I think SS and all mods are limited in how they can portray certain factors of combat, tactical and strategic, by the M2TW engine.
Firstly, the majority of casualties in medieval battles happened in the withdrawal or rout, not in the combat. Can we represent this? Creative Assembly tried this with vanilla. The result: troops rout extremely easily and cavalry capture lots of troops. However this lead to extremely short battles with no difficulty.
The consensus is that undisciplined infantry would rout if approached by a cavalry charge. However this is difficult to portray with the morale mechanics of M2TW because it makes battles too easy. While disciplined troops would be able to withstand a cavalry charge, the mechanics that allow this in game also cause the charging cavalry to take too many casualties, ie :spear_bonus. However, it wasn't until the rise of pike and shot (pikes, polearms, and gunpowder weapons) that cavalry began to lose its importance and was nuetralized on a massive scale. I think RR/RC demonstrates this in SS.
Another example that demonstrates the limits of the engine is that in the Battle of Agincourt, a battle that demonstrated the abilities of the longbow, a stock of 400,000 arrows was brought to the battle and most were loosed at the armored knights. The result is that less than 2000 knights died from arrows while the vast majority died at the infantry line killed by English men at arms, not archers. We can also discount the majority of modern arrow tests because virtually none of them consist of real battle scenarios, and contain many discrepancies with armor in use in their respective periods.
In primary sources from the 14th century and 15th century, the authors speak of the damage caused by arrows to the horses of the cavalry not the knights themselves. The arrow fire was useful in disrupting the formation of the cavalry, breaking up a charge, not killing them en masse. This effect is not able to be reproduced in M2TW because the knight and horse die together.
When we hear of the Mongols battle successes it wasn't the horse archers shooting all their ammo, whittling the enemies down to a half strength and then attacking, as in M2TW, that gave them success. It was their combined arms and communication along with quality troops that allowed them to succeed. Horse archers harass the enemy, causing minor casualties, lowering morale, and disorganizing the enemy formation leave them for slaughter by cavalry. This is somewhat recreated but Horse archers and archers cause too many causualties to armored men (heavy infantry +heavy cavalry).
Armor and its manueverability is one of the greatest debates amongst lay men but if you look at history and consensus of historians it shows that armor was well balanced and including 16th century plate armor weighed less than what the modern soldier carries. At the 35:00 minute mark of the attached video is a man in gothic armor that shows a man running at a full sprint without a problem. Later it shows 2 men dueling with polearms and they have great visibilty through their visors. Plate armour for knights and men at arms was tailor made which meant that it was balanced and fit for a specific person. So can we finally dispel the issue that knights couldn't get up once they fell and that they couldn't see around them.
http://youtu.be/NqC_squo6X4?t=35m
Finally, The main reasons for victory or defeat in battle is tactics. This has been proven again and again throughout history. Terrain, tactical movements, and experience all can cover up an army with inferior troops or armor than an enemy. This infers that a human player should be able to defeat the AI even with a significantly inferior army, which is the case. The human is always smarter than the AI. The first time I played SS with RBAI I was stunned at the AI's ability to use tactics and was mildy afraid that battles would be too hard if the enemy had superior troops or more cavalry than I did. However, after sometime I developed counters to the AI and this should be true for every player, and as you play even when something unexpected happens you should be able to assess the situation and determine the best course of action.
In closing, while the M2TW combat engine is flawed, RR/RC is the best combination of compromise and realism that is available on the forums that tempers the player's attack without disarming the AI army's ability to compete with you. I do believe that cavalry unit size and availability needs to be reduced, but this is already being implemented in RC 2.0.
Debate is welcomed.