I'm a working physicist (not in fusion, admittedly) and I think you're worrying too much.

For one thing, high temperature superconductors are starting to hit the market in a big way. These superconduct at liquid nitrogen temperatures so there's not much need for helium. Companies like American Superconductor are producing hundreds of kilometers a year and are ramping up production still further.

Also cryocoolers, essentially electronic refrigerators, allow you to keep your helium bath at 4K or below without any boiloff at all (and for any experiments can eliminate the need for helium entirely). This isn't in real common use yet but the price and performance of cryocoolers is improving fast. (For some reason they're very common in Japan, though.) Even at current cryocooler and helium prices, a lot of physicists are considering moving to cryocoolers since helium is a pain to work with for various reasons. If helium got even a little more expensive you'd see a pretty big shift. We designed an experiment recently and had a really hard time deciding between helium and cryocoolers. (We went with helium, because our experiment is extremely vibration sensitive, but that's not going to apply to most people.)

vinc -

And exactly how much energy do these cryocoolers consume per hour? If your power plant is using NG to generate your electricity, then you're still on a depletion curve. If NG becomes difficult to get hold of, then you're looking at another problem, but that one too, is related to resource depletion.

One of the points made is that the inputs for many of our materials are fossil energy - steel alloys, stainless, magnets, ceramics, silicon...  The question posed is: Can these materials be manufactured without NG. I think it is a very valid question. Heating with electricity has typically consumed more energy than heating with NG in manufacturing.

Replacing NG with electricity actually consumes more NG at the electricity generating end, and throws a very large load on the grid - that's why manufacturers have a special electric meter (called a demand meter) when they use large amounts of electricity. You get billed at the highest rate of consumption - if it's 10kw an hour for 2 minutes, then you get billed as if you are drawing 10kw an hour for everything you use. It definitely affects the bottom line, and it's there to discourage pulling a lot of juice.

These are the things a lot of us are thinking about and discussing - energy inputs and how to offset them or find alternative sources that are workable. The fossil fuel train is slowing down - we have to get off if we are to continue our journey forward.

It's going to take roughly the same amount of power to liquefy the helium at a central  plant as it will to run a cryocooler at the point of use.  Plus, you're eliminating all the trips made by the helium tanker.
Less power, I think. With a cryocooler you're just maintaining a constant temperature. Whereas when you liquefy it you have to do all the work of bringing helium from room temperature to 4 K, not to mention purifying it...

At any rate, a cryocooler that would have been much more than sufficient for our experiment would have used about 5 kW, which is comparable to central air conditioning for a large home in summer. A fusion plant would be a heck of a lot larger than our experiment, of course, so I'm not sure what their total consumption would be.

Well, presumably the key issue would be the EROEI of fusion.  The energy for the cryocooler is included in the "EI" part.  So the goal remains the same - to get fusion to the point where it's a (controllable) energy source not a sink.
An important property of a superconductor is its critical current density (i.e. the maximum current density it can carry while still remaining a superconductor). This is a particular problem for superconducting magnets since the critical current density decreases in the presence of a magnetic field. While high temperature superconductors exist (~110K) they don't have the critical current densities needed for magnets producing 13 Tesla fields where as for low temperature metal superconductors (e.g. Nb3Sn) it is just about possible. People are talking about using high temperature superconductors in Maglev trains some time in the future but the field strengths would be a lot lower than a tokamak. It should also be noted that because of all this the generally held notion that if we could get a viable fusion reactor working it would be perfectly safe is probably wrong. A full scale tokamak that produced useful amounts of energy would be huge (football stadium size) and hense the associated magnetic field would not only have a very large field strength but would encompass a massive volume. Playing around with the equation E = B^2.V/(2mu) you should be able to convince yourself that the energy stored in the field would be enormous and therefore a magnet coolant failure could be catastrophic.
The existing limits on current density in HTS superconductors are not fundamental, and I suspect they'll improve by quite a bit. (Typical high-Tc wire these days, due to manufacturing limitations, is actually made mostly of silver and other metals, with only a little bit of superconducting material. The other stuff is added for mechanical strength.)

I'm actually not a big fan of fusion research, but for other reasons...

Experiments at Purdue with sonofusion looks promising. Unlike tokomats the strong magnetic fields are not needed. Unlike claimed cold fusion experiments no precious metals are needed. Hot fusion research is a lost cause. They have had over 50 years of experiments behind them and have only produced arguments for more government grants.