Another views on Leader Check (Section 15.5.1 to 15.5.4)

[HOTEC]

We ask Gary to buy a lottery. Can we use his failure to boost the chance of our win? The chance applied to each event is mutually exclusive from each attempt and its outcome, whatsoever it is, should not be brought forward in next attempt.

The passing and failure of a leader check is an outcome of the computer generating a Random(x) value in comparison with a leader rating. The left side of the Attachment_1 shows the way to determine the formula for the chance of success (S) in the leader check. Put simply, it is (Leader Rating, R)/x (i.e. S=R/x). In the right side of the Attachment_1, it shows the relation of the chance of success in the leader check with respect to x of its variation from 10 to 100 applying to each leader rating from 1 to 9. In details, x is a sum of base value (B), the excess of the violation of command capacity (V) and the modified distance between the unit and its commanding HQ concerned (D) in the chain of command, (i.e. x=B+V+D). The B is a constant and varies with respect to the commanding level in the chain of command. They are 10, 20, 40 and 80 for the Corps, the Army, the Army Group and the High Command, respectively. As x is the denominator and, V and D would not be negative, the best scenario for success in the leader check is to make V and D both zero. Apparently, the command capacity under any circumstance must not be violated so as to V equal to zero. Regarding the last parameter in the right side of the equation, the first operation in the calculation of D is -5. If the command range of the unit and its Corps HQ is in place, D is then zero. If it is out of command range at any commanding level in the chain of command, the leader check at that commanding level deems to be failure. After deduction with positive in result, the distance between the units and the commanding HQ concerned is further modified by a command modifier applying to each commanding level in the chain of command. In a gist, D is the result of (distance-5)/4 applied to the High Command, (distance-5)/3 for the Army Group and (distance-5)/2 for the Army. The manual of old WitE explicitly describes that the end of this calculation is a rounded down operation handling the decimal in clarity. Hence, the best scenario for D in success of the leader check imposed upon the unit is to place the High Command 8 hexes [(8-5)/4=0, rounded down] from the unit, 7 hexes in range for the Army Group and 6 hexes for the Army. With the understanding of the condition that the leader check deems to be failure for the non-compliance of the command range, the worst scenario for D is the unit and its respective commanding HQs aligning themselves on a straight line. Therefore, the worst results for D are 150 hexes (90+45+15+5-5) applied to the High Command, 60 (45+15+5-5) to the Army Group and 15 (15+5-5) to the Army. The Attachment_2 shows the chance of success in the leader check under the best and worst scenario in each commanding level in the chain of command with respect to each leader rating from 1 to 9. Numerically, 0.33 and 0.01 chance of success is a huge in difference. Psychologically speaking, they are no difference - frustration of the failures, persistently. Those numbers in red shown in the Attachment_2 represent a failure mode for use during the consideration of a leader assignment to a new commanding post. In essence, I would not assign a leader with rating of 6 to an Army HQ where the chance of successes in the leader check are 0.30 and 0.22 for the best and the worst scenario, respectively, of which all are below the failure threshold of 0.33.

Each curve in the Attachment_1 shows an exponential decay in chance of success during the leader check with increasing x. A cliff drop in chance of success is expected when the leader moves from the Corps to the Army, then flattening the curve from the Army, the Army Group subsequent to the High Command. For use of Erich von Manstein as an example, he commanded the LVI Corps (Mot.) at the beginning of Barbarossa; then the 11 Army; and subsequent to the Army Group Don later renamed as the Army Group South. Assumingly, he has a rating of 9. The chance of success of the leader check upon his command at the Corps HQ, the Army HQ and then the Army Group HQ are 0.90, 0.45 and 0.23, respectively. If he is honorably appointed by the Player to the C-in-C in the OKH, the chance of success is further dropped to 0.11. It is worse than a leader in command at the Corps HQ with a poor rating of 2 which this leader has a 0.20 chance of success. Obviously, a good leader commanding at Corps HQ becomes a bad leader when it is moving up in the chain of command.

OKH controls all units which would have the same chance of success in the leader check. If a good leader instead of a bad one is appointed to OKH, it is reasonable to assume that the number of units passing the leader check increases. The quantity outweighs the quality. The second thought is how good it would be for the unit stationed in Berlin passes whereas the unit fighting in Stalingrad fails the leader check. Thus, the quality of the leader check for those units at crucial sectors matters.

Under the best scenario in accordance with the Attachment_2, a good leader with a rating of 9 and a bad leader with rating of 2 in command of OKH have the chance of success of 0.11 and 0.03, respectively. Their difference is 0.08 (0.11-0.03). As such a low chance of passing with so small in its difference, a good leader and a bad leader makes no difference at the high level in the chain of command.

The manual recommends that the commanding HQs should be moved close to the crucial sectors to boost the chance of success in the leader check. In light of this recommendation, the best and the worst scenarios shown in the Attachment_2 are the placement of the commanding HQs with respect to its unit under two extreme conditions. Using the best leader with rating of 9 as an example, the chance of success in the leader check under the best and the worst scenarios at the commanding level of the Army, the Army Group and the High Command are improved by 0.12 (0.45 - 0.33), 0.08 (0.23 - 0.15) and 0.03 (0.11 - 0.08), respectively. They are not impressive. The placement of the Army HQ, the Army Group or the High Command in the Communication Zone as shown in the Attachment_2 would have security issues for the exposure of the leader to be killed by strafing or a potential of force displacement of the commanding HQ after the breakthrough of the enemy. With such low improvement in success in trading off the security risks so as to boost the chance of success in the leader check seems to be a cost not justified consideration. Do not place the commanding HQs too close to the frontline.

In summary, a good leader should be appointed to the Corps HQ. Unless a better rating is available, the leader should stay in the post and should not be moved up in the chain of command.

A battle commences with Infantry attacks to break the enemy line. Because of the presence of the Combat Delay Movement Cost, the local CV supremacy (final odd: >=10:1) of the attacker has to be achieved to order to minimize the impact. Once broken and through which, the Panzer units in conjunction with the Infantry perform an envelopment for the remaining enemy units in contact and conduct a deep penetration to complete an outer encirclement. At the end of the drive, these Panzer units should be low in fuel, low in CPP, high fatigues with a lot of damages in hand resulting of very low CV values. As such strength ebbing, they are vulnerable if and only if they are attacked by the enemy otherwise they are safe. The forward zone for the Panzer units is designed to consume the MPs of the enemy units, increase of their fatigue/damages for the purposes to obliterate their potentiality to attacks so as to provide a degree of protection of the Panzer units. The air interdiction to be assigned in this forward zone adds additional MPs imposed upon the enemy units. Some measures have to be in place against the Soviet's rescue missions, in particular, at the flank-base of the penetrated area. In the following turn, the Infantry would mop up and destroy the enemy units by isolation and then move through the friendly hexes towards the Panzer units for consolidation. The Attachment_3 shows these maneuvers conceptually and the Attachment_4 shows one of the many ways to achieve a breakthrough of the enemy line.

As shown in these Attachments_3 and _4, the mobility of the Panzer units has to be maximized so that they can be able to overcome the impact of the Combat Delay Movement Cost, move from EZOC to EZOC behind the enemy line and drive deep in the enemy territory. During offensive operations, the mobility of the Panzer units is the most need than their firepower. When they are in defensive modes, the priority of concerns is in reverse that the firepower is more essential than the mobility. As for the Queen of Battle, the Infantry units need to punch a hole in the enemy line, mop up the pockets once encircled during offensive, to resist and counterattack during defensive. As they shoulder the most of fierce fighting both in offensive and defensive, the firepower of the Infantry units is, therefore, the focus rather than their mobility. Among the leaderships provided in WitE2, the Initiative and Administration are rated as the Mobility whereas the Mechanized, Infantry and Morale as Firepower. The Initiative is the predominant part in the Mobility because the Administration check passes automatically if the unit has not moved in the previous turn during the determination of movement point allowance. For those non-essential HQs (the Army HQs, the Army Group HQs and the High Command), the leader with high Morale rating should be appointed as it is used for rallying routed units. For those air-unit type HQs, the leader with high Air rating should be assigned.

The leader appointed to the Corps HQ may not have a higher Morale rating than the leader in the Army HQ. Under an appropriate condition, a routed unit should detach from the Corps HQ, attach to and move within 5 hexes from its associated Army HQ so as to boost the chance of success in the leader check for its recovery. If there is no violation of command capacity in any commanding level, the command capacity of the Army HQ should not be violated as this re-organization is done within the chain of command. Thus, V is zero. Since it is within 5 hexes from the Army HQ, D is zero as well. As for the base (B) for the Army level is now 12 not 20, the chance of success in the leader check for the unit directly attached to the Army HQ is improved. The Attachment_5 shows the condition of which the re-organization should be done otherwise no reassignment is needed.

The talents of a leader who could provide the most needs in the battles is a good leader. The criteria in selection good leaders among others to be appointed to a HQ are set as follows:

Panzer Corps HQ for offensive: Initiative, Administration, Mechanized, Morale.
Panzer Corps HQ for defensive: Mechanized, Morale, Initiative, Administration.
Corps HQ (non-Panzer type): Infantry, Morale, Initiative, Administration.
Army HQ, Army Group HQ, High Command: Morale.
Air HQ: Air.

A typical assignment is exemplified in the Attachment_6.

WitE 3.0:
The Theater of War is an area which might be the scene of operations. Within this area, the Germans divided it into the Theater of Operations and the Zone of Interior with an intermediate area known as Zone of Military Administration or Occupied Territory. The Theater of Operations itself is further partitioned into the Combat Zone and Communication Zone. The Corps takes up the Combat Zone. Whereas the Army and the Army Group occupies the forwards and rearward part of the Communication Zone, respectively. The Zone of Interior/Occupied Territory is controlled up by OKH in WitE2. The Attachment_7 shows the command range (Section 21.11.4) applied beautifully in the Theatre of War. However, this beauty is distorted by the leader check. In order to boost the chance of success in the leader check, the commanding HQs have to move to the Communication Zone close to the units at the crucial sector(s). Albeit the chance of success is not impressive, the Soviets may have little option as they have both low morale and poor leadership rating. An incentive has to be provided so as to make the HQs stayed inside their respective zone of operations.

To move a good leader up the chain of command is a waste of its talents. Thus, the good leader would stay in Corps HQ all the time. If its promotion is not interfered, the leader would dishearten by the development that a FM or GENO commands the Corps. Hence, I selected 84 leaders based on the aforementioned criteria and invited them to attend the Editor Conference. In the Conference, I explained to them the Gary's rules applied to the leaders. For the sake of Wehrmacht, their promotion schedules were postponed after September 1945. When the time reached, they would be promoted 2 ranks higher than the one presently held. They were gracefully accepted. As no such a boss in the world would make such commitment to the staff, their Political rating were increased to 9.

By considering the formulas below, they might provide some incentives for promotion a good leader from a low commanding level to a high commanding level and, meanwhile, keeping the respective HQs units within their zone of operations.

Chance of success of the leader check = Rating / x;.
x = 10 + V for Corps;
x = 10 + (5+15 - D)/2 + V for Army;
x = 10 + (5+15+45 - D)/3 + V for Army Group; and
x = 10 + (5+15+45+90 - D)/4 + V for High Command.
where:
x is the seed of Random (x) and x > 0;
D is the distance between the HQ unit and the unit in concern;
V is the violation of the command capacity;
(sum of command range - D) >= 0. It is 0 if negative which is out of command range (i.e. no leader rating check).

If the nest sorting of leaderships for leader were made available in the Commander Report, it would be a great assistance for the Player to select the desirable leader.

[HOTEC]

Attachment_1:

[HOTEC]

Attachment_2:

[HOTEC]

Attachment_3:

[HOTEC]

Attachment_4:

[HOTEC]

Attachment_5:

[HOTEC]

Attachment_6:

[HOTEC]

Attachment_7:

[jlbhung]

[&o]Wow, detailed and clear analysis. Attachment 3 and 4 are really good in showing the breakthrough process.

[loki100]

its not easy to work out if you are analysing the current system or proposing your own preferred version.

But, if I have understood you, this is wrong:

Therefore, the worst results for D are 150 hexes (90+45+15+5-5)

there is no HQ with a command range of 150. The max possible range is 90 for something like OKH or Stavka - see section 21.11.4 and 4.7.4 of the manual

as with your other post, its hard to work out, but I think you are misunderstanding when the higher parts of a HQ chain come into play (I could be wrong but its not easy to follow). The important part of the probability calculation is not the chance that leader succeeds but a combination of how many tests come to that leader and their relative competence (in effect its Bayesian not conventional statistics as this base rate is important).

This is why we used the tables and worked examples in the manual - and I fully realise there are other ways to have tried to explain what is going on but they all seemed to be harder to follow and less comprehensive (never a good combination).

[HOTEC]

You are right that there is no HQ with a command range of 150. The maximum command range is 90 which is the High Command's command range. The 150 is the sum of all command ranges in the chain of command. My understanding of the chain of command is that an order is passed from OKH to the Army Group HQ then to the Army HQ subsequent to the Corps and at last to the unit. The command range of 90 hexes is the range from OKH to the Army Group HQ, 45 hexes from the Army Group HQ to the Army HQ and so on. When I move the Army Group HQ, I see the number of distance in parentheses changing relatively to OKH. All the HQs including the unit concerned aligning themselves on a straight line with the command range of HQs apart (i.e. 90+45+15+5, -5 for calculation needed) would be the longest distance between OKH and the unit without violation the command range of respective HQs. I consider as the worst scenario. No matter you right and I am wrong (i.e. the command range 90 is the range between OKH and the unit), the chance of success in the leader check is still too low (my failure threshold of 0.33).

Bayesian theorem is the probability expressed a degree of belief in an event. My emphasis is that a good leader has to be appointed to the Corps HQs. To pick a good leader is based on the needs of the battles. Good and bad are a relative term. I use the Random(x)<rating as a tool for comparison among the conditions. As the example shown in the manual have the leader rating of 6 in all commanding levels, if you use rating of 9, 7, 5, and 3 for the Corps, the Army, the Army Group and the High Command and then the order in reverse, you may share my view when their outcomes are compared.

[loki100]

ORIGINAL: HOTEC

You are right that there is no HQ with a command range of 150. The maximum command range is 90 which is the High Command's command range. The 150 is the sum of all command ranges in the chain of command. My understanding of the chain of command is that an order is passed from OKH to the Army Group HQ then to the Army HQ subsequent to the Corps and at last to the unit. The command range of 90 hexes is the range from OKH to the Army Group HQ, 45 hexes from the Army Group HQ to the Army HQ and so on. When I move the Army Group HQ, I see the number of distance in parentheses changing relatively to OKH. All the HQs including the unit concerned aligning themselves on a straight line with the command range of HQs apart (i.e. 90+45+15+5, -5 for calculation needed) would be the longest distance between OKH and the unit without violation the command range of respective HQs. I consider as the worst scenario. No matter you right and I am wrong (i.e. the command range 90 is the range between OKH and the unit), the chance of success in the leader check is still too low (my failure threshold of 0.33).

Bayesian theorem is the probability expressed a degree of belief in an event. My emphasis is that a good leader has to be appointed to the Corps HQs. To pick a good leader is based on the needs of the battles. Good and bad are a relative term. I use the Random(x)<rating as a tool for comparison among the conditions. As the example shown in the manual have the leader rating of 6 in all commanding levels, if you use rating of 9, 7, 5, and 3 for the Corps, the Army, the Army Group and the High Command and then the order in reverse, you may share my view when their outcomes are compared.

then your understanding is (unfortunately) wrong, the HQ command range doesn't chain that way (don't think it does in #1 either). So OKH's command range (max) is 90 to the unit drawing on its leadership values (that unit can be a unit or a HQ depending on the test).

While I don't think this is the place, you are wrong re Bayes (the key is the application of base rate data to the basic probability) - but I suggest that is a discussion for a totally different place [;)]

[HOTEC]

Indeed.

The command range of 90 hexes from OKH to the unit raises a concern for the units fighting at Leningrad and beyond, and those fighting at Baku as such so apart that one of them may have out of command range of OKH. Although, the placement of OKH at Kiev seems to be the center of the battlefields for Leningrad and Baku, beyond the NE and SE edges of the huge hexagon with its radius of 90 hexes from OKH would be out of the command range. It seems that any battles with proper leader check in chain of command have to be carried out within this 90-hex-radius hexagon covered by the OKH.

Just make sure the release on March 25. Cannot wait to see the outcomes of WitE2 under computer vs computer. The location of OKH is interesting.

[Ynglaur_slith]

If there were only a single Corps, then certainly place the best leaders as far down the chain as possible. If you're trying to optimize for the maximum number of successes at a given level, though, it should be worth putting more senior leaders in more senior leadership roles.

[Rexzapper]

I love your graphics, HOTEC, even though they were made before the game was released and are more related to Wite1. It would be nice to have all those stat charts of the leaders also with data relative to Wite2, especially the comparisons of the other thread on the quality of the leaders by countries, I especially like that one. The problem is that the air leaders should probably have been removed from the ground statistics and vice versa, if not it blurs the results a bit since most of them cannot be used in that other function (or it is not worth doing). Regards