WORKSHOP: Universal Simulationist Military System

Masada · Nov 24, 2008

Nice one FC, I was about to give it a whirl tomorrow night

flyingchicken · Nov 24, 2008

Speaking with no authority whatsoever (unless you are a Confusion and believe in leadership by example or something like that), I say go for it (if you meant you were going to do Hastings, do another battle!), and if you have the infos go for an Analysis as well.

Perfectionist · Nov 25, 2008

BATTLE OF GOLDEN SPURS

As far as I have been able to determine, the two armies broke down as follows
Force Strengths
French: 2500-3000 heavy cavalry, 1000-1100 crossbowmen, 3000-5000 assorted infantry
Flemish: 900-1000 crossbowmen, 7000-9000 infantry

We will take the middle of these ranges.

Notes and Assumptions
1) French were the attackers
2) The surrounding terrain is sometimes described as marshy. However, the French were informed by the garrison of Kortijk castle where the best ground for cavalry was. They do not seem to have suffered any difficulty during the battle, and certainly had no problem mounting a cavalry charge. While there was apparently a stream between the armies, the French, by and large, seem to have crossed it without difficulty. The terrain will therefore be taken to be Flat-Mixed.
3) There is no mention of the weather during the battle. It may therefore be assumed to have played no role, Weather will be taken to be Dry-Overcast-Temperate, which applies no modifiers.
4) The French crossbowmen were, in fact, Genovese, and Genovese crossbowmen were universally considered the be the best in Europe. They proved this during the battle, when they rapidly worsted the roughly numerically equal Flemish. A +10% modifier will therefore be applied to the French crossbowmen.
5) The French heavy cavalry were the creme de la creme. They were all veterans, they were all superbly armed and armoured, and they were utterly confident of victory. The primary and secondary sources are unanimous that they possessed a significant qualititative advantage over the Flemish. A +20% modifier will therefore be applied to the French cavalry.
6) The Prepared Defense modifier will be applied to the Flemish.
7) The relevant OLIs are as follows:
Hand-to-hand: 23
Crossbow: 33
8) Season was Summer-Temperate

Calculations:
W_{French cav} = 2750 * 23 * 1.20 = 75900
W_{French infantry} = 4000 * 23 = 92000
W_{French crossbows} = 1050 * 33 * 1.1 = 38115

S_French = W_{French cav} * (Terrain factor)*(Season factor) + (W_{French infantry} + W_{French crossbows}) * (Terrain factor)*(Season factor) = 75900 * 0.9 * 1 + (92000 + 38115) * 1 * 1 = 198425

M = sqrt( ( (7800 + 2750 + 75900 ) / ( 7800 ) ) / ( 1 ) [Flemish had no cavalry, so denominator reduces to N_d / N_d]
m = 3.3292 - (1 - 0.9*1)(3.3292) = 2.9963
P_French = 198425 * 2.9963 * 1.1 (Season factor) = 653995

W_{Flemish infantry} = 8000 * 23 = 184000
W_{Flemish crossbows} = 950 * 33 = 31350

S_Flemish = (W_{Flemish infantry} + W_{Flemish crossbows}) * (Terrain factor) = (184000 + 31350) * 1.2 = 258420

P_Flemish = S_Flemish * (Posture factor) * (Terrain factor) * (Season factor) = 258420 * 1.5 = 387630

P_French / P_Flemish = 1.678.

Conclusions
Based on QJM, the battle should have been a clear French victory. In fact, it was a near total Flemish victory. The French cavalry was wiped out virtually to a man, and total French casualties likely surpassed 50% of the force, while Flemish casualties were probably less than a thousand. We have a problem here.

BATTLE OF AGINCOURT

Yes, I know I said it's too difficult to pin down force strengths, but it's somewhat similar to Golden Spurs and, unlike the other good candidates, I don't need to go look for a monograph. Forces used will be rough figures that I find probable.

English: 5000 longbowmen, 1000 infantry
French: 5000 cavalry, 25000 infantry

Notes and Assumptions:
1) French are attacking. No posture bonus will be applied to the English, since they assumed their position only moments before the battle commenced.
2) Season was Fall-Temperate
3) The English army was certainly qualititatively superior to the French, so a +10% factor will be applied.
4) Similarly, Henry's leadership was certainly superior to his French counterpart's, so a +10% leadership factor will be applied
5) The English morale will be assumed to be only Good, which gives a 0.9 modifier.
6) The English advance immediately before the outset of the battle discomfited the French. This will be taken as Slight Surprise.
7) The terrain was Flat-Mixed.
8) The weather was Dry-Sunshine-Temperate. This has no effects.
9) Relevant OLIs are:
Longbow: 36
Hand-to-hand: 23

Calculations

W_{French cav} = 5000 * 23 = 115000
W_{French infantry} = 25000 * 23 = 575000

S_French = W_{French cav} * (Terrain factor) + W_{French infantry} * (Terrain factor) = 115000 * 0.9 + 575000 = 678500

M = 1 / sqrt(1.3) * sqrt ( ( 30000 + 5000 +23*5000)/30000) = 1.9611
m = 1.9611 - (1 - 0.9*1)(1.9611) = 1.765

P_French = 678500 * 1.765 * 1.1 = 1317307

W_{English longbows} = 5000 * 36 *1.1 = 198000
W_{English infantry} = 1000 * 23 * 1.1 = 25300

S_English = (198000 + 25300)* 1.2 * 1.1 = 294756

P_English = 294756 * 0.9 = 265280

P_French/P_English = 4.966

Conclusions
According to this admittedly basic and hasty analysis, the French should have obliterated Henry. As we all know, the reverse happened. The calculated French superiority is so large that merely messing with terrain, morale or posture factors is insufficient to correct it. Even halving the size of the French army would not be sufficient.

Coupled with the result for Hastings, which predicted a clear Norman victory when the actual battle was an extremely close run thing, these results seem to indicate that there may be a very serious problem for the application of the QJM to NESing before about 1600, viz. that cavalry is highly overvalued. Obviously more tests are required to be certain, but it seems fairly likely that this trend will continue.

Upon brief consideration of the problem, it seems to me that this is likely because we are slotting cavalry into the place reserved in the original model for armoured vehicles. Cavalry, however, have a much greater problem with well-trained infantry than do tanks. This needs to be reflected somehow.

EDIT: Just realized I forgot a -1 in the mobility factors. Can't be bothered to recalculate, but that makes it even worse.

Disenfrancised · Nov 25, 2008

The fact that you seem to ascribe little to terrain in the Agincourt analysis is a bit of a WTH, considering how massively it favoured the English force (muddy ploughed fields were terrible to advance across, and the woodlands meant the English couldn't be flanked and the French could only send equilvalent numbers against the English at a time). Plus the English did have a chance to redeploy their palings giving a considerable posture bonus.

Perfectionist · Nov 25, 2008

Yeah, I wasn't sure about either of those, and they could certainly be changed. The thing is that the terrain possibilities didn't include 'Mud', and I didn't think I could justify calling the field Heavily Wooded or Swamp. The terrain system may need to be adjusted for medieval and earlier battles, as the current options really apply best to battlefields covering several square miles. As to posture, the English assumed their positions ten minutes before the French cavalry advanced, and the palings were, when you get down to it, pointy sticks in the ground. At most, that's Hasty Defense.

At any rate, even if you change both of those, you're not going to get anywhere close to the factor of 20-25 that we need.

the woodlands meant the English couldn't be flanked

Not actually true. The French could have gone around the woods and hit Henry from behind before he knew what was happening. One of the French leaders wanted to do it, and it's why the English panicked when they heard that the baggage train was under attack.

Kal'thzar · Nov 25, 2008

Final Words
I say, the horrid comparative lack of mobility by the Englishmen cost them the battle, and not even the fumbling of the Norman King could have saved them. I did not factor in training and leadership and stuff because I think they were roughly equal (i.e. both sides sucked).

Also, QJM rocks so far, though I cannot do QJMAs because I don't have big fancy books or scholarly websites detailing exactly how far winners went and held land.

Oh and using Google as a calculator is a lot easier than Windows' calc.

I was under the impression that Harold held a fairly strong position and only when unit cohesion was lost (i.e. they thought they'd won and broke their shield wall to charge down), and Williams units rallied and counter attacked did he actually start 'winning'.

Indeed wikipedia agrees with me, all in all in reminds me a bit of one of the Napoleonic battles, forgive me for not finding it straight away: http://en.wikipedia.org/wiki/List_of_Napoleonic_Battles

:/

Birdjaguar · Nov 25, 2008

Kal'thzar said:
Indeed wikipedia agrees with me, all in all in reminds me a bit of one of the Napoleonic battles, forgive me for not finding it straight away: http://en.wikipedia.org/wiki/List_of_Napoleonic_Battles

:/

That would be Marengo in 1800. The Austrians had won the battle and unexpected reinforcements for the French turned the loss into a great victory.

flyingchicken · Nov 28, 2008

Suggested crude solution to show the relative weakness of the horse in very tight combat situations (very pre-gunpowder): multiply the OLI by 0.25 (same as Air Mounted modifier, only without the "Air", hah) or something.

In the meantime, with my suggestion and a few adjustments of my own, here are:

Spoiler reassessed Hastings :

Spoiler reassessed Golden Spurs :

Spoiler reassessed Agincourt :

I also propose: 1) more comparisons with the QJM books and the examples provided and 2) a Battle of Agincourt with different numbers (haha).

If you guys noticed I used Wikipedia to change Perfectionist's numbers and assumptions a bit, but I swear I only did it only if it fudged things.

Masada · Nov 28, 2008

I don't find the results that bad, sure the system can't deal with outliers like Harold biting an arrow in the eye, I get the feeling the system was designed for the sixties, where outliers in warfare are comparatively rare when you have uniform troops, formations, organization, etc.

We just might need to accept that for outliers approximation is not possible, I am sufficiently satisfied that the system combined with some discretion on my part would produce predictable, and rational outcomes from battles. But then most if not all causal factors only become apparent with retrospect so were splitting hairs

. Random chance when it decides to play its hand is not easily predicted, in the absence of pure random chance in an event [a battle] it provides a good predictive result.

flyingchicken · Nov 28, 2008

Masada said:
But then most if not all causal factors only become apparent with retrospect so were splitting hairs .

I believe the "problem" with retrospect in the system is a non-factor if it is going to be used in ahistorical games.

Symphony D. · Nov 28, 2008

Retroactively deleted.

Masada · Nov 28, 2008

I believe the "problem" with retrospect in the system is a non-factor if it going to be used in ahistorical games.

True, but the vast majority of battles are easily predictable, you can say they will win, or them or it will be a close run battle. It's only a small minority where the results diverge all that much from the norm, just think of them as normally distributed. Only the smallest number are outliers, they tend to be the ones we remember simply because they are unpredictable. Look at a given campaign in the Ancient World, and I'm sure you'll find 5 unmemorable forgone conclusions for every upset [probably more]. Sure we can't apply retrospect, but we can act as the arbiter of the facts, balance them, and use our discretion where necessary to vet the final results. There is nothing in NESing worse that arbitrary war outcomes, they tend to revolve on pure numbers, or retrospective reflections of our history. It's better to have a system than to have a series of dice rolls.

So Symphony. Here, here on bringing this system to us.

flyingchicken · Nov 28, 2008

I'm gonna try to do some Gulf War battles if you guys don't mind (just to double-check if the system is applicable 6 years ahead of Dupuy's book's time).

I'm gonna dump stuff on this post, starting with:

Spoiler 125mm 2A46M smoothbore gun :

Spoiler 7.62mm PKT coax machine gun :

Spoiler 12.7mm NSVT antiaircraft machine gun :

Tired. I could have calculated these stuffs a lot faster with my custom OLI calculator if I had it with me, I just realized.

Possible typo from Appendix A: "Machine guns with a caliber of 10-15cm are assumed capable of hitting two targets per strike." Those are some mean machine guns! Speaking of typos, there is also an instance where "x" from what I assumed was "six" is replaced with a middle dot.

Symphony D. · Nov 28, 2008

Retroactively deleted.

flyingchicken · Nov 28, 2008

Okay, though this is unorthodox and highly inapplicable, and for the benefit of those who don't do the maths: taking the total coalition forces totally owning the Iraqis in every battle in the Gulf War, we get the best case total armored casualty rates (these are percentages, so meh) of:
2.80 * 0.60 * 0.70 * 5.4 * 0.80 = 5.08032%/day

While a similarly very very rough approximation of progressive Iraqi losses would be (this is assuming total crushing victory every time for the Coalition):
1.50 * 0.60 * 1.60 * 5.4 * 0.80 = 6.2208%/day
1.50 * 0.80 * 1.60 * 5.4 * 0.90 = 9.3312%/day
1.50 * 1.00 * 1.60 * 5.4 * 1.00 = 12.96%/day
1.50 * 1.50 * 1.60 * 5.4 * 1.00 = 19.44%/day
1.50 * 2.00 * 1.60 * 5.4 * 1.50 = 38.88%/day

For me this looks very impressive at first glance, but I realize that 5.08032%/day is really heavy compared to the actual armored losses ("15 tanks and perhaps 25 other armored vehicles" for the entire war).

This article whose important parts I barely skimmed through (and from which I got that last statistic), says that indeed through superior training and not equipment superiority that led to the amazingly low loss rates.

Having said that: if we can get accurate P/P values for battles between Soviet and Western MBTs and then compile all that data, I see the extension of the Opposition Factor table (found on Page 27 of the rewritten Appendix A) based on what we get, in a way to be later discussed, as the best solution to this problem, maybe at the risk of illegally recreating bits of the modern TNDM. :lol:

---

Consider this post as another one of my spammy ads to draw in help (i.e. help in calculating stuffs and getting sources for battles involving Soviet and Western tanks [hint hint: the Gulf War/s, 'specially the first]) from the peepz.

Perfectionist · Dec 2, 2008

Spoiler Pavia :

Symphony D. said:
The terrain and its extreme effects upon French performance are not being accounted for with the proper seriousness.

I've never really liked that explanation, but I suppose that's beside the point. How would you handle Agincourt in-game? "You rolled snake eyes for terrain, so kiss your whole army goodbye" seems pretty unsatisfying, but I can't think of any other way of doing it. For that matter, how would terrain in general work? A table of probabilities based on general location and strategy?

flyingchicken said:
1) Terrain will be Rugged

In Flanders?

based on Wikipedia's description; just because the French knew where best to use their horses doesn't negate the terrain, I believe

Rather less to do with that and more to do with the French not having any problem with the terrain. They didn't straggle into the Flemish in twos and threes; they charged in en masse at nearly full tilt. It just wasn't enough.

Consider this post as another one of my spammy ads to draw in help (i.e. help in calculating stuffs and getting sources for battles involving Soviet and Western tanks [hint hint: the Gulf War/s, 'specially the first]) from the peepz.

I'd help, but I don't care about anything after 1650. Sorry.

Symphony D. · Dec 3, 2008

Retroactively deleted.

Symphony D. · Dec 8, 2008

Retroactively deleted.

Perfectionist · Dec 8, 2008

Symphony D. said:
- France receives a 70% mobility penalty due to battlefield topography boosting the density of their formation to 200% its initial value and reducing progress to a crawl.

I don't know where those guys got their numbers, but they're . .. .. .. .. A minimum width of seventy yards is ridiculous. Every other source gives a width of somewhere between 800 and 1000 yards, including the primary sources.

- France receives a 50% fatigue penalty due to the extreme physical difficulty of advancing through such difficult terrain (requiring some 40s for cavalry and 180-240s for infantry) and then having to fight a pitched battle at the end of it.

Why don't the English receive a similar penalty? The French only had to go 200-300 yards to reach the reset English line, while the English advanced two or three times that far, through terrain that was described - I believe by the chaplain - as a recently ploughed field such that men could scarcely stand.

P_fr/P_en=0.8155
P_en/P_fr=1.2263

DEFINITIVE ENGLISH VICTORY

But not sufficiently definitive. Agincourt did have a nearly 100:1 casualty ratio, after all.

Symphony D. · Dec 8, 2008

Retroactive deletion.

WORKSHOP: Universal Simulationist Military System

Koi-san!

Deity

Angel of Verdun

Beep Beep

Angel of Verdun

Deity

Hanafubuki

Deity

Koi-san!

Deity

Deity

Koi-san!

Deity

Deity

Deity

Angel of Verdun

Deity

Deity

Angel of Verdun

Deity

Similar threads