When starting a liquid rocket engine, the pragmatics of valve timing, dribble volume, turbopump spin up (if applicable) and other transient fluid dynamics mean that you're going to naturally have one propellant "lead" the other into the chamber.

It's common design practice to bias this intentionally toward one propellant or the other to give a consistent startup behavior. So, which one? Do you bias toward a fuel lead or an oxidizer lead? Well, that depends on the propellant.

The disadvantage of an oxidizer lead is that you will have a transient operating condition where you have a high flame temperature AND an oxidizing environment. This is an environment in which metals burn. Steady state operating is usually fuel rich.

In hydrogen engines, these concerns have led designers to generally use a fuel lead startup. But hydrogen has a unique property that makes this the obvious choice, it has nearly symmetrical flammability limits.

Flammability is depicted on a ternary diagram. These can be a little odd to read without practice. Focus in on the left diagonal line. Basically this says that a binary mixture of hydrogen and oxygen is flammable at hydrogen concentrations as low as ~5% vol and as high as ~95%.

Compare to the diagram for methane. Unlike hydrogen, it shows substantial flammable asymmetry in binary mixes with oxygen. Still flammability at ~5%, but concentrations above ~60% won't light. This is typical for hydrocarbons, and the following roughly holds for kerosene too.

As a simple model for a fuel lead, imagine injecting oxygen into a chamber flooded with hydrogen gas, with perfect mixing. At 5%vol, it lights and reacts with ~10%vol of hydrogen. To a first order, the remaining 85%vol is just buffer gas that doesn't participate in the reaction.

Now, imagine an oxidizer lead as injecting hydrogen into a chamber flooded with oxygen gas. At 5%vol, it lights and reacts with ~2.5%vol of oxygen. ~7.5%vol of the gas burns. 92.5% is buffer mass. A little better, but fuel or ox lead, the percentage of gas that reacts is low.

Try an ox lead with methane, and the result is similar. Ignites once you reach 5%vol, and reacts with 10%vol oxygen. 85%vol doesn't participate. Similar to a fuel lead with hydrogen.

Now, try a fuel lead with methane. Introduce oxygen into a chamber filled with gaseous methane. Ignition doesn't occur until you hit 40%vol oxygen, which reacts with 20%vol methane. 60% of the gas volume in your combustion chamber ignites.

This can cause what is called a "hard start". Chamber pressure rapidly spikes far above steady state design pressure, and can cause severe or catastrophic engine damage. This is a great way to get a "blue" nozzle. As in, "it blew way down over there" (it's a better verbal joke).

Now, this is all a very simplified model for very complex combustion physics, and the specific conditions of a given engine may vary, but it is still instructive. The takeaway is this. Most, if not all, hydrogen engines start with a fuel lead. Hydrocarbon engines often don't.

For hydrogen, it makes sense to avoid the hot, oxidizing transient point during startup, and precluding a hard start is relatively easy either way.

For hydrocarbon engines, and especially ones with hard to light, heavy hydrocarbons like kerosene, it has been common to accept the hot, oxidizing transient, design to get through it quickly, and lead with oxygen to avoid the risk of hard starts.

Peroxide engines have an additional complexity added here. The oxidizer vapor is detonable. For a catalyzed engine, this isn't a big deal. You crack the peroxide in your catalyst bed and are functionally doing an oxidizer lead with very hot oxygen into which you inject fuel.

For a "wet" peroxide engine though, you have to ride the zone between a hard start due to the fuel flammability limit coming from the rich side, and the risk of peroxide vapor detonation coming from the lean side. Valve timing is important. Fast hypergolic ignition helps.

Wet peroxide engine with a torch igniter? You can do it, but you should use an easily vaporized and mixed fuel. Wet peroxide engine with kerosene? That sounds like something I'd love to watch provided suitable shrapnel shielding.