Well if you look at the 21.0 tests, and add three db across the board, you get an idea how the 18.2 would look as they share the same motor structure and amplification(per driver) and proportionally more airspace. So it should be nearly matched for long term output under ~30hz, with similar distortion at max output, lower distortion at just below max as already the 21.0 is very close in distortion given the same output level(115db sweep for 21.0 compared to 105db sweep for S2) under 30hz when both are down a couple sweeps below max. Then it walks away for output over 30hz and distortion over 50hz. Not really a fair comparison though considering the cost. The S2, and proportionally the S1 shows awesome performance for the money under 40hz, the 18.2 shows what it takes to match that performance under 40hz at nearly half the size. Massively more power and cost, is what it takes to push the laws of physics hard up to the wall.
Comparing to the 18.0 is a little more fair as they are closer in box size and power per driver, although the 18.0 still costs more. Considering the S1 is said to be basically half the S2, so -5-6db across the board. Factoring that in the S1 and 18.0 look near equal 30hz down for long term output with the 18.0 having a clear advantage over 30hz, as it compares directly to the S2 40hz up. Burst output is very close across the board to the S1, each having a little more than the other in some areas, as well as distortion is comparable 30hz down. The advantage the 18.0 has is over 50hz where distortion and power compression are much lower, lower even than the S2 compared directly at similar outputs in that range.