re: avg damage vs just hitting: I am in the camp that would say average damage is better than 'just hitting' as a way to figure overall effectiveness. Everything else seems to balance itself out: sometimes a high-damage weapon will 'waste' damage on someone who's almost dead (or had low HP to begin with), sometimes the arrow can't make it through DR, sometimes 'just hitting' is better than doing more damage, esp if the target is just about dead. In the long run, the better here/worse heres I think tend to nullify each other and avg damage capaicity is better off.
In the case of statistics, it's actually rather interesting:
On a roll of 2 d 20, you have 400 possibilities. To determine what the % chances, it's simple:
Assume 11+ to hit.
There are 200 cases where 11+ will come up on one die, but not the other (ie, when die 1 rolls 11, there are 10 cases (1, 2, 3, 4 ... 9, 10) where die 2 also doesn't roll an 11+, when die 1 rolls 12, there are 10 cases (1 ... 10) where die 2 doesn't roll 11+, etc).
There are 100 cases where 11+ will occur on both dice (ie, when die 1 rolls 11, there are 10 cases (11, 12, 13 ... 20) where die 2 has also rolled over 11, etc).
There are 100 cases where neither die will roll over 11 (ie, die 1 rolls 1, there are 10 cases where die 2 also rolls less than 11 (1, 2, ... 10), etc).
You can pick any number needed to hit and work out the results.
So, this works out to:
200/400 = 50% chance 1 hitting
100/400 = 25% chance 2 hitting
100/400 = 25% chance 0 hitting
Now, this is a simple case, where the chance to hit is 50%. But one can work it out for any chance to hit... say, for example needing a 17+ to hit (pretty tough!), the rapid firing dude would have: 72% no hits, 26% one hit, and 2% two hits.
A full analysis could be done to see how that compares to a single wielder with a single, standard attack. 15+ = 30% chance to hit, 3.5+3.5+6 = 3.9 average damage.
BUT, if we do that simple, we have NEGLECTED the fact that the single wielder may have taken some feats, A) giving him better chance to hit and B) possiblitity for even more damage. We can make the single wielder ON PAR with chance to hit, but with a +3 to damage due to Weapon Focus and Power Attack, for example. That's a much more valid comparison. And in that case, pure damage wise, the greatsword wielder comes out ahead, (at least at level 3+ when he can use 3 points in power attack }

.
Are archers useful? Yes! Are they powerful! Sure... are they OVER powered? I'm not leaning that way at this time.
At least from a pure, non-magical point of vue.
I DO agree, however, that GMW's 50 arrows/time could be what is broken in this case, and/or the stacking of magic bow/arrow. As green slime said, if someone wants to spend a tonne of GP on getting magic arrows, then stacking is fine... when its sooo easy with GMW, then not so fine.
Assuming your party cleric is willing to cast GMW, of course }
Kannik