A number n is selected, perhaps the number of candidates. Each voter lists their top n choices, in order of preference.
A first-place rank is worth n points, a second-place rank is worth n-1 points, down to an nth rank being worth 1 point. A candidate's score is the sum of the number of points they received. The highest-scoring candidate is elected.
In the trivial case of n=1, this is mathematically identical to plurality voting.
The potential for tactical voting is large. Voters are encouraged to list choices they believe are popular lower than they actually believe they deserve to be ranked, so that they don't compete with their top-ranked choice. It also encourages voters to assess the viability of each candidate, so as not to "waste" their vote on those with little chance of winning.
This was much discussed in the case of the 2000 elections for Most Valuable Player in the [American League]? of Major League Baseball, where there were two top candidates, and partisans for one were found not to have included the other at all on their ballot [1].