Back in 2005, I posted an article on a history forum where some really cool intellectuals and military minds in America and Europe were discussing history. Recently someone asked me what the result of that was. This is part of my thinking and interest in Prediction, and worth adding into the discussion about The Future – which clearly concerns all of us, not just in Southern Africa but elsewhere in the world.

The Reader wrote:-
Hi,
I have run across your mathematics of war threads.

Did you manage to do the modeling you were attempting? How did it go? What models did you choose? Did they work?

When I saw the discussion, I became curious about how things went.

Bernie

Here is Jan’s original post in 2005:-

I am looking for information on what one might call the Mathematics of warfare. By this I am referring to equations such as the Lanchester equation, Col Du Puy’s QJM/HERO… and anything else.

I’m looking for models and maths to calculate:-
o battle casualties, advance rates, logistics, etc.

I know what I’m asking for is a very wide topic… but if people can point me to different sources I would appreciate it. Also, I’m looking for any info on any era. I just want to get good sources and good reading material with a statistical/mathematical flavour.

Also… does anyone know of any mathematical models for guerilla warfare… for calculating the outcome of a guerilla type war… e.g. like Napoleon’s penninsula campaign or Vietnam, etc… i.e. when conventional troops come up against irregulars who are launching surprise attacks, ambushes, etc. I am assuming that this is an area of study where little if anything has been done… but I’m asking anyway.

Its one thing calculating the outcome of a single battle… but its another when you’re dealing with a vague concept like guerrilla warfare… and yet… I would think it must be possible.
Thanks,
Jan

One particularly important reply came from an American Artillery Officer whose posts I’ve read many times and who actually fought in the first Gulf War. I regard him as an extremely intelligent man – based on his many observations in the past. He did knock the wind out of my sails a bit with his comments. But I think he took them in the wrong vein. Nevertheless I have a great respect for his views:-

Howie,
You’re more than correct citing Murphy. I haven’t seen Murphy wrong yet.

The Dupuys have done the same thing Jomini did-take the human element out of consideration. That is not a wise thing to do.

We did computer simulations before going to Kuwait in 1990. According to the computer, the US and its allies lost every time. So much for computer simulation.

Patton said the one constant in warfare was man. Tactics, techniques, and technology change, and man does not.

You can apply mathematics to logistics, artillery fire direction, and a few other sciences that are included in the military art. However, Napoleon didn’t consider warfare an exact mathematical science, though he did consider the study of mathematics essential for officers-undoubtedly because math tends to discipline the mind to think logically (for a change). He always planned carefully, but changed his plans as the situation dictated, and went with the flow if he had to. He also made his own luck, which he considered as the ability to exploit accidents.

‘First, one must see…’

Napoleon also considered the simplest moves to be the best and was a master of the nine principles of war.

If anyone wants a ‘mathematical’ application to warfare, see Scharnhorst’s opinions about his Prussian contemporaries before Jena. they are quite enlightening.

Sincerely,
Kevin

Hi Bernie,
The posts and replies were interesting. There were certain interesting discussions and comments, but nothing more that I could use in the way that Colonel DuPuy’s ideas could be.

The short answer is that I did indeed go ahead and I did indeed built a model along the lines of what I envisaged in that email. I even played two hypothetical campaigns using historical and invented data. It was an interesting exercise.

I did later find some papers on the Internet which I printed and read. One was a fascinating paper about how Colonel DuPuy’s formulae prove the Lanchester Eqations (of battle) true. My feeling was that there is a lot of truth in this. The key thing in all these calculations (and which I think Kevin misses in his pooh poohing of numbers), is the fact that you have soldiers of different quality. DuPuy deals with this in depth in his analysis of the world wars. I think it is fair and obvious to say that not all soldiers are the same and any simulation must make the correct assumption of quality of troops.

What Kevin never did say was whether the basic assumptions of the Iraqi army did indeed take into account the lower quality of the Iraqi forces. Also, did they simulate General Schwarzkopf’s surprise attack through the desert and did they take into account the effects of a surprise attack on the Iraqi army? Surprise is a huge factor.

I am convinced that when they did the modelling they must have made errors in their assumptions.

It is interesting that the US Army and senior commanders did still attack even though their models told them they’d fail. Does it mean they don’t trust their own models or do they sense their own models may be wrong?

In the end, lacking new ideas of how to model, I simply went and took all DuPuy’s writings and started building a set of programs based solely on Du Puy’s writings. Du Puy changed his ideas as time went by too. I did not have access to the very latest ideas – the ones which supplanted his QJM and HERO systems – so I merely took what he wrote in the 1970s and 1980s and I started writing a web-based wargaming program.

I would have liked to model tanks and aircraft, but I did not get that far. The tanks, artillery, etc instead need different loss and replacement rates and I did not get around to modelling that.

But to test the basic idea I went and did this. I wrote a simple game that can simulate infantry and cavalry and I then went and created a hypothetical and extremely simple scenario which I wargamed. It was terribly simple but I actually had a lot of fun with it and I want to do more of it.

I ran my biggest simulation to simulate 77 days of warfare. It took me several weeks, playing a little bit each evening.

The main thing I wanted to do with DuPuy’s equations, tables and historical data was to take them in the form he gave them, build a computer program around them and then run them “raw” and see what happens. I did not want to invent too many rules (except where absolutely necessary) – because I wanted to see if one could get a “feel” for the “logic of war”.

Let me elaborate. I did not want to have preconceived ideas about certain things – I wanted to run the game and then see what logic works in the game.

Its best if I explain two things I learnt after playing my simple simulations. My first simulation was truly simple – it was really in order to debug the program. It was a tedious thing. I simulated about 120 days of battle over about 2-3 months to debug the program. When it was pretty bug free, because of limitations on my mapping (this is where my game ultimately broke down), I decided to take the original game map, expand it, add a nice hypothetical story twist and then play a simple, but heck of an interesting game.

Now the original “test battle” was simple. I have a lovely map of South Africa in 1890 – complete with roads and rail. So I superimposed very simple graphics (truly primitive stuff) over it so that my units could move over the map and I could see the entire map. I wanted to test the quality versus quantity concept.

So because building logic for tanks, artillery required more work – I only wrote up DuPuy’s tables and calculations so that I could simulate: Infantry and Cavalry. Since he has a very wide variety (all of history) set of weapons, I then made some educated guesses and I had the Zulus, armed with the equivalent of Roman javelins/spears, and I gave them some guys with muskets. Then I chose the Orange Free State (OFS) in South Africa as my battleground. So I played a surprise Zulu invasion of the Free state. I looked at historical precedents like the Battle of Blood River, etc and did some quality calculations. So I settled on relative quality between the Boers and Zulus. I allowed for historical factors like the Zulus being able to march much faster than normal infantry.

As an added twist, I added in individual runners/horse riders as messengers in both armies. Since the battleground is hundreds of kilometres in size, I wanted to see how a campaign would be affected when you have bad communications – no telegraph – only riders, and troops and settlements spread over hundreds of miles. So the first test game (for debugging) was simple – a column of about 11,000 Zulus armed with spears and a few hundred armed with muskets sneak over the border and fall on town after town with the element of surprise. Then as the first town is caught in a panic they send off messengers to other towns, riding on horseback. This creates a big time delay factor in the battle. I played the commanders, and I made copious notes of what info each would receive, when, and what would he know at that time. Then he issues orders based on what he does and does not know.

Among the things I built in the game from DuPuy were:-
1. Quality of soldiers.
2. Daily attrition
3. Battle casualty calculations.
4. The building of hasty defences (takes up to 10 days).
5. Different terrain as it affects marching & horses
6. Different terrain as it affects battle outcomes
7. I built in the full effects of surprise and its diminishing effect.
8. I later added trains including the loading and unloading of trains based on my best thumbsuck guesses. Loading 1000 troops and horses on to a train would take much longer than loading say 200 infantry on to a train, etc. So I made the best guesses I could.

I did not:-
1. Build in random weather – I selected set weather appropriate for the area
2. I did not build random effects into the battle calculations – I stuck 100% to his formulae. I wanted to build randomness in, but later found the whole thing so interesting and tricky that I decided to just leave it running as is.
3. I did build in “incorrect intelligence” into the game. I did this to simulate the errors which scouts make. So, unless two forces have actually been in battle, I used these “estimates” (which could be too high or too low) as the information the commanders would use. So if 100 Boers were approaching the enemy – their scouts might report it as being anywhere from 50-200. This means that commanders can get partially wrong figures on which to make decisions of whether to attack, defend and retreat.
4. I did not build in armoured logic.
5. I did not build in the loss rates for artillery.
6. I did not build in logic for river crossings (DuPuy’s own rules on this were not easy. I did think of a nice way of doing it, but its a lot of work & I felt it was not crucial for my simulations. I could estimate/guess some of it.. and that was good enough.

When I was happy that the logic was working… I played a 2nd and much more interesting game on the same map.

The second game I played was a complex scenario I thought up. I used a hypothetical invasion of the British and Zulus along with rules that the British wanted to arm various natives they came into contact with. So as the British invade in 3 columns (one from the west, another in the east and another in the south), they bring with them wagons with muskets which they will use to arm any black natives they come across who will then join their army. I gave the British and Boer small numbers of artillery – a single cannon here and there. My artillery loss calculations would thus not be too affected. I wanted to see what effect a single cannon of the time had on a battalion-sized battle (it proved very interesting since one cannon was effectively worth a good few hundred men).

So I used the same surprise attack theme. 3 British battalions, along with a big Zulu force cross the borders. This gave the battles a great quality since there were mixes of Boers versus sometimes only natives and sometimes only British and other times a mixed British/Native force.

In the east of the OFS are the Drakensburg mountains, and that was a different feature to the very flat lands which existed elsewhere. And I wanted to see the effect of this on the battle. It slowed the movements and battles down a lot. It was great for defence. There was one very surprising battle where I think 200 entrenched Boers held out successfully against 2,000(43)+ Zulus in the game. I had expected them to be defeated… but they held on until much stronger reinforcements arrived. Eventually they were driven back.

But since the game was based on a surprise attack with no communications for the Boers except horse-riders it was fascinating. It took about 2 weeks for the news to reach the Boer commander in Bloemfontein about the attack on all fronts. I did include in this scenario a railwayline to see how that would affect the Boer strategy and I had a lot of fun with it. I expected the game to be a defence from a central position with the Boers using mobility to dash north and south fighting various armies coming at them. But that’s where the game became very strange.

So it took 2 weeks for the Boers to fully realise and appreciate the onslaught coming from all sides. It took another 2 weeks for them to send riders to the distant towns and to raise the commandos in each town and order them to fall back on various positions. So the first month consisted mostly of retreat and losing towns with small fights here and there.

Where the terrain was flatter, the battles moved faster. Action in the west occurred weeks before action in the east. Lots of interesting things occurred because of the time delay with orders and the movement of information since both the British and the Boers were always several days behind the events which were actually in the field and sometimes information would take 2 weeks or even longer to reach them. It also made the game tedious.

I enjoyed the strategic mobility that the trains offered. It was actually fabulous. I really loved the single railway line and could see what power it offered to a commander in those days.

I tried to stick to relatively historically accurate quality ratios and troop ratios. I made 80% of the Boer force mounted and had only a few infanty. The British were mostly infantry with a few mounted troops. The Zulus were mostly armed with spears and had a few hundred with muskets.

What I wanted to do was make DuPuy’s formulae and tables come alive. The troops marched along roads and trails that I laid out, and nowhere else. They marched and fought from town to town. DuPuy allowed for withdrawals. I modified it somewhat to distinguish between troops just walking away, and those withdrawing in an orderly fashion. I also had to think up logic for a unit which is being attacked from the front and the rear at the same time. I did this successfully without warping his ideas.

I did not allow for morale. I made that decision by myself. I eventually adopted the rule, based mostly on Anglo-Boer war instances – that if a unit fights a continuous battle for 1 or more days and loses 25-30% of its strength then it will surrender. I also introduced prisoner exchanges between the sides later. Units did not fight to the death. I never allowed that since that is unrealistic.

So, what did I learn if anything?

The main thing that surprised me, and it answered a burning question was: Concentration of Force. One hears about concentration of force all the time, but I found myself, almost unwittingly, splitting forces into ever smaller units sending them to watch over roads, towns and covering possible invasions. I found myself whittling away reserves and sending them piecemeal here and there – wasting time and achieving little. It took a TREMENDOUS conscious effort to resist this desire to split forces and be everywhere and think in terms of concentration.

I found this “splitting concept” to be the most interesting result of the simulation and I can now easily realise why in the real world commanders make this error. It is extremely natural. I actually enjoyed discovering that I was falling into this trap constantly and sometimes I was so guilty of it that the whole campaign was in jeopardy.

I think I didn’t do too bad in making a lot of DuPuy’s stuff come to life.

What eventually caused me to stop was a technical bug in my map. It proved to be a big nuisance and made the game really tedious. My next aim was to simulate the first 6 months of the entire Anglo-Boer war – which at a point resulted in a British army of 250,000 men. I did start building up copious notes of the order of battle for both sides along with estimates and guesses for the parts I did not know. But my technical map glitch meant that setting up the map for the entire Boer war could take me weeks. It was far too tedious. So I let it be. I needed to find a proper technical solution. I have now found a solution – and need to build it in. Once I’ve done that, I will be able to run simulations again.

Note – the simulation I produced is actually web friendly – it generates HTML and can be put on the Internet. But, running it is tedious. It is not point-and-click. You have to sit and input data tediously, give orders exactly correctly… its a long, slow process. The calculations run a long time too. DuPuy’s calculations are for 1 day at a time. I had to modify them because of the problem of forces running into each other. So I do calculations for all forces every 15 minutes. So I break a day into 15 minute chunks so that forces will march and meet each other at a realistic point. So for example, two forces may meet at 11:30, and then they fight for part of the day. So its not a full day’s battle. These 15 minute chunks result in the calculations being extremely slow, and it takes about 5-10 minutes for it to do all the movements and battle calculations for a single day. If I simulate the whole Boer war, then it could easily take 15 minutes or more for it to process the actions of one day.

For the Anglo-Boer war though, I would abandon the whole “rider/runner” concept since they used telegraphs and heliographs which were very fast. I would have a simple rule that would allow battle information to travel long distances in 1-2 days.

I’d like to wargame 6 months of the ENTIRE Boer war… with all the major forces simultaneously.

To make the simulation work for the modern era, the main missing thing is armour. I need to build in the armous rules for movement and for losses and replacement. Ditto for artillery. But once that’s in, I can then also do something new – I could compare Infantry with body armour’s losses versus infantry without body armour. There is no question in my mind that two infantry battalions, of equal size, same quality, similar weapons but the one having body armour and the other not – that the ones with body armour would steadily whittle the others down and eventually destroy them.

If I do the above, and fix the map, I could then run a simulation for any country, anywhere in the world with land forces. Air is more difficult – I can’t simulate air properly except in a vague fashion. I think one could bring in air… in a certain way if really was needed… but I’m focussing more on ground battles.

I would like to test armour and Blitzkrieg tactics.

From what I picked up of my simple simulation, artillery is a serious multiplier.

For fun, below are the notes of the Boer side of the war as seen by the Boer commander. I noted the dates he got the messages versus the dates when the scouts saw it. All the references to “I saw” and “I was at” are merely notes as I wrote them up as a scout would report them. Its very basic, but it will give you an idea of the flow of the battle. By this time the program was pretty well debugged. But it did crash once or twice and the battle at Smithfield caused it to crash. But I merely took those results and carried on anyway even though the outcome of that one battle was faulty. The rest of the campaign flowed smoothly.

I’ll also send you a map of the battle after 77 days – when I stopped. The blue squares are the Boer units, the Red are the British/Zulu/Native forces. The big box which reads Bloemfontein, is merely showing you what units are in Bloemfontein at the time. My “armies” and units were small – a mere few hundred men. Post=Posture – as defined by DuPuy. You will see one set of black crossed swords on the map – that is when two enemies are in close proximity and a battle is about to begin or is in progress. At this late stage the British were driven out of the west, and a much bigger set of battles was yet to be fought in the east, primarily in the mountains. I estimated that the Boers needed at least another 3 months or more to completely drive out the invaders – so my guess is the whole campaign would have taken about 6 months.

And note… you play the game one day at a time, issuing new orders only at the end of each day – which is quite realistic enough for me.

Note, I included in this the “quality” issue. I based it on Boer versus black battles and Boer versus British battles and worked backwards. The result was that the average Boer soldier was worth more than 2 British, and I ranked the Native/Black troops as the lowest quality. So in this you will see interesting British/Black armies fighting smaller Boer forces and losing heavier numbers of men or even losing. These calculations are exactly as per DuPuy’s formulae. I thought the results were pretty realistic. In one battle you’ll see the Boers losing 21 men while the British lose 50 and the Natives lose over 20. (It also depends who was attacking). The fieldguns made a big difference. I only allowed about 5 fieldguns in the whole scenario and having even 2 in one battle meant overwhelming firepower.

I think the entire Boer army I allowed for in this scenario was somewhere round about 1,000 men. The British started with 1,800 and the Blacks numbered a few thousand – plus – if the British reached Thaba Nchu, I would have awarded them another 2,000 blacks, 1,000 of which they could arm with rifles. But I stopped the game before it went that far. I can still carry on with the game, since all the games I played are saved and I simply can continue with them.

So here, for fun is the Boer Commander’s view of the 77 days of battle, of this hypothetical war along with his centralised “summary of knowledge” of what he does and does not know and surmises at that point in time. The Secret invasion starts on 1880-10-01:-

1880-10-03:
30 British mounted troops were spotted at the Boshoff Turnoff heading in the direction of Bloemfontein! It looks as if we are being invaded!

1880-10-04:
In the late afternoon, a British force of: 730 men, of whom 70 appear to be mounted were spotted 14 Km from Philippolis, moving in our direction. We raised the alarm. Our commando is retreating towards Fauresmith.

1880-10-04:
In the late evening a massive force was spotted 6 Km from the border post on the Natal side. It consisted of: 70 British mounted troops, 3000 Zulu Impis, 380 British soldiers. We expect them to attack us. We are abandoning the border post.

1880-10-06:
Scout for the 30 British mounted troops. Try to ascertain what is going on.

1880-10-08:
With 3 witnesses, we spotted 60 BMI (British Mounted Infantry) busy digging in 100Km from BFN on the road to Kimberly.

1880-10-10:
The Fauresmith & Philippolis Commandos are digging in to the south of Fauresmith. They have no news from Jacobsdal.

1880-10-10:
General mobilisation ordered across the OFS. Reinforcements sent south & west to counter-attack the British.

The Bfn General has ordered the following:-
We are to go to Kroonstad & then to Johannesburg with the following message:-
1. A general mobilisation of all Commandos in the Free State has been ordered.
2. Kroonstad is to mobilise Heilbron, Frankfort, Bethlehem & Harrismith.
3. Kroonstad must organise Riders to check whether the British in Natal made any hostile moves.
4. The Frankfort & Heilbron Commandos must go to Kroonstad.
5. The Harrismith & Bethlehem commandos must go to Harrismith.
6. Kroonstad must keep only 50 men. All the rest must be sent immediately to Bloemfontein.
7. Winburg must mobilise & must protect the Winburg siding in case of attack from Hoopstad.
8. Ficksburg & Ladybrand commandos must protect Ficksburg.
9. Jacobsdal, Fauresmith & Philippolis Commandos must defend Fauresmith.
10. Bethulie, Smithfield & Rouxville commandos must defend Bethulie.
11. All Commandos must mobilise & group as ordered.
12. All Commandos must send out scouts to check for British activity along all roads in their region.
13. Probing attacks MUST be launched against all British forces unless they are overwhelmingly superior.
14. The results of all scouting and probing attacks must be reported by rider to Bfn.
15. If the British fall back after an attack then keep on attacking and drive them out of the OFS.
16. If this is not possible, then notify Bfn of how many reinforcements will be needed.

1880-10-12:
Delivered the message in Jhb. Pta telegraphed to say they could be called on if more troops were needed. They would offer their support.

1880-10-14:
Arrived in Kroonstad. Kroonstad Commandant has 50 men & asks for orders from Bloemfontein to counter this massive threat in the east. Kroonstad has no other news from the outlying towns regarding this development in the east.

1880-10-16:
We have 300 men in Bfn who can assist Kroonstad & the east. However, due to the close proximity of British troops on Boshoff road & the south, we will rather abandon any large scale counter attacks in the east until we know that the British have taken Bethlehem. Once the British have actually advanced & taken Bethlehem we will make a decision. Inform Bethlehem not to make a strong stand but to rather abandon their positions if indeed strong British/Zulu forces arrive.

1880-10-18: Sent: 1880-10-14
The Bloemfontein scouts of 60 men attacked the BMI who were digging in. We discovered there were 50 of them. In the 1st day of fighting we lost 4 men and they lost 16. We drove them from their positions. The next day we pursued them & they fought a withdrawal battle. We lost another 2 men, and they lost 8 men. Then the remaining 26 of them surrendered! Our men will continue attacking & probing forward towards Boshoff. Total casualties: 6 for 24.

They told us they are part of the 1st British Batallion of 600 men & that other men went on to attack Boshoff and Hoopstad. They are based in Kimberly. They also say they will also arm the Blacks against us!

1880-10-22: Sent: 1880-10-16:
At the Boshoff turnoff we saw a British rider racing toward Boshoff. We also spotted 150 Natives armed with spears moving towards Kimberly. We decided to attack them immediately before the British armed them with guns.

1880-10-22: Sent: 1880-10-17:
After attacking, we discovered there were only 100 Basotho. We attacked the Basotho from Boshoff in the rear. They were armed with spears. After 3 hours of fighting we had 2 casualties & they had 48. They surrendered to us. They told us that there were many British troops in Boshoff. The Natives were being sent to Kimberly to be armed & to help the British to fight us.

We will now go to Boshoff & probe the British there.

1880-10-22: Sent: 1880-10-08:
In the evening we spotted 300 British Infantry moving towards Jacobsdal coming from Kimberly. We abandoned the town & our Commando has moved to Fauresmith. We are awaiting orders from BFN as to what to do next.

1880-10-26: Sent: 1880-10-21:
Bloemfontein Scouts reconnoitred Boshoff. There are 150 British Infantry digging in. We only have 52 men. We want the Western Army to join us as soon as possible so that we can make a probing attack.

1880-10-27: Sent: 1880-10-23:
The artillery joined the South Western Army. The advance & probe towards Philippolis starts today.

1880-10-29: Sent: 1880-10-15:
I approached Jacobsdal & saw 150 British Infantry digging in to the east of it.

1880-10-30: Sent: 1880-10-15:
I reached Rouxville. They will mobilise & join the Smithfield commandos who are digging in outside Smithfield. They are waiting for the Bethulie commando to tell them whether the British are attacking in this direction.

1880-10-30: Sent: 1880-10-24:
In Bethlehem we spotted: 1400 Zulus & 760 British troops 20Km outside Bethlehem. The 105 Boers in Bethlehem decided to abandon Bethlehem & retreat to Kroonstad

1880-10-31: Sent: 1880-10-22:
I spotted 470 British Infantry 100 Km from Hoopstad heading in the direction of Hoopstad.

1880-11-03: Sent: 1880-10-30:
The Boers the 3rd Bn Scouts were pursuing turned around & attacked them.
Surrendered to the Boers after a stiff fight with the Bethlehem(43)+Harrismith Commandos which lasted almost the whole day. They lost 22 casualties. Boers lost 16 men. They took 27 prisoners.

1880-11-04: Sent: 1880-10-29: From: Western Army in Boshoff.
Our scouts estimate that there are almost 100 British retreating from Boshoff. We are down to 103 men. We estimate that the British in Boshoff suffered about 25 casualties & we had 22. The fighting was tough & has taken its toll on our men. We cannot pursue them without reinforcements. They are retreating towards Hoopstad. So maybe they will join up with other British there. We will dig in, in Boshoff until we get new orders or reinforcements. We need a minimum of 50 men, and preferably 100 men, before we can resume offensive operations.

1880-11-04: Sent: 1880-10-31:
At the Boshoff turnoff I saw 120 Natives, armed with rifles & 240 British soldiers moving towards the Boshoff turnoff. (They are either coming here or going to Boshoff). They very nearly caught me.

1880-11-06: Sent: 1880-10-28:
SW Army attacked British & Native forces dug in at Philippolis. We used our field gun to good effect. Losses in 2nd Bn Rearguard battle at Philippolis:-
51 British, 22 Native. (Boers lost 21 men). We drove the British into Philippolis.

1880-11-06: Sent: 1880-10-31:
After 2 days of fighting a withdrawal, the British 2nd Bn RearGuard surrendered to us. They originally were 348 men. They were down to 199. The following surrendered:
Basotho: Philippolis Riflemen: 57
British Infantry: 113
British Mounted Infantry: 29
Total: 199

1880-11-08: Sent: 1880-10-27:
400 British attacked 150 of us in Smithfield. After 2 days of fighting they had lost about 100 casualties & we lost 30 of 150 men. They drove us into the centre of the town. The British had a fieldgun. We need reinforcements. We will conduct a withdrawal to Tha Nchu & await any reinforcements we can get from Ladybrand or Ficksburg or Bloemfontein. We need 100-200 men. More likely 200 men. But even 100 would help.

1880-11-23: Sent: 1880-11-11:
I reached Smithfield. It was occuppied by British & Native troops who were all digging in. I saw 60 British troops, and 250 Natives. A half of the natives were armed with rifles.

1880-11-24: Sent: 1880-11-19:
Disaster! The Bfn Army fought against over 500 British/Native Troops on their way out of Boshoff. There were about 300 British & 200 Natives. They also had 1 field gun. After 1 day of fighting, 162 Boers surrendered to them. The British did suffer some heavy casualties too. They were unable to break through our lines. But our losses were too heavy. They also captured our field gun. We assume the British also destroyed the Boer Western Army which was in Boshoff.

1880-11-24:
The British inform us they have a total of 210 captured Boer troops. They want to do a POW exchange of 195 men only however.

1880-12-03: Sent: 1880-11-29:
Spotted a British force of 430 men moving from Kimberly, but they had not yet reached the Boshoff turnoff. Am not sure whether they are going to Boshoff or Bfn. Will retreat a bit to Bfn & watch them.

1880-12-03: Sent: 1880-11-30:
The British force seems to have split because only 180 men are now still moving towards Bfn. Only 1/4 of them are natives.

1880-12-03: Sent: 1880-11-23:
We spotted 2140 enemy troops 26Km outside Ficksburg, heading south towards Ficksburg, from Bethlehem. Approximately 140 of these troops are British soldiers & 270 are armed Zulu riflemen. There seem to be 1730 Zulu Impis.

1880-12-08:
A British force of 590 men, of which 148 are Natives and 442 are British, arrived outside Bfn. They might have 1 or 2 field guns. We are dug in and are 471 strong. We have a force of over 350 men & 1 field gun expected from Kroonstad. We will continue digging in, in the hope that we can hold out until reinforcements arrive. Then we should be able to overwhelm them.

1880-12-12:
The British 1st Bn, with 209 troops & 1 field gun surrendered to the Boers.

Current Boer Knowledge:-
150 British in Jacobsdal (previously 300 – where did they go? To Boshoff?)
250 British & Natives towards Boshoff & Hoopstad MAYBE 1 fieldgun????? (not?)
590 British & (1/4) Natives towards Bloemfontein MAYBE 1 fieldgun/2

140 British, 270 Zulu riflemen & 1730 Zulu Impis in Ficksburg

1000 British, Zulu Impis & Riflemen in Frankfort
760 British (43)+ 1400 Zulus in Bethlehem (are there less in Bethlehem now?)

0 British in Philippolis
0 British in Bethulie

Here, for fun is an example of the log that the calculations print out. It can make for tedious reading and I kept it mostly to browse and investigate for errors. A shortened log of the highlights is what I mainly used. For example, I had an ordering system where I issued orders telling units to move and what posture to adopt. For example, I would tell units to advance and if they meet the enemy to decline battle and to immediately start digging in, or you could tell them to advance and when they meet the enemy to attack. Things such as fatigue, etc as from DuPuy’s calculations were in there.

This is a small excerpt from a time period of 30 minutes of a day. Most of the news and the log pertains to the tedious marching by the units – since that is what they’re doing. When their orders are finished they stop and wait for the next set of orders. You will see the system calculating movement rates depending on the terrain.

76 10:15 Boer Bfn Infantry: 8 Fatigue: 1 Adv=0.2
76 10:15 (Final Variance Adj) Adv=0.2
76 10:15 Boer Bfn Infantry Terrain: Rolling, mixed. Speed: 0.2 8Km/Day
76 10:15 Boer Bfn Infantry Time: 1 Speed: 0.2 Dist: 0.2 8Km/Day From: 10.4 To: 10.6
76 10:15 Boer Bfn Infantry Order: Advance Loc: Bloemfontein to Thaba Nchu Turnoff 114Km 10.6(43)+ Posture: 1
76 10:15 Boer Kroonstad Cavalry Awaiting Orders at: Bloemfontein to Boshoff Turnoff 197Km 0- Posture: 1.2
76 10:15 Boer Artillery: 1 Boer Jhb Artillery Fatigue: 1 Adv=0.3
76 10:15 (Final Variance Adj) Adv=0.3
76 10:15 Boer Artillery Terrain: Rolling, mixed. Speed: 0.3 12Km/Day
76 10:15 Boer Artillery Time: 1 Speed: 0.3 Dist: 0.3 12Km/Day From: 27.2 To: 27.5
76 10:15 Boer Artillery Order: Advance Loc: Bloemfontein to Thaba Nchu Turnoff 114Km 27.5(43)+ Posture: 1
76 10:15 Boer ReCaptured Artillery: 1 Fatigue: 1 Adv=0.3
76 10:15 (Final Variance Adj) Adv=0.3
76 10:15 Boer ReCaptured Artillery Terrain: Rolling, mixed. Speed: 0.3 12Km/Day
76 10:15 Boer ReCaptured Artillery Time: 1 Speed: 0.3 Dist: 0.3 12Km/Day From: 16 To: 16.3
76 10:15 Boer ReCaptured Artillery Order: Advance Loc: Bloemfontein to Thaba Nchu Turnoff 114Km 16.3(43)+ Posture: 1
76 10:30
76 10:30 British 1st Battalion Wagons Awaiting Orders at: Kimberly to Jacobsdal 64Km 1(43)+ Posture: 1.2
76 10:30 British 2nd Battalion: 164 British 2nd Battalion Infantry Fatigue: 1 Adv=0.2
76 10:30 British 2nd Battalion: 1 British 2nd Battalion Artillery Fatigue: 1 Adv=0.2
76 10:30 British 2nd Battalion: 95 Basotho: Bethulie Fatigue: 1 Adv=0.2
76 10:30 British 2nd Battalion: 95 Basotho: Smithfield Fatigue: 1 Adv=0.2
76 10:30 British 2nd Battalion: 96 Fatigue: 1 Adv=0.2
76 10:30 British 2nd Battalion: 97 Basotho: Rouxville Fatigue: 1 Adv=0.2
76 10:30 (Final Variance Adj) Adv=0.2
76 10:30 British 2nd Battalion Terrain: Flat, mixed. Speed: 0.2 8Km/Day
76 10:30 British 2nd Battalion Time: 1 Speed: 0.2 Dist: 0.2 8Km/Day From: 173.6 To: 173.8
76 10:30 British 2nd Battalion Order: Advance Loc: Smithfield to Thaba Nchu 181Km 173.8(43)+ Posture: 1
76 10:30 British 2nd Battalion Wagons: 10 Fatigue: 1 Adv=0.1
76 10:30 British 2nd Battalion Wagons: 1 British Rider #1 Fatigue: 1 Adv=0.6
76 10:30 British 2nd Battalion Wagons: 1 British Rider #7 Fatigue: 1 Adv=0.6
76 10:30 (Final Variance Adj) Adv=0.1
76 10:30 British 2nd Battalion Wagons Terrain: Flat, mixed. Speed: 0.1 4Km/Day
76 10:30 British 2nd Battalion Wagons Time: 1 Speed: 0.1 Dist: 0.1 4Km

I have thought of running simulations for all sorts of hypothetical and historical games once I fix the map and later build in armour and artillery losses/repairs.

I have wanted to run simple basic Iraq/Iran invasion scenarios or some very rough, fully blown Middle East land war.
I do want to run some historical simulations. I would love to run some Roman simulations, or a Napoleonic campaign – a small one – say Italy.

But even tinier simulations could be good fun and instructive. For example, I have researched a little known scenario in Rhodesia (now Zimbabwe), where in 1979, the Zipra forces, armed with Soviet Tanks and Migs were going to send 5 Brigades into Rhodesia, crossing at Kariba, and making a dash for the capital. The Rhodesian army prepared for it including building a smallish type of wheeled tank killer. The invasion never occurred because Rhodesian SAS blew up bridges in Zambia thereby stopping the armour and mechanised force from attacking. I would have liked to simulate the surprise attack if it had occurred and like with the Boers, count the days and see the progress of these armoured columns while forces are rapidly pulled from various theatres and brought in to fight them. Will they reach the capital? What will happen to the two armoured columns?

I also have toyed with simulating civil wars, including a hypothetical future Zimbabwe Civil War and studying it. I have looked at maps and drawn various conclusions about Zimbabwe strategically, but it would be fun playing various hypothetical situations. African wars are of interest to me. Another war of interest, and worth simulating, would be the invasion of Namibia by South Africa in 1914. South Africa invaded, by land and sea, Namibia and fought the German army there and defeated it in a swift campaign.

I quite like DuPuy’s results. Its too tedious for most people. At one time I considered taking a single game, and making it available on the web allowing people to play it. But its not user friendly and its slow. You really have to be deeply interested in the technical stuff to actually make the effort.

I also spent time using DuPuy’s various formulae to calculate the TLI and OLI values of weapons. What could result from this – is the theoretical “invention” of weapons that do not actually exist – but which you then wargame as if they existed. Then you assess how good these theoretical weapons would possibly be.

This simulation business can be extremely time-consuming. The movement and battle logic is no joke. It took me weeks of tedious writing and debugging. There are still limitations in the way the units can move. Devising a simple movement system that would work with DuPuy’s formulae and allowing the units to move – was no joke. The battles are really complex.

I allow reinforcements. The way the battles run, reinforcements can arrive at any time of the battle and join in the fight and it will correctly reflect the different fatigue and loss levels. I took DuPuy’s formulae and I implemented them exactly – perhaps even better than he implemented them.

I never went beyond DuPuy’s formulae. I wanted to play with Political simulations and with Guerilla warfare simulations. I even toyed with a Spy-simulation. But I never wrote any of these. I merely sat and thought about them, but did not implement them.

On my own private website I want to run a theoretical exercise called, “The Prediction Experiment” to focus mostly on the Political prediction side of the thing.

I’m actually looking for good books on serious wargames as run by military types. I’m also looking for a military handbook on logistics. This is all tedious, dry stuff, but I’m interested in it.

I’m looking for any thinking and input regarding simulation of the real world.

I did think of simulating guerilla warfare in a very vague form. At one time, some years back, I sat and did some calculations from the Rhodesian war to see if I could detect some kind of formula, but I came up empty-handed. I think its doable… but not easy… and it has to be conceptual…

I really like what DuPuy did. Even if it has errors, he tried to quantify things. Even if its not perfect, it still gives a pretty realistic feel. But the best way to test it would be to fight historical battles and wars and compare the real results with his calculations.

I probably drowned you with my reply. I was wondering if I should post it on the forum to show them the nonsense I got up to after I asked those questions. Jan.

/pred/