Senate Races - N. Hampshire, N. Carolina, Alaska, Kansas

This chart is from a Rasmussen Poll -- done on September 10th and 11th of 750 likely New Hampshire voters, with a 4 point margin of error.

This chart was made from an American Research Group Poll -- done between September 12th and 15th of 544 registered New Hampshire voters, with a margin of error of 4.2 points.

This chart was made from a Public Policy Polling survey -- done between September 11th and 14th of 1,266 likely North Carolina voters, with a margin of error of 2.8 points.

This chart was made from a Harstad Strategic Research Poll -- done between September 7th and 10th of 709 Alaska voters (and the margin of error is not known).

This survey is from Public Policy Polling -- and was done between September 11th and 14th of 1,328 likely Kansas voters, and has a margin of error of 2.7 points.

Yesterday, I brought you a CNN Poll that said the New Hampshire senate race between Democratic incumbent Jeanne Shaheen and Republican Scott Brown was a dead heat -- with each having 48% of the predicted vote. I said at that time that the poll could be an outlier, since Shaheen has been leading in all the other previous polls -- and said we needed to look at some more surveys to know the truth.

Now we have two new polls on the New Hampshire senate race. The Rasmussen Poll gives Shaheen a significant 6 point lead, while the American Research Group Poll sets her lead at 2 points (which is within the margin of error of the poll). I think Shaheen does still have a lead in this race, but it seems that the race has tightened -- and it will require close watching in the next few weeks.

Public Policy Polling verifies what other polls have shown -- that Kay Hagan does have the lead in North Carolina. They put that lead at 4 points, which exceeds their margin of error. Also of interest is a new poll in Alaska that shows Democrat Mark Begich with a 5 point lead. I do think Begich has a small lead in Alaska, but I don't know how significant a lead it is (since the poll didn't give its margin of error).

Perhaps the most interesting race is the one in Kansas. Democrat Chad Taylor, realizing he can't win and the Independent Greg Orman has a chance, dropped out of the race. The Republicans have done their best to keep Taylor on the ballot, because they are scared of Orman -- and the case is currently awaiting a Kansas Supreme Court decision on whether Taylor's name will stay on the ballot or be removed. It may not make any difference though. A new Public Policy Polling survey has Orman beating Roberts by 7 points (41% to 34%) even with Taylor still a choice. This is a race that Republicans thought was a sure thing a few months ago -- and now it looks very likely they will lose this seat.