My credentials:

1. I work directly with the development teams at DTNA - the arm of Daimler that makes Freightliner and Western Star trucks, as well as Thomas-Built school buses.

2. I just attended the School Transportation Summit (virtual) this week, where two of the presenters were Lion and Thomas-Built buses, showcasing their electric school buses.

3. And my company's customers include the two largest long-distance freight haulers in North America, Knight and Swift Trucking.

That being established, the math/physics has already been laid out by the other posters on this thread, so I won't even dip my toe into the 'power vs energy' thing. I'll deal with the business end of things.

Charging.

School buses are a perfect model for electrification because they run established, pre-defined routes, and have long periods of inactivity throughout the day. Trucks are the exact opposite. They run for long hours in highly variable road, traffic, and weather conditions, which are all variables that affect battery efficiency. Plus, there are LOTS of trucks. And they can only stop in certain places. Getting enough voltage and amperage to a single location where 50 trucks are parked overnight and charging... that's a problem that cannot be solved, practically speaking.

Cost.

Yeah, everything comes down in cost over time. But. An electric school bus today is around 3x the price of an ICE-powered school bus. That is not in the budget for ANY school. HOWEVER... electric vehicles take almost no maintenance, every fungible part is replaceable, and could theoretically have an infinite(-ish) lifespan. Today's diesel school buses may run for 20-25 years, but they fall apart over time just due to wear and tear on drive systems. You can replace the motor and transmission so many times, before the entire vehicle just isn't worth saving.

New electric buses have far fewer parts that get far less wear and tear, and every part can be replaced for a reasonable amount of money. You could possibly be operating the same electric school bus 50 years from now. It's a weird thing to consider...

Infrastructure.

A couple miles from my house, one of our largest customers has a vehicle yard with something like 600 school buses in it. They will need to be charged, if they move to electric. What is fascinating is that local power companies are working with school districts to install charging stations FOR FREE, so schools can charge their buses throughout the day. (Which is different than long-distance truckers, who generally need to charge at the same time of day, or can't/won't run on a partial charge.)

But this is all VERY new. The business models are evolving, even over the last 2-3 months. People are throwing ideas at the wall and seeing what sticks, and thinking in really new ways about transportation.

Thinking about Bridges vs Ferry Boats.

Everything comes down to cost-benefit analyses. Sometimes it is better to build a bridge between two bodies of land separated by water, and sometimes a ferry boat is a better option.

Here in Seattle, we have both.

It is the consideration of all smart commercial fleet operators that electric will be a part of their fleet within the next year or two, and that portion will grow fast. Once you pull the trigger and get your first electric vehicle in your commercial fleet, the barriers to your second, third, ... and hundredth vehicle are gone.

BUT. There will always(?) be ICE-powered vehicles in your fleet. Eventually they may be used only for special purposes (firefighting, emergency response, and other long-duty cycle operations), but fleet managers today know they will have electric vehicles in their fleet that they need to manage by 2022.

Innovation.

At some point, someone is going to invent a denser, longer-lived battery. They may already have one in testing, if recent news reports are to be believed.

The big easy money for them will be to provide these in form factors where they can be fit into existing electric vehicles, to replace aging, failed, or less-efficient battery technologies.

And it is WAY easier to drop in a new battery pack and controller than it is to replace an engine, transmission, and drivetrain to accommodate the latest Cummins/Detroit or other diesel engine.

So fleet managers want to buy into an ELECTRIC PLATFORM, rather than a specific vehicle.

Which is where I think Musk is really on the right track... an entire line of electric vehicle products, supported by manufacturing, that could churn out replacement battery packs for everything from a 1999 Honda Insight to a 2021 Thomas-Built Bus (which is named "Jouley", in fact), as well as their own line of vehicles.