I have never seen any evidence that a "burn in" would yield higher OC results. It is possible for it to reduce temps, which could possibly increase the overclock, but the difference is probably negligible.
The reason why it could help reduce temps is that it might help the thermal compound set in, but a 24 hour burn in might not lower temps much if you've got AS5 or something similar.
From Arctic Silver 5 website:
it will take a up to 200 hours and several thermal cycles to achieve maximum particle to particle thermal conduction and for the heatsink to CPU interface to reach maximum conductivity.
not so much burning in as stability testing new components..
i do all the relevant tests when i get something new so that i can RMA it immediately if there is a problem and not later on when it fails at a crucial moment...
If something is going to break you want it to happen while it's brand new and you can replace the parts while they are still under the 30 / 90 day whatever you got no questions asked return policy. That's the big point of doing a burn in.
If your parts are not under warrenty don't do it.
I have heard you might get some kind of small gain in performance from it but it's such a small difference you would probably only notice it with testing software not real life applications.
Personally I have only done a burn in one one build and that was only because It was an ultra cheap deal from a seller I didn't trust and It only had a 30 day return on parts so I wanted to find out if it was going to break.