As for concentrating on one enemy until it goes down, that is an extremely unrealistic tactic that only works in D&D because peculiarities of the rules. No savvy tactician would ever do that in real life -- it is a certain path to defeat.
In all seriousness, is this true?
I want to use a bar fight as a test case:
So there is the equivalent of a grand melee going on in said bar and we are going to focus on 2 combatants - both are young adult males of equivalent build, neither of whom has any real skill in martial arts, boxing, wrestling, or any other form of unarmed combat. Both are partially drunk (important because they will feel less pain). So they are going at it, and one gets in a lucky shot - lucky in that it lands and actually does some real damage - let's say breaks a rib. So we will say A has injured B. Now B, being drunk gets even angrier (not being rational due to the alcohol and feeling reduced pain again because of the alcohol). Let's say combatant C enters the fray - to help the injured B. A now faces both B and C, with B badly injured (let's say he has a punctured lung as a result). I realize that A may act very differently than what I am about to describe, but I am looking for the most rational tactics, not what is most realistic. It seems to me that the most rational tactic would be to reduce the number of opponents as 2v1 is going to be more difficult than 1v1 (assume C has same combat expertise as A or B). The reason why you reduce the number of combatants is that the more combatants there are the more "actions" they get to do in a given amount of time relative to you. This is why when you see any sort of training exercises for any sort of military or police, that teamwork is so important. One person is incredibly vulnerable. But a team can function together to do multiple things at once.
So back to the example - it seems the most rational course of action would be to drop B. Then A can focus on C.
The problems with this example are as follows:
B, as he is injured, is funcitoning at reduced capacity and function, his punches are probably swinging slower, his flexibility is reduced because of the broken rib, etc. Thus, A might want to focus on C as he is at 100%. Howver, it B is still fighting, doesn't it make sense to get rid of him as quickly as possible. The reason why he would not do this is that he opens himself to C. However, if he focuses on C he does the same for B.
The problem with D&D seems more to be that the HP system doesn't take into account reduced funcitoning. However, regardless, I would think one would still get rid of the weaker person first.
I do agree that the bad guys aren't go to let up on a wounded PC merely because he is wounded. That's where friends come in handy. If the wounded PC fights defensively and steps back 5', he looks less dangerous. If a friendly PC then steps up can do some damage, that fresh PC looks much more threatening. Guess who will get targetted?
It too often seems that the guy who fights defensively - is still fighting. I would presume that the giant would interpret defensive fighting as "good, I got him" rather than oh, let's focus on someone else.
Again, I am seriously wondering because I am planning a big fight involving a bunch of giants and want to make sure that I play them tactically sound for their intelligence.
(2) It puts the DM in the position of having to decide which PC gets to die this adventure. Won't that be fun?
I agree that this is quite nasty. As I said, I have been nice in the past - but really that doesn't make sense - at least with the system we use in 3E. Also, you see the players focusing on one opponent until they go down or surrender. It will make the players act more consciously of the fact that there PCs can die though and will make retreats occur (where as they probably would not occur at all before). At least this is what I have experienced in my games - and anyone who knows my style either by having read the Undermountain thread in story hour, or by previous posts - knows that I pose tough challenges to my players.