Egypt VS Jerusalem worse units win?

I'm currently playing the crusades campagin of kingdoms. Im playing as Jerusalem but i keep fighting really bitter battles with egypt when i should be winning.

It seemed like my units would lose 1 v1, so i did an experiment in custom battle

1 unit of Eygpt dismounted Ghulams Attack: 9 Defense: 12 charge: 2
VS
1 unit of Jerusalem Templar sergeants Attack: 9 Defense: 16 charge: 4

and the Ghulams always win.

Then i try Knights of Jerusalem Cavalry (which always seem to die in any battle despite being a good unit) attack: 8 Defense: 15 Charge: 5
VS
1 unit of Mamluk Archers Attack: 7 Defense: 12 Charge: 3

And my Knights lose everytime to light missile cavalry. :help:

I may be giving a dumb reply, but does the fact that the battle takes place in the desert make any difference? I know that this has made a difference in other games.

First off, when we old dogs told CA that something wasn’t working in experiments with the Custom Battle feature in other TW games, we were eventually told that there was just something wrong with the Custom Battle feature, but the campaign game was fine. So I will skip over what may or may not be happening in the Custom Battles you mention and get to the real problem.
When you say that you “should be winning”, do you mean that you are doing everything right, but you still lose? As A Nerd posted, some units do better in the desert than knights in shining solar collectors do. Exhausted troops don’t fight well. Experience and morale are also factors. The generals can have a big effect if played right. Terrain can also be a key to your victory. Do you just go toe-to-toe with Egypt or do you use tactical skill to chain rout their armies? The computer has an algorithm that compares its unit’s numbers to yours. You as the human commander should put the other factors together.

I agree with the above. Try experimenting with the units you mentioned on a flat european map and see how it goes. You can also try with different terrain levels to note moral and fatigue. Try playing with more attention given to your general as well. Experiment with flanking and the like, then return to a desert map and see how it goes.

for some reason, it wasn't to be blamed on map bonus alone:
Case 1 (Dism Ghulams vs Templar Sergeants)
1) Dism Ghulams use mace, which had Armour piercing attributes, effectively negate the sergeants' armour in half
2) Templar sergeant's weaponary got "spear" attribute, means they will suffer -4 to attack against infantry

Case 2 KOJ vs Mamluk Archers (minor issue, Mamluk HA is not light missile cav, they are actually heavy missile cav)
1) Put aside the arrow related casualities, which KOJ had no effective counter, Mamluk archers had mace, and hence AP attribute. Which set armour of the KOJ in hal

Nice observation! I often forget about the stats for they tend to confuse me. Definetly worth looking into, though I think they can be more effectively applied with map variants. Just my opinion though. :)

Don't forget the difficulty level...

Difficulty is definetly good, but I think the best tests can be made when the AI receives no bonuses.

actually in M2TW, harder battle difficulty don't meant that AI got as ridiculous bonus as RTW get, I think they only got better morale...

and yeah, in RTW, battle difficuly should be kept on medium if you don't want to see some strange imbalancement, Hard means AI got +4 attack, +4 Defense, and +3 Morale, while vh got +7 attack, +7 defense, and +5 morale