One issue I've heard is man-rating.
Manrating is a very stretchable and ultimately meaningless term. It originally came to life when ICBMs were used as the main means to get people into space. They were then "manrated" depending on various criteria. Space X has built its hardware from the beginning with manned flight in mind. So they should be able to meet all critieria.
The EELVs should also be easily "manrateable" since they do have a rather good flight record by now.
In contrast, the new 5 segment boosters needed for AresI and V are VERY different from the shuttle boosters. But noone seems to have a problem with them being "manrated"? Heck even the people at NASA are not sure anymore what exactly they should call manrated.
So I would be careful with any speculation on that matter.
One reason was that they wanted to use space shuttle engines, since they were already cleared for manned flight, meaning you just have to do the tests on the rocket body.
Well actually because they are among the most powerful engines available and they are still being produced and we still have the tooling for them. All that makes for a quicker and cheaper process. It makes sense to me. They are very powerful and good engines, though a little on the pricy side (that might go down, if they are mass produced though).
They also wanted to just used a modified shuttle tank since the tooling was there, rather than build a new body.
That would be DIRECT then, more than Ares, I guess. It is an argument though, that is true. But then the question is, whether it is still not cheaper to just simply let the market decide that?
Fixed price contracts always suck. Yes they make a lot of jobs (both conservative and dem senators are working the system to keep jobs in their states that way), but they ultimately result in a loss of money.