> If you have an electric dipole [...] the force drops off like 1/r^3
> This result is for a single quark in isolation
Sorry, I'm confused. Doesn't this confirm what I'm saying? For multiple particles I get how it could be different from 1/r^2 (though only less than 1/r^2, not more!), but as you say, this is about a single quark in isolation, which is neither multiple particles nor less than 1/r^2, so the problem is still there right? (The fact that quarks aren't ever found in isolation in practice seems irrelevant to me, unless the claim is "quarks absolutely cannot be found in isolation due to conservation of energy", which I've never read.)
And what about something like MOND? I see the same problem, and it's about gravity!
The dipole example is meant to illustrate the fact that if you have more than one particle then you can have faster decay as you move away from zero. In the dipole example you also have opposite electric charges, which is analogous to the colour charge in the case with the quarks.
The strong force has a property called color confinement which means that quarks cannot exist in isolation under normal circumstances. This is a property of the SU(3) gauge theory that describes the strong interaction, as well as something which we observe experimentally. It's not a direct consequence of conservation of energy, although energy will be conserved.
MOND is a totally different situation. For the strong force we write down a totally solid theory which is built from all the necessary symmetries and agrees well with current experimental data. In MOND, people take Newton's second law and change it to try and fit the data better. My understanding is that this doesn't work very well, and also I see no reason why it should conserve energy a priori (i.e., maybe it does but this is not clear to me without seeing a proof). Conservation of energy when gravity is taken into consideration is more subtle, so MOND may not need to conserve energy.
A wire doesn't look like a point source from far away, unlike a dipole. Cut off a segment of that wire (so it's not infinite-length) and it won't be 1/r at large distances.
> This result is for a single quark in isolation
Sorry, I'm confused. Doesn't this confirm what I'm saying? For multiple particles I get how it could be different from 1/r^2 (though only less than 1/r^2, not more!), but as you say, this is about a single quark in isolation, which is neither multiple particles nor less than 1/r^2, so the problem is still there right? (The fact that quarks aren't ever found in isolation in practice seems irrelevant to me, unless the claim is "quarks absolutely cannot be found in isolation due to conservation of energy", which I've never read.)
And what about something like MOND? I see the same problem, and it's about gravity!