2016 Primary Map Thread

[Adam Griffin]

I started this a few days ago, but never got around to posting it. Tonight, I updated the map with the latest polling available. I wanted to do this because initially, I suspected that based on what we've seen in cross-tabs in many polls, that there would be a substantial difference in how the white vote for each candidate looked when compared to the overall primary polls.

Where applicable, the numbers shown are derived from aggregate polling. 15% in polling is the threshold for a candidate to appear.

There isn't a uniform method here for calculations, given the fact that every poll differs in the amount of information provided:

• In some cases, I have analyzed cross-tabs and found them to be inaccurate in terms of the share of the electorate when compared to the GE makeup of that state's electorate; as such, I've adjusted the numbers in these cases accordingly to more accurately reflect a presidential year primary make-up (assuming that the white share of the vote in all cases will be a slightly higher percentage in a primary than in the GE).
• Some polls do not provide sufficient cross-tabs with respect to turnout by race, in which case, I use the map linked above as a baseline for determining the make-up the likely primary electorate.
• Some polls may provide basic stats on the racial sampling of the polling, but may not provide breakdowns of each race's percentage of support for each candidate. In these cases, I used relevant polling data from other states and from national polls to project what the likely support by race is for each candidate in that state.

In all of these cases, once the preliminary figures are calculated, it becomes rather easy to calculate the likely white share of support for each candidate.


The first map in each series shows the percentage of support (color-coded) for the winning candidate in each state, as well as the results from one or more recent polls for other candidates who have at least 15% support in each state.

The second map shows the margin between the first place and second place candidate in each state.

9/18/15:



9/22/15:

[Adam Griffin]

9/23/15: TX poll added



Assumes Democratic primary electorate will be 39% White, 36% Black, 25% Latino; assumes Clinton is pulling 65% among Blacks and that Latino support for Clinton/Biden matches statewide support levels

[Adam Griffin]

Your last map seems to have removed the 9/22 states (e.g., Utah, Missouri...).

Crap - I edited the wrong one. Fixed.

[Adam Griffin]

The Elon NC poll has Hillary winning whites 45-34 over Sanders. No Biden, though.

That's quite helpful. It's very interesting to see how well that tracks with my Georgia and Texas projections, and to a lesser extent, my SC & FL ones as well - it gives me a bit more confidence that I am doing an adequate job with my projections (at least within the confines of the quality of polling). Clinton certainly isn't doing all that well with white voters in the Deep South.


I've updated my map format so that I can more easily fit the stats within the confines of most states.

9/24/15: NC & NH polls added

[Adam Griffin]

10/5/15: Added first polls for WI & LA; new TX & SC polls correct seemingly unrealistically-high Sanders performance from previous polls; Clinton closes the gap a bit in NH; Biden breaks 15% in IA & Clinton/Sanders IA margin remains the same

[Adam Griffin]

One of my friends off-site has been tracking the polls in each state and keeping track of the pledged delegate totals based on the current DNC allocation method and (obviously) the current polls. He asked me to make a map of it, so I thought I'd share. Here is a Google Fusion Tables map that shows the details state-by-state (click on a state for more info; red translucent states are those for which we have no data).

Democratic (Pledged) Delegate Tracker

Nominal:
Clinton 1730
Sanders 838

Two-Way:
Clinton 67.37%
Sanders 32.63%

Percentage of Pledged:
Clinton 45.90%
Sanders 22.23%

Percentage of All Delegates:
Clinton 38.59%
Sanders 18.69%