Why abundance of matter requires fundamental asymmetry - cannot be statistical symmetry breaking? There are rather no doubts that there is more matter than antimatter in our Universe, it is usually assumed that it was created during short baryogenesis (violating baryon number conservation) period of Big Bang.
This asymmetry is used as motivation to search for a fundamental difference between matter and anti-mater, e.g. violating CPT. I wanted to ask why it couldn't be explained by symmetric rules? For example:
In presence of baryon it is a bit more likely for "genesis" of baryon than anti-baryon.
In presence of anti-baryon it is a bit more likely for "genesis" of anti-baryon than baryon.
would lead to statistical symmetry breaking: the more (anti)matter there is, the more (anti)matter will be created - initial tiny random imbalance would be statistically amplified, through annihilation leading to final domination of one of them.
Analogously for life using D-sugars instead of L-sugars: during the origins, the more one type of sugars there was in the environment, the easier life of ancestors using this type of sugar - initial random imbalance should be statistically amplified to total domination.
Do we really need some fundamental asymmetry between matter and anti-matter to explain abundance of matter?
 A: I will turn my comment into an answer, still a hand waving one. 
CP violation is a quantum mechanical phenomenon, so a phase transition you describe could not happen in a system that is dominated by classical thermodynamics. Classical  thermodynamics enters the history of the universe when protons and neutrons are formed,

In classical thermodynamics patches  would appear.The interactions of gravitational concentration of energies separate matter in space   and there is no reason to suppose that the observable universe would be one such patch, i.e. a fine tuning would be needed. There is no observation of antimatter in the universe, checking for decay products  of annihilations .
If instead of in the classical thermodynamic stage, a positive feedback exists in the quark gluon soup, when first baryon numbers appear, as it is a quantum mechanical system the transitions might be induced by a particle similar to the inflaton, ( a violon ;) ) which would then be carrying the CP violation term, and the uniformity of the present universe baryon number violation might be explained, the way the uniformity of the cosmic microwave background is explained.
In a nutshell, CP violation is a quantum mechanical observation, and in quantum mechanics interactions  are carried by particles, so your "it is more likely" has to do with coupling constants and feynman diagrams.
A: A positive feedback is sufficient to ensure that one type dominates locally, but not globally. Two patches that are sufficiently far apart that they don't interact during the time that these baryogenesis reactions are permitted may easily reach different endpoints, so you end up with patches of matter and patches of antimatter. 
We've seen no evidence for such patches (primarily, gamma emission from annihilation reactions on patch borders), but of course the patch size could be larger than the observable universe.
A minor problem with your theory is that it doesn't mention leptons. I'm pretty sure that we need the matter / antimatter ratio of leptons to match that of baryons. Although if neutrinos are Marjorana particles then that (probably) becomes a non-issue. While there are practical difficulties in measuring the neutrino / antineutrino ratio, it would be somewhat surprising IMHO if the matter / antimatter ratio of the charged leptons isn't very similar to that of the baryons.
