I like grand battles, but I find the Pendragon battle system a little clunky. To simplify it for the players and myself, I've come up with something I would like to implement, but would appreciate feedback on before doing so:

I will prepare the enemy army composition beforehand in its entirety. Larger armies will have more dangerous units, whereas raiding bands are unlikely to have giants (plural). There could be a lonesome rampaging giant that counts as a raiding party though, or something like that. Modifiers to leaders will be implemented as per the normal rules (home turf, superior troops, highly motivated etc.)

Army Leader Battle rolls
Friend/Enemy
Success/Success = fight normally
Success/Fail = +5/-5 to all Battalion Battle Rolls
Fail/Success = -5/+5 to all Battalion Battle Rolls
Fail/Fail = -5/-5 (?) to all Battalion Battle Rolls
Crit = Special Event
- A Battalion breaks through, +5 to subsequent Army Battle Rolls. Second time opposing army breaks
- Special target exposed (Army Leader or standard bearer)
- Prisoners
Fumble = Special Event
- Army breaks, flees
- Special target exposed (Army Leader or standard bearer)
- Taken prisoner

Battalion Leader Battle rolls
Friend/Enemy
Success/Success = fight normally
Success/Fail = +5/-5 to all Unit Battle Rolls
Fail/Success = -5/+5 to all Unit Battle Rolls
Fail/Fail = -5/-5 (?) to all Unit Battle Rolls
Crit = Special Event
- Battalion breaks through, +5 to subsequent Army Battle Rolls. Second time, opposing army breaks
- Special target exposed (Battalion Leader or standard bearer)
- Prisoners
Fumble = Special Event
- Battalion breaks, -5 to subsequent Army Battle Rolls. Second time, army breaks
- Special target exposed (Battalion Leader or standard bearer)
- Taken prisoner

Unit Leaders Battle rolls (also used in Skirmishes of under 15 knights or so)
Friend/Enemy
Success/Success = fight normally
Success/Fail = +5/-5
Fail/Success = -5/+5
Fail/Fail = -5/-5 (?)
Crit = Special Event
- Flank charge (+10/-10)
- Special target exposed (Unit leader)
- Prisoners
Fumble = Special event
- Flank charged (-10/+10)
- Unit flees
- Special target exposed (Unit leader)
- Taken prisoner

I also have implemented a rule, that if all the player knights are killed/taken captive, their army loses. My reasoning for these rules are, that it makes the Battle skill important across the various levels of leadership while keeping things managable enough for everyone to unerstand what is going on and why. The only part I am not entirely happy with is the Army Leader level, though I can't really put my finger on why. It just seems a little lackluster.

EDIT: It occurs to me that the order in which these are rolled was not logical, so I changed it so that the Army Leader roll is first (which is also rolled first)

Looks reasonable enough, but while you have a losing clause (all PKs dead or captured), I don't see any winning clause? Or any way for the PKs actions to matter (which I can understand in a big battle, admittedly, unless they do something truly remarkable like defeat an enemy battalion commander).

I came up with my own version way back when that you might find interesting:
http://nocturnalmediaforum.com/iecarus/forum/showthread.php?1919-Variant-battle-system&p=16334&viewfull=1#post16334

Looks reasonable enough, but while you have a losing clause (all PKs dead or captured), I don't see any winning clause? Or any way for the PKs actions to matter (which I can understand in a big battle, admittedly, unless they do something truly remarkable like defeat an enemy battalion commander).

I came up with my own version way back when that you might find interesting:
http://nocturnalmediaforum.com/iecarus/forum/showthread.php?1919-Variant-battle-system&p=16334&viewfull=1#post16334

Right, that should probably have been mentioned. Should the PKs live (and none of the other lose conditions take place), the side with most Battalions intact and present wins, if both armies are still intact, the battle ends in a draw (it happens). The Army Leader is also the central Battalion Leader. It should also be noted, that in the Army and Battalion leader Critical roll results where it says "enemy", it should say "opposing". The actions of the PKs (unless they find themselves in a situation where they get to attack an enemy VIP or unless I decide to write a set piece in the middle of the battle for them to seize) don't really matter in the grand scheme of things unless they are battalion leaders or similar.

I'll have to read through your system in more detail, but at first glance it is a bit confusing. What are the +1s, +2s, etc.?

I'll have to read through your system in more detail, but at first glance it is a bit confusing. What are the +1s, +2s, etc.?

One of the dangers of writing something for one's own use and then just throwing it up on the Forum is that some things are left inadequately explained.

The bonuses and minuses in the Army Commander's and Battalion Commanders' rolls are the Army Situation modifiers, a way to keep track on how battle is going. The number applies to the PK's side, and the opposite number applies to the opponent's side, on the Battle rolls. I see that in the thread I was advocating waiting until +-5 to give it as a bonus, but later I suggest giving it straight away. Either way would work, waiting for +-5 would make it more even, while +-X would let the PKs to start reaping the benefits of their actions sooner.

Note that this was meant for a relatively small army, with a PK at the leadership seat. If it is a bigger battle, I could see simply increasing the limit to -10 for retreat and -20 for rout (matching Book of Battle quite closely in that regard), and halve the bonuses. Why do it this way? Well, it is a big battle with bigger armies, so there will be less sudden reversals and more reserves to help to keep the fight going. So it is less likely to be decided on one roll.

Riight, now I understand, it's an excel transfer? It just looks a bit...wonky.

I can see the appeal of that kind of system, particularly in larger battles. I might do a bit of a change though, namely that the Unit Leader Roll would be from my system. Let's see if I understood your system:
Initially the Army Situation is +-0
Step 1: You add situational modifiers to this (Superior troops, home turf, disheartened, outnumbered...)
Step 2: Army leaders roll opposed Battle skills with the Army Situation modifiers. So if two generals with Battle 15 are facing off, but one is fighting on their home turf the skills would be 20/10. Look at the table to see who (if anyone) gets a bonus to their Army Situation.
Step 3: Split the battle into three separate "mini-battles" for the Battalions and keep track of the "Battalion Situation" separately. (This is where I got a little lost, does the old Army Situation carry over?). Roll Battalion commander opposed Battle rolls, look at the table to see who (if anyone) gets a bonus to their Battalion Situation.
Step 4: PKs roll Battle and get to pick an opponent if successful or fight the enemy commander on a crit. If most succeed in their combat rolls, the situation changes +1/-1 and if most fail -1/+1.
Step 5: Check Battalion Situation, if at -5 the commander attempts the Glorious Duel, if at -10 the Battalion routs and the surviving Battalions get -3 to their situation.

If I understood this correctly, I see an issue that with only a few lucky rolls one army will rout (or at least flee) in one or two rounds. If I didn't understand correctly, please correct me.

If I understood this correctly, I see an issue that with only a few lucky rolls one army will rout (or at least flee) in one or two rounds. If I didn't understand correctly, please correct me.

Like said, it was designed for small, quick battles, between a skirmish and a real battle. Once things start to go bad, they will likely go really bad, since all the troops are committed already. So yeah, that is more of a feature than a flaw in that sense. Although I will be honest and say that I didn't get the chance to test it out.

For instance, I could see making the army commander's roll follow the same table as the individual battalion commanders, which means that success-fail + success-fail would 'only' result in +4/-4 rather than +5/-5. So you'd at least require that the Army Commander and the Battalion Commander and the PKs are all winning. Or even tweaking it so that success/partial success is still a draw, and success-failure is +1/-1 instead of +2/-2, but this might make it drag a bit more (maybe limit that to the Battalion commander). Best way to find out what modifiers work would be to run a few test battles and see if the results are reasonable. I am less worried about the absolute outliers like a crit-fumble. That is such a fluke (1 in 400) that it will end in history books as a major military blunder on the part of the defeated. :)