How FPTP Silently Sabotages Fair Union Elections

There seems to be a problem with FPTP voting in that “most votes wins” is a highly intuitive system for a single-winner election, despite the fact that this can result in a winner who would lose in a two-person run-off against every other candidate.
There is also an assumption that the winner under FPTP is genuinely the most popular, while the winner under IRV (Instant Runoff Voting) is merely a compromise. In reality, this isn't true—the winner in a run-off system is much more likely to be the most popular overall.

This phenomenon occurs because of the spoiler effect: several similar candidates can split the vote, while a very different candidate, disliked by the majority, can win by appealing to a large enough minority. Under FPTP, the winner could easily be someone who would lose in a two-person run-off against any other candidate.

The flaws of FPTP are particularly problematic in union elections, where there is no polling data showing who is likely to win and people don’t always see the final results.
It would not surprise me that it’s not entirely uncommon for the vote to be so highly fragmented that the selection of the winner could be considered arbitrary.

Why such a flawed system persists, despite better alternatives, I can only attribute to its simplicity and intuitive appeal.
Allowing people to rank candidates in order of preference and selecting a winner by IRV would mean the winner is never arbitrary and is someone who always has a broad range of support from across the electorate.