I don't have a problem with 'severe ambiguity' or with a method of approximation being 'wronger.' The question, as I understood it, was, 'how high into the air could you launch an electromagnet with a lightning strike?' I seriously doubt the poser of the question was looking for more than an order-of-magnitude calculation.
Turin, I find quibbling over the validity of arguments based on the conservation of energy and efficiency coefficients to be quite pointless. I certainly wouldn't embark on a more sophisticated analysis of the scenario without first determining the maximum theoretical amount of energy that could be derived from a lightning strike, and the maximum altitude I would expect as a result of this. Briefly considering many of the issues you mentioned and more--loss from the capacitor, less energetic lightning strikes due to low altitude, poor coupling between the electromagnets as distance increases (most of the acceleration would have to happen in the first few meters), and especially losses due to air resistance and the inherent instability involved in balancing one magnet atop the B-field of another--I felt that one percent is a very reasonable scaling coefficient. That a realistic altitude is on the order of single kilometers should suffice for most purposes: it tells you that this would not be a very good method for earth-to-orbit launches, even if it were highly refined and improved, but also that the electromagnets will certainly not stand still when the strike occurs.
I am more interested in your point that the high-inductance nature of the system is anathema to the lightning; even if a lightning rod attached to the hot end of the assembly, it seems likely that it would sooner make a second arc through the air than travel through such an arduous path. Perhaps this is why schemes to 'harness the power' of lightning strikes are doomed to fail; the energy levels are high, but the system seems unwilling to apply itself to anything save reaching ground and being converted to heat.
P