If you've calibrated it properly and it still has such awful numbers, take it in and get it replaced. It shouldn't be that bad, not even close.
My 2011 Air has now 703 cycles (I am a heavy mobile user) and still 88% (and 4-5 hours in real-world use).
----------
You have no facts to back up that statement.
That is false. The threshold is 80%, not 85%, and the cycle limit is 1000, not 250.
On the one hand, I am inclined to agree with you because battery decline is not linear, but rather an exponential acceleration characterized by a minor decline over a large number of cycles followed by a punctuated decline rapidly thereafter.
However, I do contest your numbers interpretation of numbers. Apple's own numbers report 80% health after 1000 cycles, as you mention, which is what the poster you quote and say have no facts to back up is basing that off of. They assumed linearity and that it would be 90% at 500 cycles.
Given my own anecdotal experience with dozens (probably hundreds?) of Macs over the years, and my knowledge of how these batteries perform, I would posit that you can reasonably expect MORE than 90% health after 500 cycles, as the decline again is not linear, but rather sigmoidal if you can imagine the curve.
In either case, the OP's battery is atypical and problematic. If the calibration, which as you correctly say, does not affect performance but rather reporting of #s, does not change the information, it clearly is defective and should be replaced. If you are firm and clearly unsatisfied with the life, Apple will replace it. You just have to be firm on your demands, but of course polite and reasonable. Courtesy has never failed me with dealing with people.
Edit: Also, nice FAQ by the way. Very definitive, but that's also probably why people don't want to read it.
----------
Also, to everyone who posted about this, do not be concerned about minor fluctuations down from 100% while plugged in and charged. It's just Apple's way of doing automatic power conditioning.