ORIGINAL: rpg711
Remember that the more current flows through a wire, the hotter it gets, and the hotter it gets, the higher its resistance is. This is why power lines are always running something like 50000v. Higher voltage on a lower kV motor, or on lower gearing(effectively lowers the amount of power the motor needs to draw to reach its peak torque/power) results in a cooler and more efficient motor. You didn't give enough information so we can't really tell you anything. With 3s with any buggy/truggy/whatever you will need to put heavy heavy oil in your center diff, or you will kill your diff by doing wheelies(front and back on ground, functions as normal diff, once the front end lifts off, it will begin ''diff out'' and reducing power to the rear... that kills your gears in no time), especially with the instant power of a brushless, if your battery can provide it, and your esc can support it, the motor will draw the max current that it requires to generate its max rpm. If there's something that has LESS rated amp capacity, that means the internals -or- externals(wires, pathing, used hardware) cannot support the current and will start heating up, which is unfavorable because the more hot it gets, the more resistance it has, and the more resistance it has, the hotter it gets, starting a constant heatup until power is reduced. It's my belief that anyone looking at electrics should read on ohm's law, and just general electrical theory.
In practice this doesn't happen in RC though, it's not that uncommon to hear about an ESC toasted or a battery burnt up from overcurrent/overtemperature. And in these cases the wiring is typically fine. Sometimes people are 'saved' by poor solder joints on battery packs and things like that the solder joint is bad, it heats up, solder turns to liquid and the connection is broken. But that's about it.
Except in cases where the wiring leaves much to be desired, the ESC or battery is going to fail first. The potential for positive or negative feedback that you describe isn't typical at all.
"if your battery can provide it, and your esc can support it, the motor will draw the max current that it requires to generate its max rpm"
The motor draws current based on throttle input. Primarily when you accelerate. Hard acceleration is where your current is highest, just cruising along at max speed (max RPM) doesn't draw a lot of current unless you are on heavy terrain. This is just like an ordinary street car. Softer acceleration up to a constant speed is much more efficient than harder acceleration up to that same speed.
It's important to know when to apply theory and when to draw from experience. You apply theory when it's useful to simplify, model, or demonstrate experienced results. The statement "starting a constant heatup until power is reduced" can be very confusing. The increasing resistance (component, wiring, etc.) just isn't going to be enough of a factor to create any kind of negative feedback. I recognize that increasing MOSFET resistance may create a little bit of positive feedback here, since the ESC typically has much less resistance than the motor or anything else in the chain (so as that MOSFET resistance increases, current only drops slightly, power disipation increases -> positive feedback). The statement is only true in retrospect, the device heats up, it fails, and only then is power reduced, typically to zero (since a failure typically opens the circuit). There can be cases of partial failure, but they are rare. If you are lucky, you may end up simply thermaling the ESC (temperature sensor inside the ESC shuts it down or restricts usage to save itself). But these can never be counted on...