Translating historical army lists to wargame scenarios

[mmcv]

I'm interested to know how people generally go about preparing army lists for historical scenarios? Do you try and match up like for like for every unit? Pick one unit to be representative of a few? Assign units based on their battle impact rather than actual number on the rosters?

For previous scenarios I've either had reasonably well defined OOBs (such as for the ECW) where I can design up a like for like army list then cut it down to more realistic levels from there, or very poorly defined OOBs (like the Crusades) where numbers involved vary wildly from source to source and proportion of troop types are little more than educated guesses.

I've hit up against something in the middle. I'm sorting out some army lists for my Sengoku Japan project and looking specifically at the battle of Mimigawa where 30,000 Shimazu bested 50,000 Otomo.

Those numbers seem to be reasonable, though likely rounded up, but I've not found anything disputing them, and while I don't have an exact break down of number of guns, bows, spears etc I do have information from the time period that lets me out together a reasonable distribution.

Looking at the Shimazu army there are a number of generals named and their positions are well known but the number of troops under each is a mystery. The only one with reasonably clear details is the garrison force from a castle that sallied out and hit the Otomo rear. They consisted of 1000 Shimazu troops (a large unit) and 500 vassal troops (a standard unit). Keeping the pattern of numbers consistent I would need around 60 units to represent the whole army. And 100 for the Otomo. This is obviously unrealistic so sensible to bathtub it somewhat. Therein lies my riddle though. For that known force that sallied out, do I keep them at their historical number where they were a reasonably effective presence in the fight or reduce them to a single standard or even small unit if cutting the overall number represented in half or a quartet.

The other positions for their army are the main body, the headquarters and the left and right flanking units. Historically the main body staged a feigned retreat over a river, before stopping and fighting back supported by the headquarters and the flanking units hit the enemy in the sides while the garrison sallied out to the rear.

My inclination is to keep the garrison force at 1500, since they wouldn't be particularly effective if any smaller, then treat the flanking units as fairly small detachments and have a larger main body and headquarters. Doing that gives me about a quarter the number of troops and 15 units, a much more manageable number to put together. The flanking units were fairly effective too though so having them quite small may be ineffective. Equally though I don't want to peel off too much from the main body as it's still the bulk of the army and acts as the decoy to draw the enemy in.

Interested to read how others put together OOBs for historical armies where the details are a little hazy.

[John Cook]

Generally speaking I pick a battle that interests me and research it to death.  I use the historical OB, the formations (Brigades and Divisions) and units reflect the historical ones which is relatively easy to find for battles from around the mid-17th Century onwards but for earlier battles like Hastings or Hattin, all you can usually find are approximate strengths of sections of the army and the proportion of mounted to foot troops, if you are lucky.  In these cases, in the absence of any detailed information, you have to extrapolate.  For me this means average unit strengths and, often, complete guesswork where numbers of units and their types are concerned.

The best I could find by comparing multiple sources was that Alan Fergant's Breton contingent at Hastings , for example, comprised from approximately 500 to 800 Cavalry, 800 to 1100 infantry and 300 to 500 Archers.  To arrive at an OB for wargaming considerable extrapolation was necessary and in the absence of any information I invented unit identities, the names being derived from historical personalities but not ones necessarily associated with the Norman invasion, and their strengths a plausible division of the overall numbers.  

I used the same methodology for the Norman and Flemish contingents, and for my Kingdom of Jerusalem and Ayyubid armies at Hattin.

[Big Insect]

Generally - I try and start with the proportions of different troop types - e.g the ratio of foot to mounted (for example) but I also look at the shape of the battle - as often you can have a battle that is won by the smaller % of the original army (say the Foot units played no part  for example).

I then look at the component elements of the different troop types - was this an army where c.60% were foot archers and c.30 armoured dismounted men-at-arms etc. and the remainder was 'oddities' - Irish kerns or Gascon crossbowmen etc.. I then try and ensure that I have at least 1 representative unit from each troop type - so in this current example I'd probable have a single small unit of Kern LF javelins and another of Gascon LF Xbows, but the Archers (longbows) might be in bigger formations, with the dismounted MAA in intermediate size formations.

I also look at the 'impact' a small unit might have had on the battle historically.
So at Tewkesbury - for example - the small (c.500 man) Yorkist ambushing force (regardless of whether you are of the view that they were mounted spears or a foot unit) had a significant effect on the outcome of the battle by falling unexpectedly onto the flank of Somerset's division - as it attempted its own enveloping move. It might be that <500 men could not warrant/justify a full 'unit'  in your own ratio of troops to table-top figures, but their impact is significant enough to include them.
Often the same will apply to artillery or a few Elephants (such as the 80 @ Zama for example). Or in a 'classical ancients' context - Palestinian Clubmen in a Mid-Imperial Roman army, warrant inclusion because of the significant impact they had upon the battle, by negating the Sassanid Persian Cataphracts armoured advantage.

Certain, periods are easier than others - ECW and TYW battles (as long as you are trying to replicate battles in the 'core heartlands' of the conflicts) tend to have relatively few troop types.
But even then it is easy to get overly hung up on the 'points of difference' - such as Hasselrig's 'Lobsters' at Devizes (for example) - when their true impact on the battle was really not that much. But sometime a small unit - such as Dragoons defending hedges - can have a bigger impact than their numbers would warrant otherwise.

I'm aware I am counter-arguing my point here - but I think it is all about balance - I've seen large table-top ancient battles, such as Pharsalus (for example), where the focus on the non-legionary elements in both armies was such that they resulting table-top battle ended up being 'lost' by Caesar before any legionaries actually got to combat, as Pompey's auxiliaries and cavalry (which outnumbered Caesar's) had destroyed enough break-points of Caesar's auxiliaries to meet the victory conditions. Or a game based on Zama, where it was impossible for the returning Numidian/Roman cavalry to hit the back of the Carthaginian infantry lines (which did happen historically) to tipped the game in the Romans favour.

I think it is about looking for the 'core points of difference' in the battle. Pharsalus is a good example, as if you boil down the numbers you end up with a lot of legionaries facing each other in a long line. The real 'differentiator' in the battle IMHO was the differing quality of the two sides legionaries and their commanders (& command structures) - think of Pompey as CV7 in BKC terminology and Caesar as CV10 and you get the idea  but it is not about the auxiliaries and the cavalry - which is what a lot of gamers get hung-up on (as it is the exciting element - not hundreds of units similar legionaries on both sides!).

Not sure if that is in any way helpful?

Mark

[steve_holmes_11]

There are two approaches based on the detail available.


The "Early system": where the record goes into little detail about units - or when the concept of a unit didn't exist as it does now.

Try to obtain best estimates of numbers of each side, and any breakdown of specific troop types.
Then apply a pinch of salt to the 600,000 Persians or 250,000 Gauls who ended up on the losing side.

Sometimes a contemporary account might list different contingents, but it's generally guesswork.
Find your figure scale, and apply long division to yield your army figure count.


The "Later system": where records stretch down to unit level.

Find one or more accounts and make a best estimate of numbers.
Identify units involved as far as possible, or at least get their numbers.

Figure out how many units your rules need (Will differ of your rules deal in battalions or brigades) and attempt to assign correct units to commands as per battle.


One difficulty of the later system is reconciling estimated numbers with "paper strengths" of units.
Marching units will generally be below strength for a number of reasons.
Some sources will help with this, other times, a knowledge of the particular army will explain discrepancies.

Ultimately, there's only so much you can do.
You're looking to provide a best estimate as a basis for a game.
So don't lose too much sleep over it.

[mmcv]

Thanks for your thoughts! Seems to line up fairly similarly to my own approaches. I don't necessarily get too tightly constrained over exact numbers etc, just try and find something that is reasonably representative. Particularly when dealing with accounts without too many references.

but I also look at the shape of the battle

I think this is probably a key point, particularly for historical scenarios, where you want to model the effect rather than precise numbers. This is where a straight division of the numbers involved by the figure scale doesn't always match up. I think it's particularly notable when talking about smaller numbers. I.e. the difference between an army of 49,000 vs an army of 50,000 would be considered roughly equal, but an army of 1500 vs an army of 500 is a whole other story, obviously because of the ratios rather than the numerical difference. So for this scenario I'm not sure if quartering the 1500 men (1 large + 1 standard unit) down to 350 men (1 standard unit) would be particularly effective. I could either keep it as 1 large and 1 standard to have a reasonably big impact, or possibly half to 1 standard + 1 small or even 3 small units to provide a little more flexibility. I know approx. 1000 of the men were under Shimazu Iehisa and 500 under Yamada Arinobu but don't actually know how they were divided or deployed, though can extrapolate ratios of troop types based on standard armies at the time and troop muster lists from other battles.

Because most of my units are mixed anyway I don't need to mind too much on the proportions, just their role in the overall battle. I maybe need to run some play tests with a similar scenario with the rules. See for sure how much impact a single unit can make on the rear of a bigger formation vs a couple of small units doing the same.

The other bit I need to work out is the flanking divisions. On my initial impressions I had thought these to be mostly ranged troops pouring fire into the enemy but it looks more like they engaged in close combat, so likely just keep them mixed rather than ranged specific units. Need to do a bit more reading on other battles though as the Shimazu used the feigned retreat and flanking tactic something like eight times, so might glean more information from other battles around the time!