what is the advantage of the higher ohm arm?

That's not a correct assumption, but I think I can explain this in non aspirin inducing terms...

This is kind of a trick question because we are usually not simply comparing the difference in armature resistances for the types of motors you are probably thinking about, say a 15 ohm standard TJet arm versus a 6.5 ohm TuffOnes or A/FX arm. If the only difference between two otherwise identical motors is the armature resistance, there would be absolutely no benefit to the higher resistance armature unless you were trying to go slower with a less efficient armature.

To understand what's happening you need to look at the torque equation for a permanent magnet DC motor. A simplified version of this equation is:

T = Ka x I

where T is torque, I is armature current, and Ka is something called the "armature constant." Putting Ka aside for now, this equation tells us that the motor torque is proportional to current. You already know that the armature current is inversely proportional to armature resistance (Ohm's Law), so a motor with lower resistance will have higher current and therefore greater torque. That's why I said there would be no benefit and only a detriment to having "two otherwise identical motors" where one had a higher armature resistance. The higher resistance arm would suck. If you were to plot the torque of each motor for a given armature voltage, the higher resistance arm would always be significantly lower. The shape of the curves would be the same, but the magnitude of the torque would always be lower for the high resistance arm.

But, and there's always a but, what about Ka, the armature constant? That's where we have to be careful about what we are comparing because that's that's where things get interesting. The two arms I mentioned earlier, the TJet 15.0 ohm and the Tuffy 6.5 ohm, well, they are not exactly the same. What's different between the two is much more than the resistance, it's also the Ka. These two motors have different armature constants. Because they have different armature constants the shape of the torque versus voltage curves between the two motors would be different because the Ka factors are different. The magnitudes would also be different, but because of the different resistances.

So what is the armature constant? Without getting into too much detail, it's the electromagnetic force constant, or something that describes how good of an electromagnet the armature pole is. What's the biggest variable in figuring out the electromagnetic force constant, or Ka? The number of turns of wire on the armature pole. The more turns, the better the Ka.

So... when you are comparing a 15 ohm TJet arm to a 6.5 ohm Tuffy arm, you are comparing two arms that were designed with two different strategies in mind. The TJet arm is designed for a higher armature constant Ka, which yields greater torque at lower controller voltages, and the Tuffy arm was designed to give up some of the armature constant Ka (fewer turns of wire) but deliver more torque by upping the armature current by reducing the armature resistance. Why did the motor designers choose the strategies they chose? Perhaps the power supplies available for use in toys at the time of the TJets inception were expected to be run with less current available or maybe they just hit on a formula that worked well and they liked the results. They chose wisely, at least for home race set use. If you really wanted to see the difference between a TJet arm and a Tuffy arm you could develop some graphs that show the torque curves. Or you could slap the arms in a car and see how it all plays out in the real world. The empirical results would obviously support the calculated ones.

The ideal case of course would be to have an armature with a high armature constant and a low resistance. This sounds trivial enough, but in reality the mechanical design of the armature and the amount of space available on the armature poles presents a hard limit. Lower resistance windings are physically larger so stuffing a lot of turns on to the pole is more difficult. Using lower resistance winding conductors, say silver wire, to keep the size small would theoretically help, but the thermal properties and cost issues would come into play. Engineering is all about solving real problems and providing real solutions that apply to the real world. The design of motors both large and small must be amenable to engineered solutions and all engineered solutions represent a compromise between multiple possibly conflicting requirements and concerns.