The problem with "burn in" is that if a product genuinely changes, it opens up a lot of liability for the manufacturer.
If a headphone has a one-year warranty, then someone could claim that the product is defective because it changed and make a warranty claim.
Management, legal and accounting would be all over engineering if products were actually changing with use. Further, products are tested for thousands of hours before they go on the market. Management reviews the tests. If a major change occurred, they'd make engineering fix it.
Even if they found that X number of hours were necessary for a particular design, they'd burn them in at the factory, just to make sure it didn't change further and potentially incur warranty costs.
Burn in is magical thinking. People love to think they're doing something special that makes their experience unique.
If you really want to do something that changes your listening experience and learn more, go get a soldering iron and build a CMoy. You'll actually learn something instead of performing an audiophile folk ritual. After the CMoy, apply what you learn to building amps, DACs, or any number of cool projects. Those will improve your listening experience. A ceremony won't.
The thing is that the change isn't heard by the naked ear when people just listen to the headphones for enjoyment. It's not something that you listen for if you don't know it's there (many consumers don't even know about this). They won't hear it unless you listen for it.
Also, there has to be a reason why so many people are hearing this? You can't just say they are all magically thinking or eating placebos, it just doesn't work that way. If a huge group of people experience a phenomenon, then there has to be some truth behind it, especially when more than one person report back with similar results, if not the same. Look in the HF5 thread (in praise of HF5) to see another person with similar results.
There is no doubt that sound changes in headphones, it's been tested with equipment, CNet did an article on it with frequency response curves from another source (they used Grados to test with). Note that a small change in dB is actually a big change sonically since dB is listed in a logrithmic graph when the actual pressure level is linearly measured. So a 20dB drop from 20 to 0 does not equal a 20dB drop from 80 to 60. Math as follows:
dB to sound pressure:
20dB = 10^2 Pa; 0 dB = 10^0 Pa = 1; 100 - 1 = 99 Pa change
80dB = 10^8 Pa; 60 dB = 10^6 Pa; 10^8 - 10^6 = 99000000 Pa = 99000 kPa
The change is huge in comparison.
As long as it is measured the same way both times, and they show change, we know it exists.
Now we have to figure out why it exists, and why there are conflicting results for it. Saying that burn-in is a magical thinking or an audiophile folk ritual doesn't say anything without facts to back it up. Just because you didn't hear it, doesn't mean you can put down the people who did. You have to come up with a logical reason why these people are coming up with the same exact results (or even any results). You can't deny data that a person observed, you can only say it isn't true with other facts/reasoning for why it might have happened.
For example, one reason a person might not hear or notice burn in is because the change in change of sound (sounds weird, mathematically speaking it'd be dC where C = change in sound @certain frequency) is small and not detectable. Different ears hear differently. This is just one reason, there may be others. This does not mean that it is happening to you.
All I'm saying is that there has to be a reason why people are hearing things the way they are, we have two distinct groups, and there has to be a reason why both hear these changes (or lack thereof) the way they do. Placebo is not an answer for either since the number hearing it are too large. Neither are magical fairy tales.