Here's something simplistic but reasonably accurate (I think). Right now with GPU/FPGA you can get around 0.0009 Th (=0.9 Gh) for $600 = 0.0000015 Th/$ (or 1.5 Mh/$) (very rough - I'm thinking of the BFL single, but GPU is reasonably comparable). The SC MR is 1 Th for $30,000 = 0.00003 Th/$. So, very roughly speaking, BFL ASIC is 20x better in terms of hashrate/$.
Ignoring power cost (which varies widely anyway), BFL ASIC mining will be more profitable than mining pre-ASIC at least until the post-ASIC difficulty reaches 20x pre-ASIC. After that, it will still be better depending on your power cost. How long to ROI? That's much harder to estimate, but assuming you could pay off $600 invested in 0.0009 Th in about 200 days pre-ASIC, BFL ASIC will do better than that at least until post-ASIC difficulty reaches 20x pre-ASIC.
So, to boil it down to something very simple but not totally off-base, investing in BFL ASIC will be a better investment than what is currently available so long as you get your ASIC equipment quickly enough that the difficulty doesn't increase 20x over what it is now within 200 days after you get your equipment. Of course, you could still pay off the investment, even within 200 days, if difficulty increases faster than this.
Let's use these numbers and assume that the same amount of $ gets invested in BFL ASIC as is invested currently in the entire network. The current network is 12.5 Th. That divided by 0.0000015 Th/$ is $8.3 million. $8.3 million worth of BFL ASIC would be 250 Th. 250 Th is exactly 20x the current network speed (this shouldn't be surprising). So, basically, BFL ASIC will be more profitable than what we've had pre-ASIC recently until miners pour more money into BFL equipment than they have so far building the entire current network (and assuming that BFL or a competitor doesn't drop prices before this happens).
I'm skeptical that this is going to happen very quickly. Some people (bitlane for example) are, I think, investing a good deal more in ASIC than they had in pre-ASIC, but with so many others taking a "wait and see" or "I refuse to do business with BFL" or "I'm out for good" stance, I think it's going to be a while before miners collectively throw $8.3 million BFL's way. At the very least, I think people getting their ASIC equipment early in the game will get it paid off OK, and have a decent run after that. Also, there's the point that it will take BFL a while to produce 250 Th of gear.
I know this doesn't directly answer the OP's original question, but I fail to see how it's helpful to know what you would earn with an SC MR right now, if, per impossible, you had one. By the time you get one, the difficulty will already be rising dramatically. So, I think it's more interesting to ask what needs to happen before you need to start worrying about an ASIC investment getting less profitable than what's currently available. And the basic answer is, miners need to throw as much $ at BFL as they have so far building the entire network, and then all that money needs to be converted to shipped units actually hashing. The question you need to ask yourself is, how long do you think it will take for that to happen?
I swear I'm just going to refer the 90/unconfirmed questions about ASIC this or ASIC that to your post...