Not necessarily, they can and do form Hydrogen Peroxide (H2O2) - so predetermined it is not.
And the rest of your argument therefore falls apart.
Hydrogen peroxide is not easy to form, directly, from oxygen and hydrogen. Hydrogen peroxide exists at a higher energy potential than water, therefore, once formed H2O2, still wants to react and/or decompose, to form water. Hydrogen peroxide is a detour toward water. If you use hydrogen peroxide, it gives off oxygen bubbles, and leaves behind water.
Hydrogen peroxide formation requires additional atoms/molecules so it can remain stable at its higher potential. This is also predetermined. Once you find the correct catalysts, the formation of hydrogen peroxide is reproducible in the factory, since it is meant to happen that way. Chemical manufacturing, that makes millions of pounds of product each year is based on predetermined results or else the product would not be reliable and salable.
One natural predetermined potential that helped the formation of life is connected to the water and oil affect. If we mix water and oil they will separate into two layers. If we agitate this mixture, we can randomize the mixture and form an emulsion. If we stop the agitation energy, being used to increase the system entropy and randomness, and allow it to settle, the induced randomization will reverse itself and two layers of order will reappear.
Water causes organic materials to phase separate in ways that minimize the potential of the water. In doing so it can cause organelles to appear, if you have the correct combination of organics; predetermined. The water and oil affect is the main reason other solvents do not work very well when forming life. Solvents like alcohols and ammonia tend to dissolve organics instead of induce phase separation. You would end up with a random blend or emulsion that cannot self assemble in a reliable way. Water sets the bar higher than the rest, allowing order to appear from random organic emulsions.
In modern cells, proteins are observed to fold with exact folds. There is no randomness in folding, unless we alter the system conditions to induce this. This observation demonstrated a flaw in the statistical assumption of life. This observation was not predicted by random centric theory, and once established as a verifiable fact, there is still not statistical explanation, even after 50 years.
The way you explain perfect folding of protein is, liquid water is so stable, due to self bonding, as evident by its high boiling point for sometimes so small, that organics are force to conform in ways that minimize the potential of the water. This need to accommodate water packs proteins in predetermined ways. It is like a water-oil affect at the nanoscale.
Proteins pack in layers with the moieties of highest potential in water; hydrophobic, having to phase out first to form the core. Below is an energy landscape diagram of a protein. The top diagram is a protein with little water or low hydration. The peaks reflect higher energy. Once you add water, you get diagram 2, where a;; the energy levels lower to minimize potential in water. These diagrams are predetermined for each protein.