The Las Vegas strip
Public Policy Polling (PDF). 10/20-23. Nevada voters. MoE ±4.4% (7/28-31 results):
Shelley Berkley (D): 45 (43)
Dean Heller (R): 45 (46)
Undecided: 10 (12)
Back in January, when the disgraced John Ensign was still a senator, Heller led Berkley by a 51-38 margin. Now half a year of service in the upper chamber has taken a toll on Heller, who puts in his worst performance in PPP's four Nevada polls to date. From Tom Jensen:
Nevadans definitely don't seem to like Senator Heller as much as they liked Representative Heller. In January his favorability was 46/23 and in April it was 43/29. After his appointment to the Senate we started asking job approval and in August he was at 38/35 and this month he's at 39/35. It definitely doesn't appear that his Senate appointment gave him an unfair head start for a full term—in fact it could prove to be a negative in the end.
Tom also points out where the movement in the crosstabs is taking place:
On PPP's last poll, in late July, Heller had led 46-43. The main thing that's happened since then is Berkley has consolidated the Democratic vote. Previously she was getting just 75% of the vote from within her own party, but that's now up to 82%, pretty comparable to Heller's 83% of the Republican vote. On all 3 previous PPP surveys of the race this year, all of which Heller had held the lead on, he was getting a much larger share of the GOP vote than Berkley was getting of the Democratic vote. With that gap erased, so is his lead.
But I think this is the most important observation of all:
Berkley is leading 74-16 with Hispanics. That's pretty consistent with what Barack Obama got in 2008 and what Harry Reid got in 2010 with that group, and if Republicans can't do better with Hispanics it's going to make it very hard for them to win competitive races in the state moving forward.
Berkley only held a 52-39 edge with Latinos back in July, so this is a huge improvement. The difficulty, though, is that the sample size is super-tiny, maybe about 50 respondents, or 11 percent of the sample. But as Tom notes, the current spread makes sense. The key for Berkley will be locking that in, especially since exit polls showed Hispanics making up 15 percent of the electorate in 2008 and 16 percent in 2010—half again as big as the poll's sample. It's not unreasonable to expect they'll be at least as heavily represented if not more so next year, and that's good news for Team Blue, as long as we can keep this up.