>>105656541
That's a literal <0.2% difference (noise, run to run variance is almost always bigger than that) and the colder phone could just have been in a colder environment (because of air conditioning, being in the Southern Hemisphere right now or because the phone was in a fridge)
It's meaningless.