was on the defensive strategically.

After selecting a new frontline, there are two concerns: (1) can we move there and (2) will we be in supply when we get there. The second is actually easier to calculate. Only after the best withdrawal possible is worked out can we then estimate what the withdrawal entails in terms of units left behind and hex rows lost. This is very difficult stuff to work out. Even the very best human players don't make these moves quickly - it is rare for withdrawal moves to be obvious.

Flanking maneuvers are something I have worried about when writing the AIO for other games. One of my previous AIOs was for combat in North Africa. MWIF will have a similar problem in NA with the southern flank always hanging in the air. And as you noted, there is also the potential for invasions behind the line to cut off retreats and supply. In both cases the solution is to refuse the exposed flank (i.e., to angle the flank away from the enemy). Against flanking maneuvers this forces the enemy to travel farther to get around your exposed flank. It also causes him to have to extend his line, which requires more units and that he be capable of extending his supply lines at the same time. In difficult terrain (and/or weather) the latter becomes quite important. Against invasions, refusing a flank means the invading troops have to land farther from the rest of their army, which makes linking up more difficult and provides you with more time to kill them off in isolation. Paradrops are much harder to defend against. They usually require a garrison of important hexes as well as a mobile reserve.

If we are talking about an AI value of a hex or sea zone.

What should be considered:

(1) Objectives - you must have a plan to be efficient.
(2) Supply - with out supply you will not be able to take your objectives.
(3) Resources - with out resources you will not be able to build units.
(4) Force projection - this would be for both friendly and enemy units.
(5) Weather and Terrain - this will effect the number of units required to achieve the objective.

Therefore a numeric value could be derived for any hex or sea zone at any time.

What I have been using as a universal metric for the AIO is a combat value (CV). You can think of these as a currency for measuring the worth of everything in the game. The primary building block for a CV is the combat strength of an infantry unit with a movement factor of four. Thus, a 6-4 is worth 6 CVs. Without going into detail, different movement allowances increase/decrease the CV for the unit. Different land unit types have modifiers (multipliers) for their unit type. These modifiers are situation specific, changing as the weather and terrain changes (the multiplier for a ski unit in the Artic circle versus the same unit in the jungle.) Other branches of the armed forces have their own variation on CVs (air CV = ACV and naval CV = NCV). Both ACVs and NCVs are converted to land CVs and again the conversion multipluer changes depending on situation (a convoy point for the Chinese versus one for the Commonwealth).

Resource points and factories are translated into CVs based on the average CV value of the last turn's production for the major power being analyzed. If a build point produced 4.5 CVs on average, then a resource point (or factory or production point) is worth 4.5 times the production multiple. Obviously these change as production multiples increase. Excess resource points have little or no value if there aren't factories that can use them. Etcetera. Convoys have the same value as a resource point, provided they are being used to transport resources. Oil is similar, but can be worth more if needed for reorganizing units. Rail lines are trickier because it is the entire, intact rail line that has value, not an individual rail hex. Ports, in the same sense, can be a crucial hex in a rail line.

Supply I haven't worked out in detail but my general feeling is that it can be handled as a multiplier for a group of units or even a single unit. Taking out an essential hex in a supply line is worth the reduction in strength that occurs for the units now out of supply. If you are trying to attack, then the value of cutting supply is the total CV strength of the units cut off (i.e., a lot). When the units involved are only defending, it is less.

Complete conquest of a country is the total CV value of the country's units (i.e., a lot).

The effect of multiple hexes on a single game event are troublesome. This comes up with communication/transportation/supply lines and also with taking all factory hexes to achieve complete conquest.

I think I have covered everything in your list except force projection. I'm not sure if that is separate from the other elements. The way I look at these things is to tie them back to specific game rules. Force projection, per se, isn't in the WIF FE rules.

And now we come around to the value of a hex. If the hex has a resource or factory, then that has an obvious value. If it is also the capital (e.g., Paris) then that has a value because of incomplete conquest effects. I can calculate the value of hexes adjacent to these 'objective' hexes based on the improvement in odds (the improved probability) for capturing the hex. I can calculate the value of different defensive hexes based on the difference in expected combat losses (e.g., terrain benefits). The last includes the difference in the assault table versus the blitzkrieg table.

At this point I stumble to a halt. In France I can get pretty close to evaluating every hex between the German border and the 'objective' hexes in France. Mostly because there aren't too many hexes involved. In Russia, I need a more robust approach for the analysis, otherwise the sheer volume of hexes to consider can swamp the AIO's analysis ability.

At this point I stumble to a halt. In France I can get pretty close to evaluating every hex between the German border and the 'objective' hexes in France. Mostly because there aren't too many hexes involved. In Russia, I need a more robust approach for the analysis, otherwise the sheer volume of hexes to consider can swamp the AIO's analysis ability.

Maybe you could cut Russia into several pieces for the AIO.
- Ukraine (Odessa, Kiev, Dnepropetrovsk)
- Bielorussia (Minsk, Vitebsk, Smolensk)
- The North (Leningrad, Novgorod & Pskov)
- The extreme North (Murmansk, Archangel)
- Moscow Area (Moscow, Kalinin, Tula, Ryazan, Gorki Yaroslav)
- The Factory Line (Rostov, Stalino, Kharkov, Kursk)
- The Caucasus (Tiflis, Baku)
- Urals (Perm, Sverdlosk, Kazan, Vyatka)
- ...
Maybe those area could be considered as separate countries the AIO tries to conquer / defend.
Just an idea.

That is simular to what I was thinking of:

(1) Objective - Ukraine (Odessa, Kiev, Dnepropetrovsk).
(2) Supply - The AI's supply situation what it will take to get there. Any chance to disrupt enemy supply lines.
(3) Resources - What resources are in the area that can be taken.
(4) Force projection - Forces available in the area.
(5) Weather and Terrain -What is the weather and terrain like. Terrain could be assigned a value from Zero a clear hex during clear weather to what ever would be the worst terrain and blizzard conditions.

Sea zones are much easier the number convoys trace the supply and resource lines.

Steve,

OK, I read your outline of how a hex may have a CV calculated. Based on the calculation the FM will make a decision on whether to maneuver, disrupt or attack.

Using the defense below what would the AI do?

Base assumptions,

- grey stacks are German infantry
- black stacks are German armour
- orange stacks are Russian infantry
- red stacks are Russian tanks
- all German stacks total 15
- all Russian infantry stacks are 10, tank stacks are 14
- it's Jul/Aug
- there is no weather modifier
- its the German impulse
- the next priority objective hex Germany wants to take is Moscow
- the Germans have four LND2 tac value of 5 each and two FTR2 combat value 6 each - the Russians have one LND4 tac value 3 and one FTR2 combat value 4 which cannot intercept any hexes

Russian stacks 1,2,5 and 6 have a CV of 20, 9 and 10 a value of 14 and 3,4 and 7 are CV of 10. Is that correct?

Everybody else,

What would you do as the German FM?

An interesting problem. Thank you for going to the effort to pose it.

I am making the following additional assumptions:
(1) HQ units exist somewhere in the unit mix and supply is not a problem for either side.
(2) All units are face up.
(3) There are 3 units in each hex.
(4) AT and AA units (and other special capability divisional units) are not involved.

The first task for the AIO is to determine whether to attack of defend. There are 7 German infantry stacks of 15 points each for a CV value of 105. There are 4 armor stacks of 15 points each. Since I am currently weighting armor at 3 times the CV value of infantry in this type of terrain (mixed with clear), that gives a total armor CV of 4 x 15 x 3 = 180 CVs. The Germans have 285 CVs. Similar calculations for the USSR yields 164 CVs. This ratio, given the number of units involved (33 versus 30) places the Germans strongly on the offensive.

The air units, as described, mean that the German tactical bombers do not need escorts. Therefore the German fighters shadow the USSR tactical bomber flying against it whenever it takes off on a mission. It might be feasible and desirable to rebase one of the fighters so it can do that job better. For now, I will consider the USSR bomber stymied by the German fighters and of no consequence this impulse.

The AIO has to determine the front lines. This is actually rather hard to do. If I had a pencil, I could show you easily. The answer in words is that each hex in the ZOC of a USSR unit that can be reached/occupied by a German land unit (which still stays face up) is in the USSR front line. Note that part of the frontline is already occupied by German units, part of it is empty but in a German unit ZOC, and some of it is empty of both German units and their ZOCs.

The AIO then checks for which of his units can reach the USSR frontline, which in this case is everyone. The USSR units in the USSR front line (and therefore by definition attackable) are 1, 2, 3, 4, 5, 6, and 7.

I agree that the USSR stacks 1 and 2 have a CV of 20 (doubled against both land and air attacks). The CV value for the other hexes are not modified by terrain. Instead the CV of any attacking units are halved if they are coming across a river. The distinction is important because attacking air units are unaffected by the river. And vice-a-versa for forest hexes.

Each of the 7 USSR frontline hexes occupied by the USSR are examined by the AIO for the number of attacking hexes available. Here is that analysis in table form.
Hexes is the effective number of hexes from which the AIO can attack. I am assuming that the German stack #1 can reach the forest hex NE of USSR stack #1, and still be face up and in supply. The two DRM (die roll modifiers) are with and without armor shifts. In all cases I am assuming the best units are used in the attack. This is clearly impossible for all the attacks since there isn’t enough armor or tactical air units to do them all simultaneously. [There are enough land units to do that though.] One other restriction is that some hexes are used in multiple attacks. Clearly only one stack can attack out of a hex, so attacks on #3 and #4 as decribed above is not possible (the hex between them would be used twice).

The weak points from the AIO’s view are 3, 4, 6, and 7. The attacks at +4 DRM on the assault table are far too risky with 2 of the attacking units possibly lost and the others turned face down. These attacks should only be done if a war of attrition is underway and that doesn’t appear to be the case with the total unit counts so close. The +6 attack on the Blitz table doesn’t look that good either given that it is coming from a single hex.

The attacks on the four remaining hexes assume 11 tactical air units exist, so somebody isn’t going to see any planes flying overhead. The first place to remove them is from the attack on hex #6 since the odds are +14. Without planes it goes to +10. That still leaves a request from the Field Marshal to the Air Marshal for 7 when only 4 are available. Those same 4 bombers might be wanted for ground strikes too.

Note that the AIO is not concerned about counter attacks given that the situation assessment is for a strong offense. If the USSR counterattacks, that is probably all for the best for Germany. It only becomes important when deciding about advance after combat when there might be only one German unit in a hex and/or supply lines are vulnerable.

Tactical mode choices are: (1) destroy enemy units, (2) disrupt enemy units, (3) maneuver, and (4) push the enemy back. #2 is out because there are too many face up USSR units and not enough tactical bombers. A lot of the USSR units are in terrain that protects them from air units as well. #1 is attractive because it looks feasible. #3 is also quasi-attractive because the USSR line is not contiguously occupied by USSR units (i.e., there are holes).

An analysis of maneuvering against individual hexes should reveal that 1, 3, and 7 are vulnerable to this tactic. It is possible to threaten to isolate #1 and # 7 from supply. Doing this to #7 is especially nice (unless they are cavalry) because it forces the units to choose between: (1) staying where they are and being put out of supply, (2) withdrawing into the swamp and becoming disrupted, and (3) withdrawing due east into a clear hex that is unprotected by rivers. Maneuvering against #1 also has its attractions, since it can be herded back towards the center.

The AIO should decide to attack hexes 3 and 6, and maneuver against #1 and #7. I don’t know how to make that happen just yet. At least not beyond what I have indicated here. The proximity of Moscow (the objective hex) should make the attack on #3 much more attractive than the attack on #7, though in other respects they are quite similar. The tactical air should not be used for ground strikes because its effectiveness is probabilistic, while in a ground support we know the effect it will have on the combat odds. Two of them for #3 and the other two kept in reserve. This will give the AIO 2 attacks: on #3 from three hexes at +10 DRM Assault (unless the defender chooses Blitz). These should all be infantry units. The attack on #6 will be +10 DRM Blitz (all armor units) with the prospect of advancing into the clear hex due east of USSR #7 to isolate it.

If all goes as expected, the USSR will have two fewer stacks with half the attacking German units face down. The expected German losses are .67 units in the assault with 36% face up afterwards. USSR losses are expected to be 1.59 with the hex left empty 64% of the time. The Blitz attack has expected German losses of .18 units and face up German units 69%. The USSR losses are expected to be .45 units plus 1.13 units shattered. The hex will be left vacant 90% of the time. I did these calculations in my head so they might be slightly off.

The following impulse should see German units one hex row closer to Moscow. USSR units #1 moves back. #2 holds its ground. #7 has troubles. #4 and #5 shuffle along the river towards Moscow. #8 becomes part of the front line without moving. And the USSR player might have to put an armor unit or two into the front line.

Tactically, this is a combination of modes 1 (attack to destroy) and 3 (maneuver). Only indirectly is 4 (push the enemy back) involved.

Can I now look the answer up in the back of the book?

What would you do as the German FM?

As the Field Marshall, I would screen the russians 4, 5, 6 & 7 stacks with German infantry stacks 4, 5, 10 & 11, and I would mass the Panzers to blitz Russian stack 3 on the 1st Impulse, and 8 next impuls, to open a hole in the Russian line and try to threaten Moscow before the pressure on my sides are too heavy for my Infantry Stacks. If I can threaten Moscow credibly and soon, the pressure will fall by itself with the Russian units trying to protect Moscow.

Can I now look the answer up in the back of the book?

I'm happy that your projected AIO did what my WiF player's sense made me write in my previous post (attack stack 3). I must admit that I chose that only guided by the positions on the map and my gut feeling, and I did not calculate the possibility of an attack on stack 6 (or anyone else), which is a good idea if it does not take units from an attack on stack 3.

Just one tiny thing to criticize. Your AIO exemple does not take into account the possible counter attack from 2 Russian armo stacks on German units after their possible half flip on attacking stack 3 (2 armor stacks attacking in an plain hex are +6, which is enormous all by itself). The AIO must think about this possibility and maybe reorg the most exposed units instead of the most powerful, or rebase necessary planes to support the endangered german corps, or maybe ground strike those 2 armored stacks with the best TAC to prevent them from being a too strong threat.

Anyway, thanks for buckyzoom to have proposed the example, and thanks to Steve for having answered it in a great and complete way [:D]

ORIGINAL: Shannon V. OKeets
Each of the 7 USSR frontline hexes occupied by the USSR are examined by the AIO for the number of attacking hexes available. Here is that analysis in table form.

Code: Select all

 Stack	Terr.	Def.	Hexes	Att.	Tact.	Land	+Air	Arm.	DRM	DRM	CRT
 1	Sw	20	2	30	10	3:2	2:1	-	4	4	A
 2	Sw	20	2	30	10	3:2	2:1	-	4	4	A
 3	Fo	10	3	45	10	4:1	5:1	-	10	10	A
 4	Cl	10	1	15	15	3:2	3:1	+4	6	10	B
 5	Cl	10	1/2	7.5	7.5	2:3	3:2	+2	4	6	B
 6	Cl	10	1 1/2	20	20	2:1	4:1	+6	8	14	B
 7	Fo	10	3	10	10	4:1	5:1	-	10	10	A
 

A detail: Armor bonuses on hexes 4, 5 and 6 should be halved when attacking across river, so an attack on 6 would not be quite as beneficial as the table indicates. (The difference between +11 and +14 is pretty huge). Also, using a stuka for ground strikes would most likely be more effective than using it for ground support.

Does the reduction of ARM bonus due to the driver change the AI's decision to attack RU stack 6?

Also, assume all INF stacks move 3, ARM stacks 5, everyone is in supply unless they are isolated by ZOC or hex control from their 'line'. Also, Germans have seven re-org points. Russians four.

Each stack contains one black print unit and one white print unit. (i.e. defense of 1 and 3 when OOS respectively)

I bring this up because the AI could have attempted to ground strike RU stack 7 and then isolate them for an easier attack. How would the AI resolve this?

Another option is that the AI choose to go 'AIR' ground strike RU stacks 4, 5 and 6. If the fourth LND2 could take a crack at stack 9 this should result in five or six face down RU units. The RU player would be in a position to burn re-org early in the turn or give the Germans easier attacks. Using four of their re-org to turn the LND2 face up.

If you look at the hex line formed by German units 8, 7 and 10 you will see that most of the RU forces are on or below this horizontal. Pinning the RU forces on or below the line greatly reduces the number of corp RU can move to defend Moscow. How does this planning fit in to the FM (and the probablility of the above ground strikes occuring)?

Finally, how do you handle the fact that RU stack 6 is a 'blitzable' hex? Its clear and the German FM could concentrate on this spot to crack the mini-river line. Two LND2 could ground strike with the average result being that two units are flipped (call it +3). The other two LND2 could ground support (another +2). Armor bonus is a +3 (halved for river). This is +8 before odds. After river effects the odds will be 23:10 or 2.3:1. Another +4 giving a +12 blitz. Average result is */2B.

Having nine units the German FM could occupy the two hexes directly east of RU stack 7. (Probably only need to occupy JUST the hex two hexes east of stack 7 to permanently isolate it.) It also sets up RU stack 5 for next turn. I'd take a hard look at this attack if I were the German player. From the AI's point of few though this moving away from Moscow.

How will the fact that this is statistically the longest turn of the game influence the AI's decisions? Is there an average re-org required per impulse to sustain the attack?

ORIGINAL: hakon

ORIGINAL: Shannon V. OKeets
Each of the 7 USSR frontline hexes occupied by the USSR are examined by the AIO for the number of attacking hexes available. Here is that analysis in table form.

Code: Select all

 Stack	Terr.	Def.	Hexes	Att.	Tact.	Land	+Air	Arm.	DRM	DRM	CRT
 1	Sw	20	2	30	10	3:2	2:1	-	4	4	A
 2	Sw	20	2	30	10	3:2	2:1	-	4	4	A
 3	Fo	10	3	45	10	4:1	5:1	-	10	10	A
 4	Cl	10	1	15	15	3:2	3:1	+4	6	10	B
 5	Cl	10	1/2	7.5	7.5	2:3	3:2	+2	4	6	B
 6	Cl	10	1 1/2	20	20	2:1	4:1	+6	8	14	B
 7	Fo	10	3	10	10	4:1	5:1	-	10	10	A
 

A detail: Armor bonuses on hexes 4, 5 and 6 should be halved when attacking across river, so an attack on 6 would not be quite as beneficial as the table indicates. (The difference between +11 and +14 is pretty huge). Also, using a stuka for ground strikes would most likely be more effective than using it for ground support.

Perhaps I missed something somewhere, but I do not see that the armor bonuses for attacking in clear hexes are affected when coming across a river. The 2 Die 10 Land CRT specifically mentions that they do not apply when coming across a fort, but does not mention rivers.

You are right about the stukas in the clear against three units in a hex. It's more iffy against the stack in the forest, and I prefer the guaranteed +2 DRM, over the proverbial 3 in the bush.

It is on the far bottom of the 2d10 chart. "Halve attacking bonuses when the combat factors of the attacking units are halved".

(The latest version clarifies that this does not apply to HQ support.)

ORIGINAL: Froonp

Just one tiny thing to criticize. Your AIO exemple does not take into account the possible counter attack from 2 Russian armo stacks on German units after their possible half flip on attacking stack 3 (2 armor stacks attacking in an plain hex are +6, which is enormous all by itself). The AIO must think about this possibility and maybe reorg the most exposed units instead of the most powerful, or rebase necessary planes to support the endangered german corps, or maybe ground strike those 2 armored stacks with the best TAC to prevent them from being a too strong threat.

I stand by my original position that counterattacks by the USSR are good for the German player. Germany can pretty much ignore USSR attacks. Foolishly advancing after combat such that there is a single unit in a hex or the supply lines are endangered excepted.

If the USSR gets to attack a German stack containing a facedown armor and a facedown infantry corps, they might get 3 hexes on it for 38 factors versus 15. The German flies a stuka so it is now 2:1 (roughly). The USSR has 4 armor against 1 for +3 and another +4 for disrupted Germans. This is a +11 DRM (which is a very good attack and most likely the best the USSR can hope for). To get the shifts, we are using the Blitz CRT. 10% of the time all the USSR units are all going to be disrupted and an additional 27% of the time, half of them will be disrupted. The German losses might be annoying, but 5 (or 9) successful ground strikes on the best USSR units would require opening a bottle of champagne where I come from.

After the counterattack, the USSR units would be concentrated, not behind rivers, and most likely some of them in the clear. Juicy. Who's holding the rest of the line against the other 30 German units? Only 21 USSR units remain uninvolved. My basic view is that the USSR has too much to do to defend well to commit so many units to counterattacks. The risk of disrupting a large number of his best units could very well result in a pyrrhic victory. The USSR has to suck it up and accept that counterattacks are verboten unless: (1) they put a lot of Germans out of supply, (2) can be done with super odds using only a few units, or (3) are near the end of the turn when face down USSR units aren't as painful to look at.

Thanks for the clarification.

Close inspection of the situation, makes me belive that the optimal move would be to attack force 4 with maximum strength, as well as force 3 from the remaining 2 contact points.

By ground striking 4 with 2 bombers, 2 units should be flipped. Average bonus should be +3. By using 1 bomber for ground support, odds should be 2:1 (20:10). Assuming 3 armored units from each point, net armor bonus should be +6/2 = +3. Total DRM would be 10, which would achieve a B result on 9 or higher on 2d10. With HQ support from a good HQ, the DRM should be possible to increase to +11 or +12.

A breakthrough would enable forces to reach the hex to the SE of force 3, creating zoc on all sides of that force. By attacking force 3 from 2 contact points, using the last bomber to ground strike that force (for 1 likely flip, lets say +1), that force could be attacked at a +6 to +8 blitz (avg +7). A roll of 9 or higher would eliminate the entire stack. This attack, too, could be boosted by a HQ support, if available.

A brave german would also make sure to enter the hex directly to the east of where force 4 starts. By doing so, the Soviets have a really hard time sending any defenders to stop the germans from going directly for moscow, and any such attempt would likely create easy targets of the infantry that would have to remain to the west. The downside would be that Russia would be in a position to attack that stack at reasonable odds.

ORIGINAL: buckyzoom

1. Does the reduction of ARM bonus due to the driver change the AI's decision to attack RU stack 6?

Also, assume all INF stacks move 3, ARM stacks 5, everyone is in supply unless they are isolated by ZOC or hex control from their 'line'. Also, Germans have seven re-org points. Russians four.

Each stack contains one black print unit and one white print unit. (i.e. defense of 1 and 3 when OOS respectively)

2. I bring this up because the AI could have attempted to ground strike RU stack 7 and then isolate them for an easier attack. How would the AI resolve this?

3. Another option is that the AI choose to go 'AIR' ground strike RU stacks 4, 5 and 6. If the fourth LND2 could take a crack at stack 9 this should result in five or six face down RU units. The RU player would be in a position to burn re-org early in the turn or give the Germans easier attacks. Using four of their re-org to turn the LND2 face up.

4. If you look at the hex line formed by German units 8, 7 and 10 you will see that most of the RU forces are on or below this horizontal. Pinning the RU forces on or below the line greatly reduces the number of corp RU can move to defend Moscow. How does this planning fit in to the FM (and the probablility of the above ground strikes occuring)?

5. Finally, how do you handle the fact that RU stack 6 is a 'blitzable' hex? Its clear and the German FM could concentrate on this spot to crack the mini-river line. Two LND2 could ground strike with the average result being that two units are flipped (call it +3). The other two LND2 could ground support (another +2). Armor bonus is a +3 (halved for river). This is +8 before odds. After river effects the odds will be 23:10 or 2.3:1. Another +4 giving a +12 blitz. Average result is */2B.

Having nine units the German FM could occupy the two hexes directly east of RU stack 7. (Probably only need to occupy JUST the hex two hexes east of stack 7 to permanently isolate it.) It also sets up RU stack 5 for next turn. I'd take a hard look at this attack if I were the German player. From the AI's point of few though this moving away from Moscow.

6. How will the fact that this is statistically the longest turn of the game influence the AI's decisions? Is there an average re-org required per impulse to sustain the attack?

Gee, I thought this was just a quiz. It appears to be closer to a final exam. All good questions though.

1. Yes.

The choice of tactical mode is based on the defender's weak points and what attack odds can be achieved against them. Given the poorer attack odds against #6, I expect the AIO to opt for "maneuvering against a strong defensive line that has holes". #3 and #7 can be forced to retreat simply by moving German units so as to threaten them with isolation. The British call this winkling them out, I believe. USSR #7 will have the three choices I described in my first post/analysis on this position.

Looking at the map more closely than I did before, #1 can be isolated immediately by moving a unit NE of it and another due west of #2. A unit might be sacrificed by occupying the hex NW of #1 (going face down for the Germans is almost like losing it in this situation). Though ground striking #1 then looks very attractive, I would hope that the AIO keeps the tactical air in reserve despite this temptation.

The German units should thin out to 1 corps per hex at the ends of the frontline and mass in the center. USSR #3 is either going to go back one hex (and sit in the clear where it can be attacked by two stacks), or behind the river (where it can be attacked in the clear by 1 1/2 stacks), or into the forest occupied by #8 who shuffles northward (maybe putting #1 back in supply?).

When the Germans next become the phasing player (3rd impulse), they should be one or two hexes closer to Moscow, still have all their offensive capability, and have weakened the USSR line. The last occurs because the USSR has to either give up good defensive terrain and/or lengthen their frontline.

2. A ground strike against 2 units in a hex, in swamp terrain, is going to fail 56% of the time (and hit both 6%). It seems that the stukas will eventually find something better to do in this long turn.

3. You are going for a disruption strategy here. I don't think the Germans have enough tactical air to achieve that at this time against 30 USSR units. Perhaps after forcing the USSR units to flip (moving into swamps) or after successfully attacking them this might work. Right now it looks like using all the tactical bombers to force the USSR to use all his reorganization capability. Not a trade off I like, if it ends up with almost all the USSR units still face up.

4. This is definitely the most interesting of your questions. I was aware of it as a player but not sure how the AIO can be programmed to perceive it as well. I have to think about it.

5. I prefer tables of distributions to average results when making decisions for the AIO. It is harder for people to do over the board, but trivial for the AIO. What it gives is the ability to look at the best and the worst things likely to happen (10% probability at each end of the distribution) as well as the expected results. But yes, it can be made into a good attack. The forces committed are too extreme for the benefit. Specifically, the 4 tactical bombers. There are a lot of impulses left in the turn and the tactical air will be delightful to have during those impulses. For example, do we expect to attack Moscow this turn? The flat out attack without the tactical air is too risky in my opinion. I do not want to risk burning reorganization points for the likely benefits. Maneuvering threatens USSR #6 with being attaced from 4 hexes next impulse. Does the USSR really want to remain holding the river line then? I think a lot of the benefits of attacking #6 can be achieved simply by maneuvering between USSR #6 and #7.

6. This is answered throughout the above. Instead of reorganization points per impulse, the AIO will measure the disruptions per impulse. If too many attacking units are being disrupted, then the offensive will grind to a halt. The calculation involves # of impulses remaining, # of attacking units, and the # of reorganization points available. It is updated at the start of each impulse. It this example, it drives the tactical mode towards maneuver and away from attack.

ORIGINAL: hakon
Close inspection of the situation, makes me belive that the optimal move would be to attack force 4 with maximum strength, as well as force 3 from the remaining 2 contact points.

By ground striking 4 with 2 bombers, 2 units should be flipped. Average bonus should be +3. By using 1 bomber for ground support, odds should be 2:1 (20:10). Assuming 3 armored units from each point, net armor bonus should be +6/2 = +3. Total DRM would be 10, which would achieve a B result on 9 or higher on 2d10. With HQ support from a good HQ, the DRM should be possible to increase to +11 or +12.

A breakthrough would enable forces to reach the hex to the SE of force 3, creating zoc on all sides of that force. By attacking force 3 from 2 contact points, using the last bomber to ground strike that force (for 1 likely flip, lets say +1), that force could be attacked at a +6 to +8 blitz (avg +7). A roll of 9 or higher would eliminate the entire stack. This attack, too, could be boosted by a HQ support, if available.

A brave german would also make sure to enter the hex directly to the east of where force 4 starts. By doing so, the Soviets have a really hard time sending any defenders to stop the germans from going directly for moscow, and any such attempt would likely create easy targets of the infantry that would have to remain to the west. The downside would be that Russia would be in a position to attack that stack at reasonable odds.

Well, you'll know the results of your ground strikes before moving units and announcing attacks, so that should help the decision process. If you fail to flip any units in USSR #3, that attack might have to be called off (it would have a 37% chance of failing to retreat the defenders at +6 DRM). The attack on #4 has a sweet 73% chance of getting a breakthrough.

If things go as planned (which should be roughly 70% of the time), things will look rosy for the Germans. But not hopeless for the USSR. There will be no breakthrough against USSR #3 so the USSR units #2 and #8 can relocate to defend Moscow and the USSR armor work its way around to Tula and the forest due east of Tula.

If things don't work out (30%) then all the tactical bombers are gone and most of the reorganization capability too. I don't know how the 4 corps attacking USSR #4 advance but they have 3, 4, or 5 hexes they would like to hold and two of those hexes are on the wrong side of the river. No air support to help.

The trade off is between making this combination attack versus maneuvering. I am not convinced either way. However, I would hate to lose that 73% attack on the first impulse of this long turn and have no tactical air and little reorganiztion capacity remaining.

Based on your feedback I think this is how you described your move.