Two of the leading pollsters this cycle (using cells) in terms of accuracy both delivered a statistical tie in Nevada. CNN’s latest is 48%/47% Clinton leading Bernie in a poll with a marginally high 6% (MOE). Qunnipiac’s latest is 44%/42% Clinton leading with a very reasonable 4.12% (MOE).
Qualitatively speaking this is a caucus system like Iowa; where you can negotiate for your candidate before forming 15% viability groups. Because O’Malley is long-gone neither candidate will have a problem reaching viability here. There where a few very small rural areas where this was a problem for the leading two contenders. (As a side note, there is same day registration here which created 30,000 new voters in 2008).
The majority of the 2008 exit polling data is pretty useless since Nevada is a very transient state along with having a housing crisis. The one key point is that about 2/3 of voters are located in Clark County (Nevada/suburbs.) Obama lost this county by 10% points in 2008; therefore losing the state by 5% because there are not enough voters in remaining counties to make up the deficit.
As someone whose worked on campaigns (including a congressional one) that went into re-count; this has only two scenarios 1- race is decided by +/- 2 points 2- Benie wins by 3-6 points on Caucus day. Because this is a lower turnout event; the swing in new voters can make a significant upshift in Sanders support. I don’t think a 3rd outcome (w/Clinton) winning by many points is possible because she started with a 25+ point lead. (I don’t know of a case of a candidate losing a 25 point lead (only to regain it by a good amount at the last minute).
Key Figures & Trends (pollsters have different screening methods)/both use cells. For this analysis won’t back calc N/A number with CNN data (would mess up their calcs)
Qunnipiac- 44% Clinton /42% Sanders 11% undecided
CNN- 48% Clinton/47% Sanders 5% Undecided (doesn’t provide makeup of voters by demo -need to backcalc numbers)
key figures to parse for your own analysis.
Definitely decided numbers
CNN- 62% / Q-polling- 67% (shows Clinton 64%/Sanders 67%). CNN shows 67% white folks are decided. Doesn’t show non-white figures. But indicates more non-white voters are undecided.
. Key factor here favors Sanders slightly since candidates have to form favorability groups.
Q-Polling /CNN
Dems 47% Clinton Sanders 39% (76% of electorate) / CNN- 53% Clinton Sanders 41% (doesn’t show electorate)
Indies 53% Sanders 33% Clinton (24% of electorate) / CNN doesn’t show figures Key factor here moderately favors sanders because of same day registration. Massive factor in a caucus setup on a Saturday.
Ethnic makeup (CNN shows very little here).
Q-polling 63% White /37% Non White
White- Bernie 50/38 (9% undecided) — Hillary has a white person problem.
Non-white Clinton 55/30 (14% undecided) Bernie is down, but has plenty of room to grow here.
Clark County (Culinary union won’t endorse Clinton key here)
CNN- 48% Clinton/47% Sanders (matches topline) (2008 Clinton 54%/Obama 44%) Massive shift from 8 years ago showing Clinton has flatlined in so many areas of the country. Outside southeastern African American areas; Clinton has no real protection.
Gender
Qpolling — 44% Male/56% Female (some indication Soft Hillary supporters didn’t pass the screen).
Economy
CNN — Sanders 50%/47%
Honest/Trustworthy
CNN 87% Sanders/60% Clinton
Prediction - The indication here is that every (one to three /one to four) voters are still deciding to vote for on Saturday. Voters have had 25 years to decide about Hillary; therefore Sanders has all the momentum here in terms of all voters.
Same day registration, Saturday/caucus voting, high undecideds, 25 point erased lead, Culinary union (not endorsing Clinton), & 25 year national scene candidates --gives Bernie all the momentum.
Bernie Sanders 52
Hillary Clinton 48
CNN internals
i2.cdn.turner.com/…
Q polling internals
www.quinnipiac.edu/…
Nevada Voter Registration
www.nvsos.gov/…
2008 Dem NV Results
en.wikipedia.org/…
www.nytimes.com/…
2008 exit polling —
www.cnn.com/...