Wow, this is horrible reporting. Sure, it's hardly the worst example out there, but two paragraphs in this article about Specter's primary loss really jumped out at me:

Specter is 80, but for Democrats in 2009 he was the new 60, the parliamentary crowbar Democrats needed to help muscle President Obama's policies past Republican opposition in the Senate... With Specter's loss, expect a major storyline Wednesday to be whether Obama has lost his golden political touch. His track record of late has not been so good.

Right, because someone who has been a Democrat for less than a year getting primaried and defeated from the left reflects badly on Obama. It doesn't even make any sense on its own terms; this isn't Obama's first defeat of any kind, this is way, far from Obama's biggest defeat, and this isn't a defeat at all but a positive victory for the left or liberals or whatever. If Sestak wins the general he will probably vote with Obama more than Specter did, it's too early to say whether he will win the general, and the article doesn't even try to speculate on that. I suppose there could be some kind of double-reverse-judo in this article intended to drum up support for Obama by making him appear centrist because he's opposed by the left too, but that's giving the writer too much credit for intelligence. If I was still a reporter and was trying to write an article like that or was getting input from my edit pushing me in that direction, my brain would melt out my ears.

And "expect a major storyline to be" means, of course, "we're making it a major storyline, and fuck you if you care if it's true."

Also, you know what's missing from that article? RESULTS. Sestak beat Specter 54-46, but I found that out on Wikipedia, which cites Politico. Neither the CNN article I was just complaining about, nor the one summarizing all the primaries, had those numbers in the articles. Sure, CNN probably has them somewhere, but if they aren't in the article, the site doesn't deserve the traffic.