yeah because combat was based on random numbers and you can have 99% change and till lose.. Or it was more like who atacks first. if you both have trebuches just sucide trebuches and kill their army. To be honest combat system wasnt that great.
It wasn't that random. 99% means that if you take that fight 100 times, the most expected outcome is 99 wins and 1 loss. You can still lose even 3 units or lose none, but it is almost impossible you'll lose 40 times. That's how odds work.
It is funny how noone complains when they win 3 70% fight in a row (34% chance for that to happen).
Also, although your first unit may die at 99% percent, but the remaining defender will most likely be almost dead and you'll only need to mop it up. That is why you bring more troops into an offense.
Ok, there is one case where game trolls you: it can't really calculate odds when first strikes are included in the mix. It underestimates them. Since all Archery units have first strikes, they are all actually a little bit more powerful.
Anyway, playing both games, I am sure Civ4 is more tactical. Your placement is much more important as units are also slower and can't be all around the battlefield. Also, it is more realistic to have massive army. I mean Russians literally bumped into stronger (and more expensive) german tanks at full speed to roll them over, and did so successfully. Mass of army means a lot in real life. And when you have masses, you cannot lose because you had some unlikely losses. How come people who are really good at the game never lose games because of luck factor?
In Civ5 you have Super Artillery to win you the game, relatively strong unit in front to defend it, and one mounted unit to reveal you the terrain as super promoted Arty has ridiculous range and damage. I usually had 3 and they chewed Deity AIs for breakfast.