What Markdown is
A Markdown table is rows of pipe-separated cells with a dashed rule under the header. Alignment is set by placing colons in that rule.
README files, GitHub issues, documentation sites, Obsidian and Notion.
What TSV is
TSV is CSV with tabs instead of commas. Because tabs almost never appear inside a value, quoting is rarely needed and the files are easier to read by eye.
Bioinformatics, log processing, and anyone who has copied a block of cells out of a spreadsheet: the clipboard format is TSV.
What changes when you convert Markdown to TSV
Leading and trailing pipes are optional and both are handled. A pipe the author escaped with a backslash is read as data rather than a column break, which is the usual reason a hand-written table parses one column too wide.
Pick TSV over CSV when your values contain commas. Pasting the result straight into a spreadsheet fills the cells correctly with no import dialog.
What carries over from Markdown to TSV
Markdown records 2 things about a table that TSV has no way to hold.
TSV has no separate idea of a header, so the column names become an ordinary first row and nothing in the file marks them as different from the data.
Anything visual in the Markdown - weight, alignment, colour, column widths - has no counterpart in TSV and is dropped. The values are what survives.
The identifier 007 comes out of the TSV as 007. Reading it as a number would have made it 7, and an id that changes value is worse than one that stays text.
The role "Analyst, data" survives with its comma, in one cell rather than split across two. That is the first thing to check in any converted table, and the usual place a TSV file goes wrong.
The quotes around Jonah "Jo" Pryce are doubled, which is how the delimited formats escape a quote in the TSV.