"An ICE engine can hit about 30% efficiency." Yes, but they don't operate near this most of the time. This only happens at a particular engine speed (RPM) and load. This might happen on the highway if the motor is at its optimum speed. You're giving ICE way too much credit on efficiency. There are no measured numbers available that I'm aware, but if you factor in most people's driving habits, which include start-up, warm-up inefficiencies, stop-and-go which runs RPMs out of optimum bands and brakes which throw energy away, I would guess 15% is much more likely and even 10% or less for many drivers. By contrast, power plants run at optimum points all of the time.
You also forgot EV regen, which gives them a significant boost in stop-and-go driving over ICE.
ICE: 15% * 92.5% = 13.8%
H2: 30% * 80% * 50% * 85% = 10% (see below)
You also left out something very important. Gasoline/Diesel REFINING!!! Estimates by EPA put the energy cost at roughly 6 Kwh per gallon of gasoline. Also add in transport and pumping. Try factoring that into the ICE equation. I'm guessing it puts it way below 10%.
Your H2 calculation has a serious problems too. Electrolysis starts with electricity, so it gets the 50% hit plus 98% transmission hit before electrolysis even starts. I'll give you the benefit of the doubt and assume you factor that into your 30% number. But you left out transport, and compression (10,000 psi) and pumping 85%. Then the fuel cell charges a battery because it can't produce sufficient power on demand. So you have to apply the 80% charging hit you applied to the EV.
"EVs and hydrogen in inextricably linked in this way"
Hell no, they're not! Hydrogen comes from natural gas now, which will always be cheaper than electrolysis. The hydrogen economy is a natural gas economy.