Solo Campaign – Siege Tweaks Re-tweaked

If something is worth doing, my grandmother used to say, it’s probably worth doing over again. Here I was, quite happy with my new Hassle-Be-Gone automated siege rules, and then some insightful comments from Ross on my recent post and an unusually coherent email from De Vries the Impaler sent me back to the drawing board.

One of the truly great things about blogs is that you can get other interested parties to shine some light on your own thoughts, and you can learn a huge amount. [By the way, any fans or students of Water Logic? – I used to be a firm believer in all that creative evaluation stuff, though I seem to have forgotten about it since I stopped being paid to think. I might do a post about it sometime – you have been warned.]

The purpose of my mathematical, off-table siege rules is to strike a workable balance between convenience and realism such that sieges can be handled easily in the background while the campaign rolls on. The tricky bit is finding the correct balance – rephrase that – an acceptable balance.

The part of the siege under particular scrutiny here is the actual assault or storm. For a start, Ross raised the very good point that not all sieges are the same. If the defenders are unusually determined, it can change things. I carefully avoid the use of the word “fanatical” here, since it has kind of rabid overtones. Let us merely identify that there are certain situations and certain armies where the defenders would be prepared to fight for every building, and to sustain unusually high levels of casualties. De Vries’ original suggestion was that the defenders might be “Spanish or mad”, but that won’t do at all. 

Further, De Vries cited the Agustina Effect (after the heroic lady celeb from the Siege of Zaragoza), where the civilian populace are prepared to help with manning the guns and the barricades – i.e. commit to a level of active combat over and beyond merely trying to defend their own property. We also agreed that there might be situations (though I’m struggling to think of an example) where the citizens are on the side of the besiegers, and take a part in the attack on the garrison. To put all this into effect, I have changed the calculations of ASS and DSS (as defined in the rules below) in the storm – the defenders can get an extra dice if they are ready to fight for every building (the Suicide Dice - suggestions for a better name will be most welcome), and either the defenders or the attackers might possibly get yet another bonus dice (the Agustina Dice) if the civilians are prepared to fight on their side, during the actual storm. All storms take a week, however they go.

Agustina de Aragon - "No - it's OK. If she really wants to stand there when
we fire, just let her get on with it..."

Ross raised the matter of levels of loss – applying an overall factor to the complete besieging army’s strength to get the casualty figures is over-simplifying things, and may give inconsistent or illogical results. Prof De Vries also pointed out that calculating the besiegers’ losses retrospectively for the whole siege, based on the “total force employed” is, to use his terminology, dumb, for a number of reasons:

(1)  Though the total force, represented by the variable Assault Value (AV) may justifiably be regarded as all at risk during the weekly routine Bombardment Phase (which includes all kinds of missile fire, mining, sorties, hunger, disease, bad breaks and random demoralisation), this number AV will vary from week to week, apart from losses, as a result of troops being detached from the siege, or new troops joining it.

(2) During the actual storm (as Ross also mentioned), only a portion of the total available AV may be called upon to actually assault the place – losses for that week should be restricted to this subset.

(3) In a campaign where weekly returns are made for all units, it makes no sense at all to do the casualty calcs for a siege only when it has ended. It is much better to perform the calcs week by week, as AV varies up and down (or is subdivided), and carry forward the actual totals.

Though still determined to keep this manageably simple, I accept all of this, and the re-tweaked section of the Siege Rules now reads thus:

11.3.3 Storming:

Defenders’ Storm Strength, DSS = FV + GV + 1D6 + the Suicide Dice + the Agustina Dice

Attackers’ Storm Strength , ASS = AV(st) + 1D6  + the Agustina Dice     [BV, the Battering Value, does not count in a storm]

Where:

* The Suicide Dice is a bonus 1D6 available to the defenders if they are prepared to fight for every building.

* The Agustina Dice is an extra bonus 1D6 available to either side if the civilian population of the town will fight for them.

* AV(st) is whatever subset of the full current AV the attackers commit to an assault.

Results:

* If ASS > DSS then the fortress falls and the garrison surrenders. Attackers lose 0.25 x DSS (rounded to nearer whole number) from AV(st). Defenders lose 0.5 x ASS from GV.

* Otherwise, if ASS <= DSS, storm is repulsed; attackers lose 0.5 x DSS from AV(st); defenders lose 0.125 x ASS from GV

[Losses in GV and AV are not simply casualties – they represent all manner of loss of ability to continue – and note that GV and AV can become negative].

Each week during a siege, losses for each side are calculated as one tenth of the percentage loss in AV or GV for the week. During a storm, AV(st) replaces AV if it is different. Thus, for example, if AV is reduced from 7 to 6 during a particular week, the actual loss to the besieging army in killed and wounded is 1/10 x 1/7 = 1.4% of the troops present/engaged.

Solo Campaign - Siege Rule Tweaks

Since it looks as though my campaign may produce a siege quite soon, I was encouraged to go back to my mathematical siege rules, since there were some bits in there I wasn't sure about. There was a post on this system a few months ago - I explained there that, though an algorithmic system for sieges is certainly not a big attraction from visual and fun-generation aspects, it is (sadly) necessary to handle sieges in this way in a map campaign, since a siege will last for a number of campaign moves (and thus must be able to coexist with armies marching and fighting elsewhere on the map in a different timescale) and also since it might be necessary to have more than one siege running concurrently.

The particular bit I wasn't happy about was the section on Storming. Without getting too deeply into the nuts and bolts (again), the idea is that the defenders have a couple of numbers associated with them - a Fortress Value (FV), which represents the strength of the place and its guns, and a Garrison Value (GV) which indicates the fighting capability of the guys in the fort - this is a kind of lumpy amalgam of numerical strength, attitude, and their current ability to carry on - for whatever reason. Similarly, the besieging force have a Battering Value (BV), which is a measure of their heavy artillery capability, and an Assault Value (AV), which is the amount of force they could bring to bear in the event of (you guessed) an assault, but this number also makes allowance for the men who are available for digging ditches, carrying stuff and just constituting a threat.

The detail of the siege rules is set out in the orginal notes, here and here, so I won't go through all that again, but the idea is that bombardment and (implied) sorties and mining etc wear down these numbers. At the point that a storm is attempted, the appropriate section of the rules is set out in its new form below, with the odd explanatory annotation here and there. The inclusion of a dice roll for the defenders and the besiegers is intended to reflect performance and luck on the day. The calculation of losses bothered me - something struck me as counter-intuitive. If the totals for ASS and DSS (as described below) were very close - in other words, if the result of the storm was a close call the casualties would tend to be relatively light, which intuitively seemed completely wrong. A close-fought storm might have the heaviest casualty rates of all, so I've made a couple of changes - I now use the absolute values of DSS and ASS, rather than the difference between them, when calculating loss, and have changed the formulae slightly. It's a minor tweak really, but I'm a bit more comfortable about how it works now. In a campaign, losses have a lasting significance.

Here's the revised section from my Campaign Rules:

11.3.3 Storming:
Defenders’ Storm Strength, DSS = FV + GV + 1D6
Attackers’ Storm Strength , ASS = AV + 1D6  [BV, the Battering Value, does not count in a storm]

* If ASS > DSS then the fortress falls and the garrison surrenders. Attackers lose a final, further 0.25 x DSS (rounded to nearer whole number) from AV. Defenders lose 0.5 x ASS from GV.
* Otherwise, if ASS <= DSS, storm is repulsed; attackers lose 0.5 x DSS from AV; defenders lose 0.125 x ASS from GV
[Losses in GV and AV are not simply casualties – they represent all manner of loss of ability to continue – and note that GV and AV can become negative].

Whenever it is necessary, at any moment during the siege (or when the siege is broken off or completed), actual casualties may be computed as one tenth of the %age loss of AV or GV since the start of the siege. 

Example – if a successful besieging force started out with AV = 8, and end with AV of 6, then they have lost one tenth of 25% = 2.5% of the total force present; if the defenders started out with a GV of 5 and end with GV = -1 then casualties are 1/10 of 120% = 12%; if the fort surrenders, the remaining 88% will become prisoners.

[It occurs to me that if I don't actually get to a siege in the campaign then it doesn't really matter that I've improved the rule, but it's the principle of the thing!]

Trouble Next Door

Tantallon Castle today

After Cromwell's victory at the Battle of Dunbar, in 1650, raiding groups of Scottish horsemen known as the Moss Troopers and the Desperado Gallants continued to operate against the English army, and they were based at the East Lothian castles of Innerwick, Dirleton and Tantallon. Innerwick was soon abandoned, but Dirleton and Tantallon required formal sieges before they were suppressed. Tantallon is of particular interest to me since it is near to my house - right next door in rural terms - and, while I know it well as a local ruin and tourist site, in the 12 years I have lived here I have never got round to finding out much about it, a situation which may seem as strange as it is lamentable.

I am currently reading a lot about the English Civil Wars, and, of course, I now have the opportunity to make good a little of my lack of knowledge. One valuable advantage of living in Scotland is that you can't move for history - it sort of drips off the walls. There are a great many fantastic sites worth visiting, and many of them are well-maintained and accessible - the National Trust for Scotland is a fine institution. Good so far - you can get there and you can look around - but the history itself is not so straightforward. It depends who you ask...

I am not a native of Scotland. I have lived here most of my life, but I did not have the advantage of learning the history and the traditions as part of my upbringing and education. I've made a couple of brave attempts to get to grips with the history, but I didn't do awfully well. I got halfway through Prof JD Mackie's standard work and realised I would have to start again - I had literally lost the plot. Second time through I was taking notes and everything, but it was still very heavy going, and I was very glad there was no exam at the end. Convoluted. Scotland has had periods when there were several kings on the go at the same time. Some of them were Danes, some were Picts, some of them may even have been Scots. The constant conflicts that went on are confused by the superimposition of family, religious and political divisions, and the fact that half of them seem to have been called David doesn't help. And then, of course, they intermarried, and murdered each other, and the French and the accursed English kept getting involved. Even when formal warfare was not current, something very like it would be carried on by the main families.

I apologise wholeheartedly if this seems dismissive or in any other way disrespectful - it is not intended to be so - it is just a summary of the struggle that I, as a relative outsider, have had to understand what went on here - and we are speaking of matters which people have lived and died and fought for over centuries, and from which many of the factions and the grievances are still around today, so I really do have to watch what I say.

Anyway - back to Tantallon. There have been a number of sieges there over the years, but the last was in 1651. It was severely damaged by General Monk's men, and I believe that the Desperado Gallants were granted surrender terms which were much more generous than they expected. Subsequently the castle was substantially demolished to prevent its future use as a military base. The account of the 1651 siege also relates that the walled community of Castletown adjacent to Tantallon had to be fought over first. These days Castleton is merely a farm, with a steading and a Victorian house and a line of 4 cottages - it doesn't look like a military objective. Obviously it was a thriving little town in the 17th Century.

There is some good material and some enthusiastic artwork on a website here. Mostly I am reminded that I am overdue another visit to the castle - it is one of my son's favourite days out, but you have to pick your day carefully. If the weather is not completely favourable, it is potentially the windiest, most wretched place you can find around here. Anyway, I have some more reading to do before I'm ready.

The Siege of Tantallon 1651

A more traditional moody view of the castle, this one painted by Thomas Moran about a hundred years ago. This picture is of personal interest to us since it is obviously viewed from a location we know as 'Our Secret Beach', which is only accessible at low tide - and it's hairy enough then. Either Moran was a quick painter or else he took sandwiches and blankets and toughed it out for the duration

British Artillery Caissons, and Some Very Big Guns

Delayed by a late decision to strip the limbers, here are two examples of what Carl Franklin, in his lovely book, describes as the British Two-Wheeled Ammunition Car. A quick glance, of course, will confirm that the car is hooked up behind a standard limber, so it is in fact a four-wheeled vehicle, but articulated, which was regarded as a big advance over the earlier rigid 4-wheeler. These are the carts which accompanied the individual batteries into action, to provide an immediate reserve of ammunition.

The models are Lamming throughout - equipment and horses, and also the drivers, as evidenced by their Easter Island profiles and the trademark Lamming elephant whip. My thanks and compliments to Clive and to Dave Watson, who somehow came up with yet more supplies of extinct artillery kit.

Since I am deep in the artillery projects box at present, I think I may take the opportunity to make up and paint some more siege guns. As these may be of some interest, here are a couple I prepared earlier. I included a more normal 9pdr gun to give an idea of scale, and you will see that these siege guns are very bad boys indeed. These are 18pdrs from Hinchliffe's (current) 25mm scale range, which should make them way too big for the Minifigs gunners. Before you laugh (and I laughed myself before I checked the sizes), be assured that I have measured these castings and they are spot-on for 1/72 of the official weapon dimensions for an iron 18pdr. Further, Clive and I once put these same Hinch 25 castings alongside a Finescale Factory model of an 18pdr, and they were exactly the same size - I am not even prepared to consider that FSF would ever make anything which was not perfect 1/72, so let's just assume this is what they were like.

Big.

Anyway, I have 2 or 3 more of these to prepare, and a 10" howitzer, so I may take a short break from painting vehicles. Note also that my Allied Siege Train and associated engineering chaps have their bases painted a fetching shade of mud brown. It seemed a good idea at the time.

Solo Campaign Siege System - worked example

This follows on immediately from my last posting, and this is me attempting to try it out with a historical example. To be specific, the Allied siege of Ciudad Rodrigo in January 1812, and - to be even more specific - this starts off as history, but as soon as the dice start rolling, anything can happen, as we know.

One inconvenience about real history is that it is not easy to get a handle on some of the data - at least in your own pocket campaign you would know all this stuff for certain.

Ciudad Rodrigo has a Fortress Value (FV) of 6 in my table. This also limits the garrison to 6 combat units. From the appendices in Belmas, I know that the French force consisted of a very large but raw battalion of the 34e Léger, a battalion of the 133e Ligne (who were Tuscans, I am interested to note), a couple of companies of Artillerie à Pied (12th and 13th companies of the 6th regiment), about a dozen engineers and various staff monkeys and administrators and that's about it. A bit less than 2000 men, and the Garrison Value (GV) of the 4 units is - well, 4. Barrié, the garrison commander, had succeeded to the post when the appointed commandant (Renaud?) had been captured while inspecting the fortress's herd of beef cattle (the cattle were also captured). To get back to the plot, Barrié did not wish to have the position, and is generally regarded as unimpressive - we'll regard him as officially "Poor" in the motivation/leadership department, so no bonus is added to the GV.

Wellesley's force is not so easy to pin down. It is well recorded that the actual storm was carried out by the 3rd and Light Divisions, with support from Pack's Portuguese brigade, but this siege system is more interested in all the troops Wellesley had at the siege. This will then cover everyone who was available for digging and providing a threat, and who could potentially have been involved in a storm. By this wider definition, we should include the 1st, 3rd, 4th and Light Divisions plus Pack's Portuguese. If we lump together the various attached rifle companies into an extra battalion, this gives a total of 34 battalions plus 4 divisional artillery companies; 38 divided by 4 gives an Assault Value (AV) for the Allies of 10, near enough, which we should raise to 11 in view of Wellesley (supported by Fletcher and co) rating at least "Good" on our leadership scale. Wellesley appears to have had about 18000 men in these combat formations.

The designated Allied battering train consisted of 38 heavy guns which were deployed as 5 batteries - that's quite a lot of guns for 5 batteries, but we'll go with a Battering Value (BV) of 5.

To summarise, then, the defenders have FV = 6, GV = 4, the attackers have AV = 11, BV = 5.

Week 1

Bombardment: French have GV of 4, thus roll 4D6 (I'm actually rolling dice, rather pathetically, as I type...) - they come up 5 3 1 1 - the 5 deducts 1 from the attackers' AV, but there are no 6s, so no hits on the siege guns (BV),
Simultaneously, the Allied battering guns (BV = 5) roll 5D6 - 5 4 3 3 1 (not very good shooting, maybe they get better with practice?) - no 6s, so no damage to the fortress (FV), but the 5 scores against the garrison, so 1 comes off GV.

Now FV = 6, GV = 3 (total = 9) for the French, while AV = 10, BV = 5 for the Allies. The Allies do not bother asking the fortress to surrender, since their AV of 10 is extremely marginal for a storm against the defenders’ (FV + GV) = 9. No storm

Week 2

Bombardment: French now roll 3D6 - they come up 6 4 2 - the 6 hits the battering guns, and reduces the Allies' BV by 1
Meanwhile, the Allied batteries (BV still 5 - doesn't get adjusted until the end of this phase) roll 5D6 - 6 5 3 3 2 - the 6 removes 1 from the Fortress Value, the 5 removes a further 1 from the garrison (GV).

Now FV = 5, GV = 2 (total = 7) for the French, while AV = 10, BV = 4 for the Allies. AV of 10 still looking risky for a storm against (FV + GV) = 7 - bad luck with the dice could be disastrous. The Allies don't summon the garrison to surrender, and make no attempt to storm.

Week 3

Bombardment: French now reduced to 2D6 - they come up 6 5 - cheers from the battlements - they have the range now! Allies lose 1 off each of AV and BV
Allied batteries roll 4D6 - 6 4 4 1 - that's another 1 from FV - those walls are looking a bit second-hand.

Now FV = 4, GV = 2 (total = 6) for the French, while AV = 9, BV = 3 for the Allies. The Allies still don't fancy the chances of a storm, and don't ask for a surrender.

Week 4

Bombardment: French roll 2D6 - they come up 1 1 - useless. There may be trouble about this...
Allied batteries now down to 3D6 - 5 3 4 - the walls are standing up surprisingly well, but that's another 1 off the garrison (GV)

Now FV = 4, GV = 1 (total = 5) for the French, while AV = 9, BV = 3 for the Allies. Another week might improve the situation, but Wellesley decides to storm the fort now rather than lose further time (it's only a test...).

The Storm:

Defenders' Storm Strength DSS = FV + GV + 1D6 = 4 + 1 + 2 = 7
Attackers' ditto ASS = AV + 1D6 = 9 + 5 = 14, which is no contest – Attackers win.

Thus the storm is successful. The attacking force lose 1/2 x GV = 1/2 = 1 from their AV in the assault, giving a final figure for AV of 8. The surviving garrison are taken prisoner. Total loss for the Allies during the 4-week operation is 1/10 of the %age loss in AV = 1/10 x 3/11 = 2.73% which, for a force of 18000, is about 490 men. During the storm, the French lose 1/2 of (ASS - DSS) = 3.5, which is more than enough to eliminate their last surviving GV point. Thus the French have lost 100% of their GV, and actual casualties are 1/10 of this - 10% of the original 2000 men is 200, and the balance are prisoners.

OK - that worked. I'll try a couple more to see how it goes. It's not as much fun as a tabletop siege with a model fort, though. On the other hand, it didn't take me an entire weekend.

Sieges in my Solo Campaign Rules

This is getting into serious Nerd Territory, so be warned. I’d been thinking about the subject anyway, and was prompted further by Clive’s posting of Paddy Griffith’s algorithmic system for simulating sieges.

The background to this is hinted at in my previous post on my developing solo Campaign Rules for the Peninsular War. Well – not The Actual Peninsular War, of course, but a similar sort of war in the same area, using similar troops, around the same time.

They had a lot of sieges - it was a feature of the warfare. It would be lovely to be able to trundle out my model fort and fight siege games as part of a campaign, but sadly it is not practicable. The campaign has a weekly turn, and an ongoing siege would almost certainly last a number of turns. For a while I considered the possibility of maintaining a table-top siege game while the map moves went on in another room (or something), but that stopped abruptly when I realised that two or more sieges might be running simultaneously – and what then, eh?

So - like it or not - sieges are going to have to be settled by mathematical means, off the table. I have a copy of the NapNuts campaign rules, which handle sieges by means of A Duckenfield’s rules which were first published in Practical Wargaming – in fact his rules are lifted straight in, which is probably a vote in their favour. Being an awkward sod, I decided to set up my own system, though it is on not dissimilar lines.

I need the off-table siege “game” to give realistic durations and casualty levels for the siege operations, and I need it to produce reasonable results, but to operate crudely enough to work without destroying what is left of my poor brain (or enthusiasm). My earliest drafts were far too complex – this is about the fourth redrafting, and I am sure it has some distance to go. The trend has been toward progressive simplification throughout. I have glossed over many things – some because it was convenient to do so, and – doubtless – some because I just hadn’t thought of them.

This is where I have got to. It’s not finished by a long chalk, but it’s coming along. If this is the sort of thing you like, you may well like this.

Sieges in the Campaign Rules

The turns are 1 week long.

Defenders

Certain of the Areas on the map contain towns which are fortified and have a Fortress Value (FV). This reflects the size, strength and position of the installation, plus (amongst other things) an allowance for some resident artillery. The initial FV may subsequently be reduced by a combination of factors – damage to the walls, loss of guns, being compromised by siege works and siting of batteries – anything which renders the place less formidable.

The fortresses on the map, with their FVs, are:

Abrantes (P)           4

Alicante                  8

Almeida (P)            5

Badajoz                  8

Barcelona               8

Bayonne (F)           6

Bilbao                    6                              Towns marked (P) are in Portugal, (F) are in France

Burgos                   7

Cadiz                     12                           This is not intended to be a complete historical list -

Cartagena              10                              it is drawn up simply for the game!

Ciudad Rodrigo      6

Elvas (P)                5

Figueras                 8

Gibraltar                15          

Granada                 7

Jaca                       4

La Coruna              8

Lerida                    6

Lisboa (P)              7

Pamplona               6

Perpignan (F)         6

Sagunto                 10

San Sebastian         7

Santander               6

Tarragona               8

Tortosa                  6

Valencia                 7

Vic                         7

Zaragoza                7

Fighting troops in the fortress contribute a Garrison Value (GV). The number of units stationed in a fortress may not exceed its basic FV. For this purpose, a unit is a battalion of infantry (strength usually 4 CCN “blocks”) an artillery unit (3 blocks) or cavalry regiment (3 or 4 blocks). Initial GV is equal to the number of units (though cavalry and militia count 1/2 each), and its reduction during a siege will represent both casualties and diminution in “resolve” of the garrison. GV may be increased by a further 1 if the garrison commander is an identified Leader rated as Good or Outstanding (2 or 3).

Besiegers

The besieging troops are not limited in number, though supply may be an issue. The initial Assault Value (AV) is equal to 1/4 of the number of fighting units present (rounded to the nearer, half up), and may be increased by 1 if the overall commander of the besieging force is Good or Outstanding. The Battering Value (BV) is simply the number of specialist siege batteries present – these may be battering guns, mortars or rockets.

Thus, at the start of a siege, the defenders will have a FV and a GV, and the attackers will have an AV and a BV. These factors may all be impacted subsequently by enemy action.

If FV becomes zero, the fortress can be entered at will and the garrison will surrender.

If GV becomes zero, the garrison is no longer able to resist, and any survivors will surrender.

If AV becomes zero, the besieging force is no longer able to continue.

If BV becomes zero, the attackers have no bombarding artillery available, and will normally call off the siege (unless they expect some to arrive!).

If AV + BV is less than FV + GV then the attackers will normally call off the siege.

Each turn (commencing at the end of the first week of siege), the procedure is

Bombardment phase (all bombardment is simultaneous, so don’t make any deductions for hits until both sides have fired - the dice throws include the effect of accidents and bad breaks as well as direct hits).

* Defender throws [GV]D6 (i.e. a handful of 6-sided dice, GV in number) – any 6s will put a siege battery out of action (reduce BV by 1 for each), any 5s cause loss and demoralisation to the attackers’ forces and engineering (reduce AV by 1 for each).

* Attacker throws [BV]D6 – any 6s will damage or compromise the strength of the fortress (reduce FV by 1), any 5s cause loss and damage to the garrison and their resolve (reduce GV by 1).

* Adjust FV, GV, AV and BV for bombardment.

Protocol phase

If besiegers have not abandoned the siege, and if the fortress has not automatically surrendered, the attackers may summon the fortress to surrender. If AV > FV + GV + 5, then the fortress should at least consider surrendering, since they could not withstand a storm. However, a storm would kill more of the besiegers, and, if the defenders are French, they will be aware that the Emperor has given strict orders that no fort may surrender until it has withstood at least one assault....

The besiegers may elect to storm the fortress.

Storming phase

Defenders’ Storm Strength, DSS = FV + GV + 1D6

Attackers’ Storm Strength , ASS = AV + 1D6

* If ASS >= DSS then the fortress falls and the garrison surrenders. Attackers lose a final, further ½ GV from AV.

* Otherwise, if DSS–ASS is positive, storm is repulsed; attackers lose (DSS-ASS) from AV; defenders lose ½ this amount from GV

[Remember that losses in GV and AV are not just casualties – they represent all manner of loss of ability to continue].

Whenever it is necessary, at any moment during the siege (or when the siege is broken off or completed), actual casualties may be computed as one tenth of  the %age loss of AV or GV since the start of the siege.

Example – if successful besieging force started out with AV = 8, and end with AV of 6, then they have lost one tenth of 25% = 2.5% of the total force present.

Relieving forces will cause the besiegers to break off the siege, or at least divide their forces.

Sorties are abstracted as part of the unpleasantness which the defenders can inflict during the Bombardment.

Thus far, I have not explicitly addressed the question of the defenders’ supplies....