Agreed, a lot of us are getting caught up in the semantics of the question. In the game, we're not actually talking about a "peasant" i.e. a civilian with no training or inclination towards combat, even if the unit is called a peasant. "Militia" or "conscript" would be better terms, i.e. a peasant that's been drafted, handed a club, and received very little training, but we're still talking about a military unit - not just some farmer or cobbler or baker off the street. Naturally the actual civilians are never seen on the battlefield, anything that appears in Elemental's tactical combat is by definition a military unit, and at least minimally prepared for combat.

As to the original question, given the above definition of 'peasant' - I think there should be room for lucky, unexpected victories in some situations, but not in extreme situations. If your 100 peasants - excuse me, conscripts - with clubs are fighting 100 veteran light infantry with leather armor and wooden spears, I think maybe 20-25 peasants should get 'lucky' and win. If we're talking about well trained and equipped knights with steel plate, maybe 1-5 peasants get lucky and somehow win. I think you get to a point though where it's hard to believe even a small chance of a peasant winning, and allowing that to happen at all - no matter how rarely - is just going to frustrate players when that unlikely chance comes up. There really should be no chance for a conscript with a club to beat a hero with the best equipment possible - think about it, do you want to be the player whose hero gets "unlucky" and dies in a trivial encounter?

To put it into RPG terms, if you're a level 10 then a level 8 should have a respectable chance of killing you, and a level 5 should pose some small threat, but you should never lose to a "lucky" level 1 -  that's like the lucky phalanx beating a tank in older versions of Civ, it's just taking the luck factor too far. If our level 10 uber hero has to fear even level 1 peasants with clubs, it neuters the RPG aspect of the game; you just can't invest yourself too heavily in any one unit, because sooner or later he'll get incredibly unlucky and that'll be it. And how seriously can we take a dragon if sufficient numbers of terrible units can overwhelm it? I can see a decent unit - say a well trained longbowman, as in The Hobbit - getting that lucky shot in, but I never want to see a peasant with a club killing a dragon. You may be able to justify these extreme situations realistically - ok, given a ridiculous enough number of peasants, sooner or later our hero will trip and sprain his ankle - but I don't like the gameplay effects it'd have.

So we now have some notion of the relative chances of victory given certain unit types and quantities, and that's all very well and good. But how are you going to implement a system that reflects these types of odds? Obviously, you can't just have the game enforce what 'should' have been the outcome, so you need a combat system to account for these expectations.

I've made some suggestions already, so does anyone else have an idea in mind of how these calculations should be carried out in the game? What combat mechanics, unit abilities and stats, or algorithms should be made use of?

That would be classified as miitia, they had some training and were armed.

The easiest & most realistic would involved a good morale system as the #1 variable, other variables are less important. A wide range of morale values would be needed. Three aren't enough. I'd range morale from something like 5 to 10 if you're using a 2d6 system (I think ASL got the morale range down very well, as one good example). I don't know how this game models morale since it's not covered in the manual; I've never seen anyone "break," and I've very occasionally seen a guy "panic" and loose his move & attack capability.

Like the other guy said, it's a matter of semantics. "Peasants" should be called something else, at least that's what I would've done, but I'm not the dev, so he can do whatever he wants with his game & what he calls his troops.

Game systems that model character statistics, like most RPGs, are designed from the ground up to be single-man (scaled) games. It's like this game is trying to be both a single-man & company level game at the same time, and it doesn't really work very well (as many other threads like the hero vs squad threads attest). I mentioned above, modelling peasents at the single-man level (townsfolk in a barroom brawl) is cool in a single-man scaled games, but if you're trying to create peasent battalions or whatever (like the OP post) then that's just not realistic for reasons I already covered in another post. 

It all comes down to what the game scale is: single man, platoon, battalion, regimental, division, army, etc... It seems this game wants to be both a single man and company level (8x12 = 96 men on tactical battles at most). Since this game is attempting to model "squads" of peasents, I would get rid of the "peasant" term and change it to something more appropriate & accurate, "conscripts" and "militia," which differ in morale. Perhaps the "conscripts" would be used in situations where an enemy is attacking the town itself, and these guys are auto-added to the defender in amounts based upon 1) city size or population size 2) amounts/types of available weapons (I would have a "tax" or "slush fund" or something that defines the number and types of weapons for exactly this purpose). These guys would have very low morale (i.e. should break and run after the first guy in the squad dies) and should be penalized when attacking (say -5 DRM). There's a couple of problems though 1) the tactical grid is too limited (if you already have 12 units on the battle, you can't add any more guys) 2) tactical battles don't allow units to "retreat" off-map, which seriously needs to be fixed, as allowing heroes to retreat but not any other unit is just plain silly. In this game "militia" could be modeled by allowing military units to be produced at only half the training time of normal units. Their penalty would be lower morale, but not as low as conscripts, and maybe just a -1 DRM when attacking. Their first level-up would be be considered a level up to the base level of newly-created fully-trained military unit. Another problem in this game though is that training times can be reduced to just one term, so you'd never really have militia if you just build the correct builidings .. just build all your squads from those cities which crank out at 1/turn, so what need for militia if this is the case? 

This would allow for hastily trained units which a kingdom could use if it was immediatly threatened, which is basically how conscripts & militia were used anyway. Oh, and get rid of the peasents (just rename it) in this game, unless you want to be a full-blooded single-man scaled game, in which case they should probably only be available in villiges and rural towns/farms for single-man encounter purposes and barroom brawls.

this would be exactly what I meant by "scale." What you're talking about is a man-to-man scale.

Frankly, I think even one peasant taking out two armed knights on the single-man scale is pretty much ridiculous without some external factor/variable: magic, lightning bolt strike happens to hit the knights, knights fall in a tar pit, peasent is armed with molotov cocktails, knights are completely un-armed for whatever reason, peasent rolls a boulder on knights from canyon wall, etc. 

I don't think this would make for a good game (the basic situation without the variables: if one peasent could beat two armed knights in a hand-to-hand melee), but that's just my opinion. I mean, why bother getting knights if even peasents can beat them? Those are some sucky knights, boy I'll tell you. A "real" peasent by defenition simply would not be able to do this without external variables .. maybe the odds of a peasent being succesful in this attack should be equivalent to winning the mega lottery, like 1 in 10 million or so. A strong farmer is not going to beat even one trained knight 99.999% of the time without external variables in melee combat. Again, look at the history of peasents vs trained military, the end result is always massacre.

1) a "peasent" does not have the military training (even if armed) to use his "pitchfork" or whatever in a military way (as in thrust, parry, lunge, parry & riposte).

2) an average "peasent" would not have the morale to attack two armed knights (unless that peasent was mentally deficient or there were external variables that would overcome self-presevation and push the peasent to suicide, such as a temporary form of insanity) ... rage would work against the peasent in the actual blow-by-blow attack and support the knight.

I understand we're discussing RPGs and we're not talking realism, so I'm just speaking for myself since I prefer realism. Instead of modelling a peasent, why not just model a mercenary or someone (gladiator, carnival knife-thrower, local archery champ), anyone that would have some reasonable level of combat or weapon-skill proficiency.  I mean, there's hundreds of possible examples of people that would have a better than Nil chance against a knight that can be used in a game, why choose someone with no chance? A peasent would be just as likely to be a 4yr old kid, baby, pregnant woman, or elderly man with a cane.

Over the course of history, many technologically/militarily advanced civilizations have lost again what might be termed as peasant armies. However, these "peasant armies" tend to be the simply the masses of people living in an area using guerilla tactics. A city would not conscript these people, but they would fight simply to protect their lives and families from what they perceive as invaders. In most cases, the invaders either make simple mistakes in tactics or just underestimate the desire of the inhabitants to welcome their advances. To use the Taliban as an example, their forces may be illiterate peasants, but they are also seasoned in the art of guerilla warfare against heavily armored opponents. One must remember that the Taliban is made up of people who successfully fought off the Russian invasion in the 80s and who remember well how the US helped them to fight the Russians, only to leave their nation devastated after the Russian retreat.

These aren't peasents, they're religous-fanatic brainwashed, *guerilla* fighters. The Taliban & Al qaida (sp?) have training camps.

You're severly underestimating the capabilities of trained soldiers/knights if you think one farmer with a fork could take on two trained knights. Did you see that movie Troy? That part where Achillies kills the big grunt soldier with just a quick charge, shift, and thrust. He was able to kill this trained veteran (who was the best solider of the opposing side) in just the length of time it took him to charge the open field which was probably about 100yards before the other guy could even lift the sword past his waist. Now sure, history calls Achillies the greatest human warrior who ever lived (and is probably right), but the point is: this is roughly the same difference between any trained knight/soldier over an untrained peasent. The knight would kill the untrained peasent in about 5 seconds, before the peasant could even lift his pitchfork.

From my own personal experience: I made the mistake of calling a challenge to a friendly wrestling match against one of my new roomates (I had just arrived at my new post, this was about 20 something years ago when I was in the military, this was like one of those newbie initiation things). I was about the same size, weight, strength, etc as my new roomate. I over-estimated myself & figured I could beat my roomate, except I knew nothing about wrestling. My roomate on the other hand was on the wrestling team in his HS. Needless to say, he pinned me in about 5seconds, literally, just like the Achillies example. As a matter of fact, now that I'm thinking about it, I think that guy was one of the state champs from his state for his weight class.

... Don't underestimate troops, if a peasant goes up against trainied soldiers/knights without training (and without any wierd circumstantial variables) he'll loose, and it'll be a massacre. History supports this.

Really, just think about this intuitively, outside the military context. How much quicker would a baker be able to bake a cake versus an untrained person in the kitchen who has to figure it out first? I say 99.99999% of the time the baker will bake the cake faster outside of any wierd variables. How much faster will an auto mechanic fix that broken car than someone who knows nothing about cars? I say 99.9999% of the time the auto mechanic will fix it faster than the untrained peasent who knows nothing about cars. How much faster & better will the software engineer's computer program be than the untrained peasent farmer who's never seen a computer? I say 99.9999% of the time the software engineer will have a better program done faster. Now let's turn the shoe a little, how much better & faster will that highly trained peasant farmer be able to harvest his crops of potatoes/corn/barely than the software engineer, baker, or knight? I say 99.9999% of the time the farmer will do a better and more efficient/faster job of farming than the engineer/knight/baker.   

True, but it's not the *mass* of people who actually fight in these cases. It's a *subset* who actually become partisans/guerillas. This partisan subset either already has appropriate training (i.e. hunters, ex military, current military) or provides some rudimentary training to the other members of the subset who are otherwise capable of fighting [i.e. those able to shoot, move, fire, conceal, etc, which would moslty be younger(ish) people but not always]. Sure, sometimes partisans move entire families to safety to escape a scourge, like the Russians did in WWII, but these *entire* peasent families did not do the resistance fighting once relocated, just a subset who had 1) some rudimentary guerilla/weapon training as mentioned above and 2) armed (either from retreating red army troops, captured german weapons, or some other supply source like an airdrop). The mass of people did not actually fight, although the masses surely supported the fight in one way or another. Russia (and the other federated states in WWII) is a highly populated country, having roughly as many people as the US. In WWII they had partisans numbering the millions (out of tens of millions peasants in the occupied areas) in 44-45. 

One of the only real examples of a true Peasant War (the actual name of the war), was called the German Peasent's War, and it was a massacre, as in the trained troops massacring the peasents ... should've been called "The peasant's massacre"

Sometime's history books use the term "citizen's army" (US army in WWII is referred as such) or "Peasent army" (similar as the Red Army in WWII), but it's misleading. It's true that that untrained citizens were drafted and initially had no training & no weapon (technically a "peasant" by definition), but they were then put through basic training and armed. At that point, they're no longer "peasents" but full-fledged soldiers. Sometimes, nations draft without training, these organized & armed but non or ill-trained soldiers are better termed militia or conscripts, and they should be appropriately modified in game terms (lower morale, lower attack, etc). The exact same principles apply to tribes, except it's usually a difference of technology (stone age vs modern .. some tribes have "veteran" soldiers, most are usually "guerilla" soldiers).

If you're comparing one guy that *knows* how to fight (like a boxer) vs 10 people that don't know how to fight: I'll put my money on the boxer every single time. That boxer will throw ten punches and ten peasants will be knocked out out the ground (and probably under 30 seconds), two at a time with left and right jabs/hooks. That is, if the peasents are fool enough to take on the boxer (i.e. if their morale is high enough for whatever external reason, or if they're suffering from temporary insanity, or drunk & stupid, or high). The reality is, the boxer would probably only need to knock out a couple of the foolhardy peasants before the others hightail it & run away ... otherwise 10 peasents will be lying on the ground knocked out and the boxer will be left standing.

If you're talking about *anyone* fighting, as in a peasent vs peasent (un-trained participants) fight, barroom brawl, or catfight, then sure you're absolutely 100% correct. But, the topic of the OP is "knights vs peasents," that would be synonomous with a "trained fighter" vs untrained civilian (or "peasent").

There's a *huge* difference between trained & untrained (anything), and not just military/fighting. Take a stock car racer vs a taxi driver who never raced and put them both in the Indy 500, the stock car driver will run laps around the cabbie every single time even if the cabbie could avoid killing himself in a wreck.