@masiarek it’s common to consider the plurality/choose one algorithm as operating on rank ballots, it just counts the number of first place rankings of each candidate and chooses the candidate with the highest count. The tallies are just those first-place counts. In these cases, since the lower rankings don’t matter for the outcome of the method, the ballots usually serve as indications of true preferences, or in contrast, as examples of tactical voting, usually to mitigate vote splitting.
In this example, you can see that under plurality voting, the two top-right voters would benefit by tactically raising C above B against their actual preference, essentially colluding with the lower four voters. In a sense, B would have spoiled the election for C. This is vote splitting and Duverger’s law in action, and ultimately the reason we have two huge honking awful parties.
It also demonstrates how Condorcet methods are resistant to certain kinds of tactical voting. In the example, C is the Condorcet winner and A is the Condorcet loser. In an election that may fail to elect an existing Condorcet winner, the majority who prefers the Condorcet winner to the alternative has a tactical incentive to collude and form strategic ballots to secure their preference.
Unfortunately this may be a slim or non-existing majority, but the hope is that it’s a broad one.