Tyll is a good guy, but before you go and read this and start screaming the sky is falling, note what Tyll summarizes:
Summary Did I show break-in exists? No. There are too many variables still. Was it simply movement? I don't know. If I did it again to another brand new pair would I get the same results? I don't know. If I did it to an already broken in pair would I get the same results? I don't know.
This is not evidence of break in / burn in. There are way too many variables that he cannot account for, and since he's reasonable about this, he's summarized as such. To improve upon his experiment, he should take the burn in at 20~40 hours and measure it 100 times separately, once per day, allowing all equipment the same cool off time and pre-warm up time, in a facility with the same exact ambient temperature and humidity every time. The problem is, you would want to do this to more than one pair of headphones, and you'd not be able to accurately measure if it's the headphone being different from the other headphone (micro differences in build quality, fitting, production), or if it's in fact "settling" as time goes on more and more.
There currently is no evidence of burn in, other than attempts at measuring it with experiments that are simply not elaborate and accurate enough, and of course done so with many golden ear subjective trials for those that listen and make the claim based solely on their ears.
This is not to be negative. It's simply to point out that the more you try to put emphasis on this ritual the more silly it gets. We're not going to be putting a pair of $300 headphones into the Large Hadron Collider to measure minute changes over instants in time while playing back some pink noise test tones.