Rules ideas - when is enough 'enough'

Or to put it more clearly when do battlefield combat formations become unable to stand in the line of battle?  As wargamers knowing when a unit is done it is one of the central measures of whether we have won or lost a game.  Most modern analysis on how long a unit might continue in combat focus on the percentage of casualties suffered and the consensus is that other than in unusual circumstances most units are done somewhere between 20% - 30% casualties.  I suspect that this focus on casualties is because it is a measurable metric.  It is also accepted that a the specific level of casualties is masking the impact that other factors have on a unit's willingness to continue in combat.  Things like level of fatigue, hunger, weather conditions, terrain, unit cohesion and leadership.

Historically rule sets have often used figure removal as a marker of a unit's decline in effectiveness (WRG ancients for example).  In such rules a unit's effect on the enemy is based on a weapon factor multiplied by the number of figures able to fight.  So reducing figure counts makes a unit less effective.  When a unit gets to a set number of figures lost a reaction test is carried out, for example in WRG 4th Edition (pulled at random off the shelf) that trigger point is 50%.  The likelihood of a unit breaking is greater the lower the morale class of the unit is there is a reduction from the dice roll for each 20% of figures lost.  A nice simple trigger and easy to remember during a game BUT it can lead to units fighting on with possibly unrealistic casualties and other factors.  There are other triggers such as taking a high number of casualties in a single turn but the preset percentage is the one where a unit is almost certain to break.

The last stand at Gandamuck 1842; but how common were they?

Yet in reality history is littered with famous last stands, some of these are statistical outliers (although that's an insult to those brave souls who fought in them) others represent situations where a unit has no  choice but to fight on as surrender means death for them or for others.  This is a situation described by Sun Tzu who tells us to always leave a defeated enemy an avenue of escape so they will flee rather than fighting to the last man.  

So how do rules writers cover this situation?  The issue is that we are modelling the effect of combat on human beings and we are all individuals (a little voice should now be shouting "I'm not" from your memories of 'Life of Brian') and while we become more predictable in larger groups we are still not clones who will react the same way to the same circumstances every time.  To be honest I'm not even sure that clones would be so predictable but lets leave nature v nurture for another day.

If we want to model the factors which create the situation where a combat unit become 'combat ineffective' then what should we be looking at and how should we model it in rules?  This is probably a good time to admit that I don't really know the answer.  I have some thoughts but no certainties.  What I think impacts the outcome should be (and bear in mind I'm primarily considering periods before  1750AD or so) are:

• Units degrade in fighting ability over time.  I have read that 20 minutes is about the limit before individual soldiers are mentally and physically exhausted and need to be pulled out of a fight.
• The level of training, morale and combat experience have an impact on their ability to continue.
• Higher levels of casualties are going to reduce the time a unit has in combat.
• Good command and control will have some effect on keeping troops motivated to fight.
• Other environmental factors do play a part.

Trying to build all of those factors into some kind of test is going to be difficult and lead to lots of dice modifiers.  The question becomes do you want long detailed tests or a quick and easy result.  That in turn depends on the kind of game you want to play and the command level you are playing at.  Equally the level the player is handling in the chain of command and the nature of the figure basing is going to play a role.  A set of rules where one base is a regiment will need different mechanisms than one where each figure is one man and the 'army' represents a squad.

The options are endless!  For example I could see a viable solo game where the player is a company or platoon commander managing his unit, with the enemy and the environment being handled by a solo mechanism to create changes to the situation that the player has to deal with.  It wouldn't even need toy soldiers on a table it could be handled with markers on various morale tracks. Covering casualties, supply, hunger and thirst, fatigue, combat stress, belief in cause, support for commander, support for the rest of the combat team etc.  The player would have tools available like sending hot food forward, rotating squads or sections (or maniples or centuries), evacuating casualties, resupply of ammunition and similar things to bolster the will to stick with it.  On the other hand at a higher level of command you leave that stuff to the officers on the ground and concentrate on the big picture.  Which is perhaps where the US obsession with body counts in Viet Nam arose.

In my latest home brew rules I have chosen to ignore body counts altogether in favour of tracking the overall willingness to continue in combat.  Units have an initial combat effectiveness rating that degrades during combat due to a number of factors (and one does sort of measure the level of casualties) but the unit never gets any smaller.  In part this is based on the concept of the most of the killing being done by a very small percentage of the soldiers with the rest just going through the motions and not really trying to harm their fellow man.  I make shooting far less impactful on a unit's state than hand to hand combat where both sides can go to pieces very quickly, As I consider close combat is going to take a toll not only on the person being injured but also on the person doing that (unless they are a psychopath and they don't tend to make good soldiers).

So this has been a post on rules ideas without a hard and fast rules suggestions, rather I have a question for you.  How do you want rules to model this do you prefer detail or a more abstracted system?  Is there a  point in the size of the game where you preference changes?  do you see differences in treatment as vital for different historical periods?  Answers on a postcard to the comment section!