that the older garden tractors were capable of doing what they do because they're "heavy". Weight isn't the only issue. It's also how efficiently the engine's power is put to the ground.
Which is what I'm saying. Would, (or could), a 2 hp engine powered HST tractor be practical? The 2 hp non-HST example you gave that would spin its wheels under load illustrates what I mean. Put that same 2 hp engine on a machine and couple it to a hydrostat and see if the results are even remotely similar. They won't be. You'll have to have the engine revved up higher to efficiently drive the pump. Then once you add up the associated losses along the way, you'll end up with a lot less power being transmitted to the wheels.
The larger the equipment is, the less it all matters. If you're designing a machine to do a job, and that job is going to require 50 hp, then upsizing a powerplant to 55-60 hp to compensate for losses in a hydrostatic system is no big deal.
The smaller something is, the more it matters. Consider a smaller and simpler "machine" designed to transport you from point A to point B using the power you can provide manually. You can hop on a normal bicycle and do the job easily. Now imagine someone designed a bicycle that you propelled by pedaling a pump, that in turn sent power to the wheel via a hydraulic motor. In that situation, the losses and inefficiencies would be greatly magnified. You would expend a LOT more energy to travel the same speed or distance.
Machines that are gear-driven, (or chain-driven, or belt-driven), are not only built that way because it would be costly to re-tool to a different type of power-transmission. It's simply that what the machines could be re-tooled to use instead of the gears, chains, or belts costs more than just the $$ spent for the bits and pieces. Efficiency falls off as well.
