There’s no clear answer to this one, because it depends on many factors including the geometry of the situation.
Mains voltage electrocution usually causes death by stopping the heart. The current required is typically rather small - a few milliamps will do it, but this current has to pass close to the heart itself.
Electricity transmission uses the Earth as one leg of the transmission cicuit, so in any scenario you have the “live” or “hot” terminal, and current flowing between it and ground. Just about your worst situation is to have a hot wire in one hand and your other hand grounded, causing the current to flow through your chest. (A nice way to do this to yourself is to firmly grasp your metal-bodied toaster in one hand and try to dislodge the stuck bread with a metal knife using the other. Get yourself a Darwin award!)
So let’s throw a single bar heater into an occupied plastic bathtub. Current will flow from the live end of the element to the grounded end. The easiest route is still through the heater bar itself, and this will take the majority of the current. However, a small fraction will flow through the water - you can visualise it as passing in a looped path out from one end of the bar and back into the other. And some small fraction of that will pass through your wet skin, the bag of salty water that is you, and back to the bar. The longer the loop, the smaller the current though, and you have a good chance of surviving this, especially if you have a ground fault trip on your power supply.
A nastier situation would be someone throwing a single live wire into the bathtub while you’re touching a faucet. Then pretty much all the current will flow through you to ground via the metal water pipes and your chances are poor.
Do the heater trick in a swimming pool and there is a definite range factor at work - someone quite close to the heater is going to get a bigger dose of current than someone at the far end of the pool. Chances are good that most or all of them would survive - that heater bar really does suck up a lot of the current.
It gets more interesting and dangerous if someone drops a live wire into one end of the pool and a grounded wire into the other. If the pool lining itself is an insulator, then you’re going to have all the current passing through the water from one end to the other. The current distribution will be complicated - further from the wires will be better than close, and corners of the pool better than the centre. Not a happy place to be in any event. For more democratic electrocution, have a live metal sheet lining one side of the pool and a grounded one lining the other, giving a uniform current distribution and frying everyone evenly.
If the pool lining isn’t an insulator you can dispense with the grounded wire or sheet. Current distribution will be altered and highly complicated - again, best to be furthest from the live wire.
For really large grounded bodies of water such as the sea, you can assume hemispherically symmetrical radial current distribution from a live wire dipping into the surface. The danger you’re in depends upon what fraction of the current you intercept. So if the total current flowing from the wire is 10 amps, at 1 metre from the wire you’re getting 2-3 amps through you which is not good (assuming intercept “area” of a human is 1.5-2 m[sup]2[/sup].) At 10 metres from the wire you’re getting 0.02 to 0.03 amps which may still give you trouble. At 100 metres you’re only getting a few tenths of milliamps and wondering why all those people near that cable are being so noisy. At great distances the hemispherical distribution is no longer valid but the current densities are so low it makes no odds anyway.