Military FAQ Part 1 - Casualties
Military FAQ Part 1 - Casualties
Introduction
This post is concerned with the mechanism whereby units take casualties and how to understand the likely casualties your units will cause or take. The FAQ assumes that you have read the relevant section of the manual which gives the raw statement about how combat work but this will leave most players with little appreciation of what this means in practice.
Combat consists of a series of combat resolutions extending over a significant period in days. There is a single combat resolution for each 5-day period, which is then applied in each successive day. This will then give similar but not identical results on each of these days. The results vary because the strength of the two sides will change so typically casualties slowly decrease but there may be more subtle effects when regiments fall out of line.
The combat types are currently 5 days shock action followed by 5 days fire action repeated in turn until one side runs out of morale. This has some important implications for battle analysis. In particular
- The number of fire days in a complete battle will almost always be less then the number of shock days and will certainly never be more
- Combat against defeated and retreating enemies will always be shock days only as they are pushed into new retreats before the first fire day occurs
How it works
Be cautious in reading and understanding this section since inflicting casualties is not the only issue in battles and you need to be conscious of actually winning the battle as well, which is covered by the morale section and results in some significantly different conclusions.
Actual combat is resolved using the following formulae. The process is the same for fire days as it is for shock days, simply using fire factors instead of shock factors, so I will provide just the one general illustration.
On each day the game evaluates how many casualties each side inflicts on the other by the following…
Attack+Dice+Terrain+Leader-Defence = score
Where
Attack = the attack rating of the friendly unit
Dice = that periods dice roll (in the range 0-9)
Terrain = any terrain adjustments, these are applied to one side adjusting both attack and defence (see below)
Leader = a positive or negative adjustment reflecting the difference in capability between the friendly leader and the enemy leader (applied to both attack and defence)
Defence = the enemy target unit’s defence score.
One of the most important issues to understand in this formula is which parts affect both attack and defence and which only effect one of them. This is because anything affecting both can be considered as twice as significant.
Terrain, for example, affects both and therefore a –1 for terrain is worth 2 points and should be considered like a –2 on the dice rather than –1. (In actual fact it matches –1 on your dice roll PLUS +1 on the enemy dice roll). In 1.2.1 cavalry receives a doubled penalty for terrain which makes river crossing and woods/mountains much more of a combat hazard for cavalry than they are for infantry.
Attack factors, Defence factors and the Dice roll only have a singular effect and therefore they only count for one point per point. Leader values and terrain effects are applied to both attack and defence and therefore their factors apply twice and should be evaluated as worth 2 points each. (Note that versus cavalry each terrain point can be considered to be worth 4 points). This makes high grade leaders
extremely powerful.
Terrain modifiers
-1 for forest, marsh or hills
-2 for mountains
-1 for river crossing
Terrain issues are adequately explained in the game manual
Having calculated a score this is then looked up on the following tables to obtain the number of casualties inflicted on the enemy. The process is…
1) Look up the casualties per 10,000 on the casualty table
2) Multiply by the relative strength of the attacking unit (compared with 10,000)
3) Multiply by the tech level modifier for unit class and combat type (see below) (seen in game on the ledger page
before the first army page)
4) Subtract casualties
The manual states that the attacker goes first but casualties appear to be calculated based on start of day force strengths.
Note: I have used casualties per 10,000 so that the results are in whole numbers rather than fractions.
This table is at least very close to correct but I cannot be certain of exactness. The formula is
(score-1) x 3.5 for scores of 2 and above (the figure 3.5 proven correct within an accuracy of better than 2%)
3.5 for a score of 1
2^(score-1) x 3.5 for scores below 1 (this amounts to halving the casualties for each point below one)
The figures in this table can be looked up in the game files but are not visible in game except for checking your current level in the ledger (page immediately preceding the first army page). The table is incomplete and only covers tech levels up to 35 (year 1735). Actual tech levels in the game extend to level 59 BUT this is dated 1825 and hence not typically reachable. The change from the patch is purely the leading values for cavalry shock, which no longer start at 4 but slowly work up to it.
See unit analysis section for more information and analysis.
Worked examples
This is all a bit complex for most players so here are a few worked examples simply designed to illustrate the workings of these formulae. This are not intended to provide insight into particular combat situations.
1.
1 Latin Knight fighting 1 Latin Medieval Infantry, both at full strength at tech level 0
2 examples with different dice rolls
Knights: 1 attack + 5 dice = 6 => 5 x 3.5 x (Tech modifier) 1 = 17.5
Infantry: 0 attack + 5 dice = 5 => 4 x 3.5 x 0.5 = 7
Knights: 1 attack + 0 dice = 1 => 1 x 3.5 x (Tech modifier) 1 = 3.5
Infantry: 0 attack + 9 dice = 9 => 8 x 3.5 x 0.5 = 14
Best possible result now gives the infantry a chance whilst cavalry, on average, still retain a significant average advantage.
This nicely illustrates the situation at low tech-levels. The tech level modifier gives the cavalry a 2:1 advantage in causing casualties and this easily overwhelms all other factors. This remains the dominant factor in land combat until infantry fire factors start to become significant and the relationship changes. Bear in mind that although this shows cavalry as superior for a large part of the game it does not make them more cost effective.
1 Caroline infantry (5/4) fighting 1 Grenzer infantry (2/6), shock action
Caroline: 5 attack + 6 dice – 6 defence = 5 => 4 x 3.5 x 1.25 (Tech) = 17.5
Grenzer: 2 attack + 6 dice – 4 defence = 4 => 3 x 3.5 x 1.25 (Tech) = 13.125
Difference is quite limited and easily overwhelmed by the dice roll.
Now trying both of the above against 1 Latin Knight
Caroline: 5 attack + 7 dice – 0 defence = 12 => 11 x 3.5 x 1.25 (Tech) = 48.125
Knights: 1 attack + 7 dice – 4 defence = 4 => 3 x 3.5 x 4 = 42
Grenzer: 2 attack + 7 dice – 0 defence = 9 => 8 x 3.5 x 1.25 (Tech) = 35
Knights: 1 attack + 7 dice – 6 defence = 2 => 1 x 3.5 x 4 = 14
(dice roll of 7 chosen to keep on the linear part of the casualty tree)
The purpose of this example was to show that defence factors become more prominent when infantry fights cavalry than when they fight each other. The conclusion being that Caroline infantry is better than Grenzer when they face each other but Grenzer does better against enemy cavalry. See unit evaluation section for more details
Repeating the above with lower dice rolls allows us to look at calculations below the linear section of the table. Using 1 Knight versus 1 Grenzer
Knights: 1 attack + 9 dice – 4 defence = 6 => 5 x 3.5 x 4 = 70
Knights: 1 attack + 8 dice – 4 defence = 5 => 4 x 3.5 x 4 = 56
Knights: 1 attack + 7 dice – 4 defence = 4 => 3 x 3.5 x 4 = 42
Knights: 1 attack + 6dice – 4 defence = 3 => 2 x 3.5 x 4 = 28
Knights: 1 attack + 5 dice – 4 defence = 2 => 1 x 3.5 x 4 = 14
Knights: 1 attack + 4 dice – 4 defence = 1 => 1 x 3.5 x 4 = 14
Knights: 1 attack + 3 dice – 4 defence = 0 => 0.5 x 3.5 x 4 = 7
Knights: 1 attack + 2 dice – 4 defence = -1 => 0.25 x 3.5 x 4 = 3.5
Knights: 1 attack + 1 dice – 4 defence = -2 => 0.125 x 3.5 x 4 = 1.75
Knights: 1 attack + 0 dice – 4 defence = -3 => 0.0625 x 3.5 x 4 = 0.875
Grenzer: 2 attack + 9 dice – 0 defence = 11 => 10 x 3.5 x 1.25 (Tech) = 43.75
Grenzer: 2 attack + 7 dice – 0 defence = 9 => 8 x 3.5 x 1.25 (Tech) = 35
Grenzer: 2 attack + 3dice – 0 defence = 5 => 4 x 3.5 x 1.25 (Tech) = 17.5
Grenzer: 2 attack + 0dice – 0 defence = 2 => 1 x 3.5 x 1.25 (Tech) = 4.375
Notice how the more linear and closer spread of the Grenzer means that this combat is very strongly dependent on the cavalry dice role. An interesting side issue is that this combat can be dramatically affected by leadership. Give the infantry a level 6 leader against none for the cavalry and they are completely protected from the higher combat results. The cavalry will vary from ineffective to very moderate whilst the infantry will always achieve reasonable results.
Do it the other way round and the cavalry are shifted to spectacular results as against modest for the infantry.
This illustrates nicely that leadership is more significant for an effective tech level disadvantage in unit scores. It is
not an illustration that leaders are more significant for cavalry.
Test worked example
This example has been left in the FAQ but is based on testing patch 1.1 behaviour. It is no longer an accurate model for the actual parameters in the game but remains an accurate analysis of how the game behaves.
This worked example based on the first few days of the Battle of Alexandria is intended to demonstrate the accuracy of the casualty rules worked out so far and perhaps provide some insight into other factors. The battle is between a French cavalry army, tech level 25 Galoop cavalry, under an excellent general fighting against a Kara Koyonlu mixed army of tech level 10.
After 1 day of battle
After 2 days of battle
By inspecting individual units I can tell that the overlapping Qara Koyonlu end units have not been attacked and all units have attacked the enemy unit opposite to them or nearest at the ends except both the right hand French cavalry have been attacked by a Kara cavalry. The red lines on the display illustrate the apparent Kara lines of attack. This gives no immediate insight into the battle AI’s operation but does suggest that in normal battles nothing very weird is going on.
Qara tech level 10, French tech level 25. Oddly enough these have no impact on shock modifiers. Infantry have 1.0 and cavalry have 4.0 for both sides. Qara units are cavalry 2/2/2/0 [morale a/d shock a/d] and infantry 3/2/1/0 whilst the French have 4/2/4/2.
The dice roll appears to have compensated for the leader but the two are not actually equivalent. Our leadership increases the casualties that we inflict and decreases the casualties that the enemy inflicts whereas the enemy’s superior dice roll only improves the casualties that they inflict.
Working out the battle we get…
Casualties
French Cav versus Kara Any = 4+3+2-0 = 9 -> 8x3.5x4 -> 112
For the whole force 112 x 7 x 6,845/7,000 = 767 casualties when, in fact, 762 were reported. This is assumed to be the result of rounding errors with casualties being evaluated separately for each unit.
Kara Inf versus French Cav = 1+5-(2+2) = 2 -> 3.5
In fact, in the middle of the line the French cavalry took 3 casualties each on the first day and 2 on the second day. This clearly shows a casualty rate of 2.5 rather than 3.5 so there is a slight discrepancy here indicating that the FAQ is not entirely correct.
Kara Cav versus French Cav = 2+5-(2+2) = 3 -> 2x3.5x4 -> 28
At the left flank this should be 80.8 casualties scaled for day 1 losses (all the exposed units in the Kara line are at 888 men). The actual losses by this end unit are 80 so we have as good as an exact match there. At the other end we have a similar match (worked out separately but seems correct).
Over all losses are a reasonably good match for the FAQ going as 155/784 on day one and then 142/762 on day 2. The declining rate simply reflecting the shrinking size of the two armies.