Don't get me wrong, i'm not disagreeing with the basic idea here, but there is one thing a major league starting pitcher cares about when he takes the mound, and that's securing the win. If they give up 5 runs and win, they are happy. If they give up 2 runs and lose, they are mad. If they have a 10-run lead in the third, they don't care about giving up 4 solo home runs on 2-0 counts and leaving the game in the 7th with an 11-4 lead. That's a great outing in their mind. If he loses 2-1, he's mad at himself for the mistake pitches he made and doesn't think he was good enough.
I'd be willing to wager that 98% of pitchers have no clue what FIP or ERA+ is, let alone what these numbers are for them at the moment. But I bet most of them can instantly tell you what their W/L record is. Because that's the end game when they step on the mound.
The pitcher quoted above pitched a fantastic game, easily his best to that point in his season, but lost a no-hit bid late. His team won, but they were not ahead when he left the game. To him, he did not do his job.
W/L record is pretty far down the list of things you should use if you are breaking down a pitcher, but to say it doesn't matter at all is ignoring the nuances of the game. More often than not, a starting pitchers W/L record is a close approximation of what they "deserve". If you disagree, show me the list of 300 game winners who were bad pitchers, show me the pitchers with winning percentages over .530, .540 over a sustained period of time who were bad pitchers, show me the good pitchers who were losers over sustained periods of time. You may find a few outlier examples, but the fact of the matter is the larger the sample size, the more likely W/L record is a close approximation of what a pitcher deserves. Focusing on small sample sizes like a rough 1-0 loss or Ivan Nova 2011 is kinda silly, but quite frankly, despite the annoying Yankee fan talk, Ivan Nova did his job last season.