While searching for opinions about answers to this question I found mostly promotions for adopting Instant Runoff Voting (IRV). This is a widespread notion and the argument for it is largely that this voting system does dispense with the spoiler effect and it does assuage the feelings of voters who want to vote for someone not from the two dominant parties. But it is less than clear that this would put an end to the polarized two-party duopoly. It seems more likely that these votes, no matter how satisfyingly cast, will simply be converted to votes to one of the two dominant parties. Ranked voting generally and IRV in particular have been discussed repeatedly and from many different perspectives, in earlier articles.
One promotions for IRV that I discovered was from The Institute for Political Innovation. While I am very skeptical of their choice of voting method, we do agree in believing that the polarized two-party duopoly needs to go. We also seem to have come up with the number five as a reasonable number of candidates to stand in the final election; this seems like enough candidates to allow voters a choice without expecting voters to become familiar with a bewildering number of candidates. An earlier preliminary election can be used to winnow down the candidates to five or so.
In earlier articles of this series we have repeatedly demonstrated how balanced approval voting (BAV) will put an end to the polarized two-party duopoly (see for example a recent article or a very early one). But a careful reading of the argument reveals that what would be ended is a polarized, two-party duopoly. Voters in favor of the candidate for one of the two dominant parties, because of the polarization, will choose to vote against the candidate from the other dominant party (provided the voting system permits that). That surely is critical to the argument. BAV puts an end to the two-party duopoly due to polarization and that leaves open the question of whether there could be a two-party system that is not polarized. In such a system, many voters would simply prefer, or at least find acceptable, a win by either of the two dominant parties.
If that were to happen, it would seem to imply that the two parties have nearly identical agendas and that agenda is shared by the voters. The question then arises whether this could possibly become a stable situation? Would divisive issues that divide the parties never arise? Moreover, if this unlikely situation did arise how harmful would it be?