Before 1900, maybe before 1912, virtually all verse translation of metrical verse was also metrical, and usually rhymed for short lyric poems, especially when the original rhymed. This is mind-numbingly obvious, and the reasons are not hard to find:
*There was such a thing as the prose translation of verse, so translators who did not want to aspire to verse translation could translated verse into prose.
*Very few people were writing their own poetry in free verse, so verse meant, essentially, metrical verse.
*Very few of the works being translated were in non-metrical or free verse either.
This obvious fact also has several secondary implications. Before going into those, I should also point out that:
Usually, translation was into the prevailing metrical norms of the target language. So rhymed couplets or blank verse for longer poems, the ballad stanza for ballads, iambic pentameter for Italian and Spanish 11-syllables lines and French 12 syllable Alexandrines, etc... There were attempts to write hexameters to translate Homer, but they weren't as frequent. Thus metrical / cultural equivalence was usually the norm.
And a third fact:
After a certain date (19??), metrical and especially rhymed translations became much less frequent. Almost all translators today take this for granted, or at most have a perfunctory statement to the effect that "I have not tried to rhyme..." The reasons are also not too hard to find.
*Readers have a different standard for how literal they want their translations. They are no longer willing to tolerate a lot of latitude in interpretation for the sake of a rhyme or metrical pattern.
*The translator is not likely to be writing her own poems in metrical verse in the first place, and might not have any skill in doing so. The idea of translating into a prosodical norm thus prevails in this period, (19??-2017), only now the prosodical norm is free verse or extremely relaxed blank verse.
*The originals are less likely to have strict metrical patterns than before, if translators are mainly interested in poetry written in the past 100 years.
The consequence of these three obvious facts is that there is a before and and after, and that verse translation fundamentally shifted its ethos at a certain historical juncture, probably about the time I was born.
It was one thing, and now is something else.
It is hard to prove a negative, but a lot of verse translation does not have a clear prosodic impulse behind it. It is essentially prose translation (remember our old friend, prose translation?) but broken into lines to match the original.
If a translator has to strain to make a verse fit a metrical pattern in the new language, wouldn't he also have to make some effort to make a line sound good as free verse, not to mention the entire poem? Would we expect that the poem would automatically have a rhythmic flow to it just because it follows the phrases and lines of the original? Now we are moving from the stunningly obvious to an analysis that needs to be extremely nuanced.