Not necessary, because although the author does believe that allowing a candidate to buy more media exposure than others can afford is something that detracts from the goal of fairness, that doesn’t imply the author thinks there’s nothing else that would detract from the goal of fairness. The author could agree that there are many other things that could be unfair; this argument just happens to concern one thing that could be unfair.

Although the author does assume that more media exposure can help a candidate evoke positive feelings from their image, that doesn’t imply the author thinks “familiarity” is necessary for positive feelings toward a candidate. More media exposure does not necessarily imply “familiarity.” And the author could think there are other factors that also contribute to positive feelings; as long as the amount of media exposure is one of those factors, the author’s reasoning can still work.

Necessary, because if it were not true — if the tendency of a candidate’s visual image to evoke positive feelings NEVER increases as media exposure increases — then we’d have no reason to think a candidate’s greater level of media exposure constitutes a problem of fairness. Even with more media exposure, that wouldn’t necessarily help them evoke positive feelings.

Not necessary, because the author doesn’t have to have any belief about what candidates “invariably” do. Even if candidates do not always buy as much media exposure they can afford, they might still buy a lot more media than their candidates can afford.

Not necessary, because even if some candidates whose visual image does not evoke many positive feelings CAN be elected, on average it can still be unfair to allow candidates to buy more exposure than their rivals can afford. The author believes that on average, those candidates can help make their visual image evoke positive feelings, which increases the chances of election.