What a bizarre question.
I like paul s-306273's comment about its relevance to life, the universe, or anything and think the anser is (to paul's question) is none. But I have no objection to irrelevant questions, they can be quite appropriate trivia to entertain when posed as QoTD.
I agree with Hugo's comment: there are no "rows" in a query plan. In fact that thought had tempted me to answer "0". But I took a wild guess that the "rows" were the things labelled in the estimated plan as queries, and we know from last time approximately this same question was asked that there will be 8 such, so I picked that wrong answer instead of the 0 wrong answer.
When I saw the explanation I thought "someone has screwed up here - the behaviour reported for SSMS / SQL 2008+SP3 is crazy" so I tried actually running it. I don't get anything but "the commands completed successfully" in the message tab when I run this in SSMS / SQL SP2008 R2+SP2. I've tried all sorts of config changes to try to make it give me a row count message, but nothing worked. If I add a select for a single row before the comit statement, I do (of course) get a row count in the message tab - and that says one row is affected; that seems to me to imply that the other lines - at least for that version of SQL Server - affect no rows, since there's no imaginable way that adding that select can prevent a row being affected that would have been affected but for the presence of that select.
So either I am missing something, or there's been a change in behaviour between 2008+SP3 and 2008R2+SP3.