There is no doubt Harry Reid is in a pretty precarious position for reelection next year, but his situation may not be quite as bad as recent polling suggests.
That's because Nevada was the most poorly polled swing state in 2008 and the error all skewed in a Republican direction. Only one pollster (congratulations AP!) ever showed Obama up by his 12 point margin of victory. Only two others (Zogby and Suffolk) even came within four points of getting Obama's margin right and most (including us) were off by 7-8 points.
The three polling companies that have been most active in Nevada this year, showing Reid's vulnerability, have been Research 2000 (underestimated Obama by 7 last year) and Mason Dixon and Rasmussen (underestimated Obama by 8 points.)
I don't know if any of them have changed anything about their Nevada polling after last year, but if not it's possible that Democratic performance in the state is still being shortchanged by the polls.
Nevada is the only closely contested state I can think of last year where the pollsters were not only pretty universally off, but also pretty universally all off in the same direction. When we went back after the election to see where we went wrong it seemed like the culprit was significantly under measuring Obama's share of the Hispanic vote. Nevada is also one of the fastest growing states in the country and that could trip up pollsters as well, especially ones like us who sample based on past voting history.
Just something to keep in mind on the Reid polling.