To play any game well, one must understand the tables that govern it. The CWB is more complex than most games because there are several tables that one must consult to resolve a combat. I have done a mathematical analysis of these tables and the results follow. You may lower your losses by 5-10% while increasing those of your opponent by 5-10% using the information from these tables with several tactics described below.

Allocating Fire Points or Whom Do I Clobber?

Have you ever had two lines blasting away at each other and wondered what the most effective allocation of fire power is? Given 14 fire points, do you make two 7 point attacks or two 5 point and one 4 point attacks?

[Ed. Note: Remember, you must split fires along Fire Level lines. I'm not sure our author is following that rule here. Future editions of the CWB series rules may preclude fire splitting entirely.]

The four tables below can help us answer these questions.

Assume the targets all have B morale. The expected losses from two shots on the 7-8 column are 2.76 hits (Table one, 1.38 x 2) and 2.2 stragglers (Table Two, 1.1 x 2), for a total of 4.94 casualties (Table Three, 2.47 x 2). The expected losses from two 5-6 shots and one 3 point shot are 3.19 hits (1.17 x 2 +.85) and 2.63 stragglers (.96 x 2 + .71), for a total of 5.82 casualties. This is an enormous difference!

From these tables we can see that you should try to maximize the number of shots on the 3-4 column, with shots on the 2 and 5-6 columns being very good also. It is generally wasteful to expend nine points on one shot. Spread out your artillery. Make many five and seven point shots rather than nine and four point shots.

Of course, there are exceptions. It may be worth concentrating firepower on a unit near breaking or with low morale in the hope of wrecking or disorganizing it. Also, broken or disrupted units may yield so many extra stragglers as to render the unit useless for the rest of the day.

Reverse Slope Tactics or Denying the Enemy Return Fire

Instead of placing your units on hills and ridges where the enemy can fire on them from a distance, there are times when it is better to set your line on lower terrain. Try to set your line such that if the attacking enemy units must retreat by the morale table, they cannot return fire. By denying return fire, you can lower your casualties substantially.

Assume two units with A fire levels and B morale are shooting at each other, There is a 19% chance the enemy will suffer a retreat result from the morale table (Table Four). The effects of this retreat on an A level return fire are (43%) Shaken-fires on 2 column .51 hits + .39 stragglers = .9 (43 %) Disorg-fires on 1 column .22 hits + .2 stragglers = .42 (14%) Routed-No fire possible

Expected Loss = .31 hits +.28 stragglers = .59 casualties
Expected Loss from the 3-4 table = .85H +71S = 1.56 casualties

Taking the retreat into account gives 81 %*(.85H +71S)+ 19%*(.31H +.28S) = .75H +.63S = 1.38 casualties.

Thus, when the enemy returns fire, it will inflict somewhat less damage (11.5%) than your unit caused. If you situate your unit so that a retreating enemy unit cannot fire on it, the expected loss to you drops to 81%*(.85H + .71S) = .69 hits + .58 stragglers = 1.26 casualties.

This is a 9% further reduction in your casualties and it costs nothing to implement. You never know when having a strength point will make the difference in fire level or becoming wrecked. If the enemy unit has a morale lower than B, your losses will drop more, because the enemy is more likely to retreat when fired on. An attacking C morale unit will inflict 11 % fewer casualties.

Obviously, you do not want to surrender key elevated points from which the enemy artillery can enfilade your line. This tactic works best when the enemy attacks you across a gently declining or inclining field without slope hexsides to benefit him. Defending in or behind a forest will also yield this benefit.

Distribution of Artillery Batteries or Hold That Line

What is the most effective use of artillery? Is it better to mass four five-point batteries in a row, or spread them in alternate hexes, or break then into detachments? It depends on the circumstances. If the enemy is about to charge into Sharpsburg, move all the cannons you can get your hands on in front of him. If you are short on artillery you can try placing a battery in alternating hexes. You can more effectively distribute your fire strength into seven and five point shots if the hexes without artillery are strong enough to withstand Close Combat.

However, there is a more effective deployment. If there is a line you want to hold, and you do not have much artillery, place a detachment of one to three gun points in each hex. (We play keeping track of fractions of artillery points fired with an extra marker.) This works best when the enemy cannot bring artillery to snipe off your cannons.

There are two benefits to using this tactic. First, you have a more effective distribution of firepower. You will have more shots on the 5-6 column. Second, because your infantry is with artillery, you will have to retreat much less frequently. This will maintain the integrity of your line, and lessen the need to limber cannons in enemy ZOC. If forced to withdraw a one gun point detachment, you can only lose one gun point. A five point battery can lose three points. This will save you gun points in the long run.

Because you retreat less often you will inflict greater casualties. The enemy will take more losses when you add the additional losses inflicted by your superior firepower. You will be able to hold your line longer, take fewer casualties, and lose fewer gun points.

Defending in Front of Roads

By denying the enemy use of roads running parallel and adjacent to the line of battle, you can limit his flexibility in directing his attack. In forest, try to defend in front of or along a road. This will greatly expand your command control, while limiting that of the enemy. You can also move units around more easily. If you are forced out of the road, retreat two hexes to deny him the road benefit to his command control.

Optional Rules

I strongly support The Gamers philosophy of not losing the forest for the trees. Adding many special case rules (Me Wilderness catching fire) adds very little to the realism of the game. Many of the suggested optional rules in Operations (unit density, artillery straggler recovery, etc.) add little to the simulation, except complexity and book keeping. We use all the optional rules in the Revised 2nd. Edition rules, with three additions.

First, we keep track of fractional artillery points. This is very easy. Use a spare marker on the artillery ammo track. You need this only if you use many artillery detachments.

Second, as mentioned in Operations #8 we use the night sleep rule. After dark, a unit must spend six consecutive turns stationary, not firing or being fired on. If you do not meet this condition, reduce the unit's morale by one for lack of sleep.

Third, we use a modification of the optional rule that units in column require one hex per B fire level. We play that a road is in use for the period of time a unit in column is moving along it. If two columns are moving in opposite directions across a bridge there will be congestion and only one column will be able to use the bridge. The other column will have to expend movement points waiting for the first column to clear the bridge. We keep track of traffic jams with extra numbered markers. We mark them from 1/2 to 6 by 1/2 increments. These represent the period of time a bridge or road is being used, and other units may not use it. The other units must wait (expend mps) until the road is clear. An AA unit takes four hexes of space. If it were two movement points from abridge, we would place a 2 marker and a 3 1/2 marker in the bridge hex. Thus, there is a window in which any unit reaching the bridge by expending 2 to 3 l/2 mps must wait for the first unit to pass. If a unit reached the bridge on its 2nd mp, it would have to expend 1 1/2 mps waiting for the first unit to go by.

Boring, tedious, and inconvenient, you say? Well, it takes a little getting used to, but it is not too bad. It is not a problem for a column of reinforcements, because the brigades are far enough apart that they do not interfere with each other. The only time this is a problem is when many units are crowded around a bridge, or two columns are moving perpendicularly across each other. A major delay will occur when a fresh corps on one side of a bridge wants to relieve a spent corps on the other side. You will be able to move units in only one direction, forcing the spent corps to wait a few turns for the fresh corps to clear the bridge. This can have a large impact on the game, and it should.

Many a plan was thwarted because some column was behind schedule, snarled in a traffic jam. Longstreet, at Seven Pines, did more than anyone to bungle Joe Johnston's attack by using roads earmarked for others. Throughout the Seven Days, both sides had major traffic problems. This rule emphasized good road networks, and planning troop shuffling movements carefully.

Limited Intelligence

A simple way to implement a degree of reduced intelligence is by the use of six to twelve blank counters to keep the identity of your units hidden. You may also place a hidden marker over a few bogus fire level markers, fooling the enemy into believing you have a strong hold on an empty hex. You must reveal an empty hex and remove the hidden marker if the hex is in enemy LOS. Now an opponent will not know if a strong or a weak division holds an area. Is that mass of troops behind the enemy's lines a division or a bluff. This adds little complexity, and makes for an interesting game.

How I Constructed The Tables

I used a spread sheet to calculate the probability of each result for each column of every table. In Table One, the second column has an expected loss of .22. 1 calculated this as follows: 1 hit x the probability of rolling a 10 (3/36) or 11 (2/36) or 12 (1/36) = 6/ 36 *.5 =.166 losses, and.5 hits x the probability of a 9 (4.36) =.055, a total expected loss of .22.

Table Two was more difficult to calculate. For each column on the fire table, there is a different probability of going to each straggler table. For each morale, I had to calculate the expected stragglers generated by each straggler table, and weigh them by the probability of using that table.

Table One: Expected Losses
Result	under 1	1	2	3-4	5-6	7-8	9-11	12-14	15-17	18-20	21+
Expected Loss	0.08	0.22	0.51	0.85	1.17	1.38	1.54	1.72	1.94	2.18	2.44
Increase Over Prev Col	-	0.14	0.29	0.33	0.32	0.21	0.17	0.18	0.22	0.24	0.26
Effect per Fire Point	0.17	0.22	0.26	0.28	0.23	0.20	0.17	0.14	0.13	0.12	0.12

Table Two: Expected Straggler Losses
Result	under 1	1	2	3-4	5-6	7-8	9-11	12-14	15-17	18-20	21+
A (.03)	0.04	0.06	0.13	0.24	0.33	0.39	0.44	0.47	0.50	0.52	0.53
B (.11)	0.12	0.19	0.39	0.71	0.96	1.10	1.21	1.31	1.38	1.42	1.44
C (.14)	0.16	0.25	0.51	0.92	1.22	1.39	1.53	1.64	1.72	1.78	1.81
D (.22)	0.27	0.44	0.86	1.49	1.88	2.07	2.23	2.36	2.46	2.52	2.56
E (.33)	0.36	0.58	1.13	1.88	2.25	2.42	2.56	2.67	2.75	2.81	2.83
Note: This table includes the expected straggler loss from the Morale Table. The number in parenthesis is the average straggler loss from the Morale Table.

Table Three: Total Expected Losses from all Causes
Result	under 1	1	2	3-4	5-6	7-8	9-11	12-14	15-17	18-20	21+
A	0.12	0.28	0.64	1.08	1.50	1.76	1.98	2.19	2.44	2.70	2.97
B	0.20	0.42	0.90	1.56	2.13	2.47	2.75	3.03	3.32	3.60	3.89
C	0.24	0.48	1.02	1.76	2.39	2.76	3.07	3.36	3.67	3.96	4.25
D	0.35	0.66	1.38	2.33	3.04	3.44	3.77	4.08	4.40	4.70	5.00
E	0.44	0.81	1.64	2.72	3.42	3.79	4.10	4.39	4.69	4.99	5.28

Table Four: Morale Table Results
Morale	BL	NE	SH	SHb1	DG	Rout	% Retreat	Straggle Loss
A+	17	61	11	6	6	-	11	0.03
A	14	61	14	6	6	-	11	0.03
B+	11	64	14	6	6	-	11	0.03
B	8	58	14	8	8	3	19	0.11
C+	6	58	17	8	8	3	19	0.11
C	6	50	19	11	11	3	25	0.14
D+	3	39	22	17	14	6	36	0.22
D	3	31	28	19	14	6	39	0.22
E+	3	28	28	19	17	6	42	0.22
E	-	19	31	22	19	8	50	0.33
F+	-	11	33	25	22	8	56	0.33
F	-	8	33	25	25	8	58	0.36
G+	-	-	36	25	28	11	64	0.44
G	-	-	28	28	31	14	72	0.56
H+	-	-	22	28	33	17	78	0.67

Table Five: Acceptance Table Results
Results	under -2	-2 to -1	0 to 1	2 to 3	4 to 5	6 to 7	8+
Dt	25	17	11	8	3	3	3
D2	61	53	47	36	22	14	14
D1	14	31	42	42	53	44	39
A	0	0	0	14	22	39	44

% Probability of Acceptance
Turn	1	2	3	4	5	6	7
D1	33	56	70	80	87	91	94
D2	17	31	42	52	60	67	72