He showed that even though individual samples will vary somewhat due to the manufacturing process, this causes some of the T/S parameters to vary in such a way that in the end, the effects cancel out. In a given box, ported the same way, the woofers of a given model will perform very similarly even though Vas, Fs, etc may vary slightly.

He stated that what really mattered were the ratios of various T/S parameters. And even though T/S numbers varied from sample to sample, they varied in such a way that these ratios were the same.