I think we will disagree with this one.
As the antibiotics are administered, the ones with no resistance will die off first. As there are several methods that bacteria can become resistant (change in drug permiability, change in metabolic pathway, drug inactivation, or simply pumping the drug back out of the bacterium), not all strategies for drug resistance are equally effective. The bacteria with the ineffective drug resistance die later during the course of medication. There may well be some bacteria with effective resistance, but we hope to get the numbers so low that they are are unable to colonize well, and that the body's defenses can handle the few bacteria that are left.
If we were dealing with a petri dish with antibiotics, we would have a resistant strain. Since the body is (unless immunocompromised- that is another issue) also destroying the bacteria the very few resistant bacteria left from a properly administered course of medication will generally be destroyed. As for the numbers question, A single (or few) bacteria are generally not enough to colonize a body. When we grew fermentation, we had to go through several fermentations of increasing volume to make enough innoculum for a 100 liter batch. Just putting a drop of whatever we wanted to grow up into the 100 liter fermenter would not guarantee any growth.
By stopping the antibiotics early, there are enough bacteria present to recolonize, and too much for the body to clean up on its own. They all have resistance to varying degrees, and they will trade DNA so the strain with the best resistance will continue to grow.
If the body wasn't helping to clean up the infection, I would agree with the statement, but the body's immune defenses are involved. The bacteria might be resistant, but the body should clear out that low number so they can't grow or colonize elsewhere.