Hey guys. So I have noticed that I have gotten a lot of traffic from Dailykos. I posted here a week ago to talk about the projections for New York. At Benchmark Politics, we take polling and apply it to a translation model that looks at demographics, education, and tons of other stats at the county level and creates county level expected results so that you can follow along on Primary/Caucus night and see how a candidate is doing early. Using this benchmark model, we can then make a call for who will win a contest with only 10-20% of precincts reporting in the majority of cases. This methodology has been extremely successful and accurate. Furthermore, the model itself has gained a huge amount of accuracy as we now are modeling based on 2k counties that have already voted. I won’t speak too much on the county projections, but you can find them here:
New York Final County-by-County Benchmark Guide - Clinton 57% - Sanders 43%
We wanted to make the actual modeling accessible in a format that all of you here at Kos can read easily, in diary form. I’ve been a member of Dailykos since the Obama convention speech days in 2004. Here we will focus on the city congressional districts, but you can read the entire state at the following link. You can use these numbers to plan and optimize canvassing, such as the case where you focus on congressional districts that are near delegate thresholds:
New York City with a link to New York State Congressional District Projections
I’ll lead in with this… This is NOT a Pro-Bernie or Pro-Hillary Diary. This is a pro-information and pro-access diary. Both campaigns can use this in their ground game to minimize the damage to their delegate count and maximize their damage to their opponent’s delegate counts. Think about it this way: If you are canvassing the Bronx, you may think twice about canvassing CD13 and 12 because they aren’t near thresholds, and instead may decide to canvass CD11 and 14 because changing their margins by 3-5% can result in a swing of 4 delegates just in these CDs alone!
That said, here is the actual guide:
----------------
Hello everyone!
(It has Congressional Districts 17-27)
Today, we have a special for you. We're going to do two articles on the congressional districts of New York. The reason why it is so important to focus on the congressional districts of New York is because New York allocates a large number of its state delegates based on who wins congressional districts. See the following:
New York Pledged Delegates
State Popular Vote Delegates: 54
Congressional District Delegates: 163
Statewide Pledged PLEOs: 30
Total: 247
As you can see, this means that the winner of each congressional district is literally three times as important as the winner of the state! To make matters tougher, the delegate allocation rules mean that candidates have to hit certain thresholds to achieve net delegates!
There are 3 kinds of delegate totals in New York:
5 Delegate Districts
3-2 Split - Winning candidate achieves > 50%, < 70% of the CD Vote
4-1 Split - Winning candidate achieves >= 70% of the CD vote, < 85%
5-0 Split - Winning candidate achieves >= 85% of the CD vote
6 Delegate Districts
3-3 Split - Winning candidate achieves < 58.4% of the CD vote
4-2 Split - Winning Candidate achieves >= 58.4% of the CD vote, < 75%
5-1 Split - Winning candidate achieves >= 75% of the CD vote, < 85%
6-0 Split - Winning candidate achieves >= 85% of the CD vote
7 Delegate Districts
4-3 Split - Winning candidate achieves >50%, < 64.3% of the CD vote
5-2 Split - Winning candidate achieves >= 64.3% of the CD vote, < 78.5%
6-1 Split - Winning candidate achieves >= 78.5% of the CD vote, <85%
7-0 Split - Winning candidate achieves >=85% of the CD vote
As you can see, this means that TONS of congressional districts will draw!
Now, lets look at the actual district breakdown of delegates, curtesy of the Green papers. Keep in mind CDs 1-16 are in the city proper and CDs 17-27 are in the State of NY:
One thing to note here is that outside of the City, districts 17-27 have 5 uneven delegate districts. In the city, there are only 3. This bodes well for Sanders, partly, because of his better performances in the state and the fewer blowouts expected there. However, a 6 delegate district can be very good for a candidate expecting blowouts, because once reaching the lower end of the threshold, they get two net delegates for a win (4:2) instead of one (3:2).
With no further adieu, here are the projections for the State of New York using our benchmark models.
Please keep in mind - We do not stand as strongly on these as we do our county projections. We have never done CD projections and because we don't have the granularity of precinct level data, there is a lot of reasoned guesswork here using our benchmark numbers and weighting them/combining them, due to the fact that congressional districts cut through counties.
For the State Portion
(it has the writeups for CD's 17-27)
For the City Portion
Scenario 1: Expected Result from a 57% - 43% Victory
Expected Delegate Result from the State (non-city) of New York
Clinton 33 Delegates - Sanders 32 Delegates
Expected Delegate Result from the City of New York and Surrounding Areas
Clinton 60 Delegates - Sanders 38 Delegates
Expected Statewide Pledged Delegates Assuming 57%-43% Victory
Clinton 48 Delegates - Sanders 36 Delegates
Total Delegate Score:
Clinton 141 Delegates - Sanders 106 Delegates
Clinton +35 Delegates
Scenario 2: Result from a 57% - 43% Victory
(Where SANDERS BENEFITS from all close to threshold CDs)
Expected Delegate Result from the State (non-city) of New York
Clinton 29 Delegates - Sanders 36 Delegates
Expected Delegate Result from the City of New York and Surrounding Areas
Clinton 56 Delegates - Sanders 42 Delegates
Expected Statewide Pledged Delegates Assuming 57%-43% Victory
Clinton 48 Delegates - Sanders 36 Delegates
Total Delegate Score:
Clinton 133 Delegates - Sanders 114 Delegates
Clinton +19 Delegates
Scenario 3: Result from a 57% - 43% Victory
(Where CLINTON BENEFITS from all close to threshold CDs)
Expected Delegate Result from the State (non-city) of New York
Clinton 35 Delegates - Sanders 30 Delegates
Expected Delegate Result from the City of New York and Surrounding Areas
Clinton 65 Delegates - Sanders 33 Delegates
Expected Statewide Pledged Delegates Assuming 57%-43% Victory
Clinton 48 Delegates - Sanders 36 Delegates
Total Delegate Score:
Clinton 148 Delegates - Sanders 99 Delegates
Clinton +49 Delegates
Congressional District: 16
Counties/Cities of note: Northern Bronx, Southern Westchester
Number of Delegates worth: 6
Expected combined spread: Clinton +32%
Clinton Delegates from Spread: 4
Sanders Delegates from Spread: 2
CD 16 was Obama's best CD of any state in 2008, where he won 95%-5%. This CD will be very tough for Bernie Sanders, and is on the border of being a 5:1 delegate split. If Clinton can manage +50 (75%-25%), she will achieve this threshold. Part of the reason why it is so difficult for Sanders here is because it includes some of the heaviest minority areas of the Bronx and all of Yonkers. Southern Westchester should remain more Sanders friendly, hence why the margins are not as high as if this were just the Bronx alone.
Congressional District: 15
Counties/Cities of note: The Bronx
Number of Delegates worth: 6
Expected combined spread: Clinton +60%
Clinton Delegates from Spread: 5
Sanders Delegates from Spread: 1
This is the absolute worst area for Sanders in the entire state. The Bronx is, by itself, benchmarked to go to Clinton about 78%-22%. However, this particular CD cuts out all of the non-minority area of the Bronx and is almost entirely made up of minorities. The college degree % in the area is very low, and the demographics very, very favorable to Clinton. Obama won 96% of the vote here in 2012, so he is clearly very popular in the area. We believe that Clinton will have a fantastic showing here. Sanders viability will be threatened in this CD. Watch it carefully for a potential 85% showing for Clinton, leading to a 6:0 split.
Congressional District: 14
Counties/Cities of note: Eastern Bronx, North Central Queens
Number of Delegates worth: 7
Expected combined spread: Clinton +20%
Clinton Delegates from Spread: 4
Sanders Delegates from Spread: 3
CD14 is going to be kinder than CD15 and CD16 to Clinton. Queens county is not expected to be particularly bad for Sanders, compared to many other parts of the city, although it's still a "Clinton heavy" county. Likewise, this area has a section of some of the more favorable parts of the Bronx. It certainly is still a CD that leans strongly to Clinton though. What is important to watch here is how close the numbers are to a 5-2 delegate split. We have Clinton winning 60% here. If she can get to 64.3% here, then she will achieve that coveted 5-2 split, valuable in the 7 delegate districts. This will depend largely on how she does in Queens. If we notice Sanders winning in Queens or keeping it close (within 5%), then it is going to be impossible for her to get that 5-2 split. However, if we notice Clinton really running away with it in Queens, >65%, then the Bronx will by far ensure that she stays over the mark to get 5 delegates here.
Congressional District: 13
Counties/Cities of note: Upper Manhattan, Western Bronx
Number of Delegates worth: 6
Expected combined spread: Clinton +32%
Clinton Delegates from Spread: 4
Sanders Delegates from Spread: 2
District 13 is bad for Sanders, but because it is a 6 delegate district, Clinton isn't likely to get a 5-1 split here - she would have to achieve 75% of the vote. We actually have manhattan as a whole going for Sanders, but District 13 includes the western Bronx which is VERY Clinton heavy, and the areas of Manhattan it includes are the minority heavy ones, and not the areas that are more favorable to Sanders. We don't actually anticipate this district being extremely contested.
Congressional District: 12
Counties/Cities of note: East Side Manhattan, Greenpoint, Western Queens
Number of Delegates worth: 6
Expected combined spread: Sanders +8%
Clinton Delegates from Spread: 3
Sanders Delegates from Spread: 3
This is one of Sanders best districts in the New York City area proper. It has a very good racial demographic setup for him, and it includes his strongest areas in Queens and Manhattan. That said, the district itself also includes some of the richest areas in the entire United States, with a per capita income of $75,000... that is not median income, but per capita, which means that this accounts for every man, woman, and child in the district. These wealthy voters have given Sanders trouble in the past. But still, the area is favorable for him demographically and if he is going to win any CD in New York City, this is likely to be the one. However, it is a 6 delegate contest, and in order to net delegates, Sanders needs to achieve 58.4% of the vote here. With a strong minority population and an extremely wealthy population, we don't see it happening. Likewise, Clinton has room to overperform here if wealthy voters end up driving the vote in the area, but even if she does, it isn't likely she runs away with net delegates. This is a high uncertainty CD.
Congressional District: 11
Counties/Cities of note: Staten Island, Southern Brooklyn
Number of Delegates worth: 5
Expected combined spread: Clinton +5%
Clinton Delegates from Spread: 3
Sanders Delegates from Spread: 2
CD11 is a true tossup county. Though the map shows it a a Sanders lean area, it overlaps with Brooklyn, which is a Clinton strong area. Looking at the actual benchmarks for the area, Clinton's got the advantage, but either candidate can win Staten Island. On Clinton's side, the area is high income and slightly older, but on sanders side, it is much less diverse than the rest of the city. It is an important CD to watch, too, given that it is one of the few odd-number congressional districts in the city area. Whichever way Staten Island goes is likely the way this county goes.
Congressional District: 10
Counties/Cities of note: West Manhattan, South Brooklyn, Wall Street, Greenwich Village
Number of Delegates worth: 6
Expected combined spread: Sanders +10%
Clinton Delegates from Spread: 3
Sanders Delegates from Spread: 3
Demographically, this is the best area for Sanders in the New York City area. That being said, this is one of those few areas where issues have a stronger chance of trumping demographics. After all, Wall Street is located in this CD and some very wealthy parts of New York City are contained in CD10. Our benchmark model has Manhattan going FOR Sanders because of its overwhelming number of bachelors degrees (60%!), it is one of the youngest areas in New York, and the median income is actually NOT as high as you'd think (69k, which means half of all income earners earn LESS than that). These factors, as well as a relatively not as diverse racial breakdown here, suggest that Sanders could be competitive here. If he is not, it is likely New York will be a blowout greater than polls suggest. However, if current polls are correct, Sanders needs to win SOMEWHERE in New York City - his margins in upstate/Buffalo/Syracuse are not nearly enough to justify his current polled support without some areas of the city going for him. So we'll peg this as his place he's supposed to do well, demographically speaking. This could easily go HEAVILY the other way, due to Wall Street. All of that being said, it being a 6 delegate district makes it highly unlikely that it is anything more than a draw, barring incredible margins on either side.
Congressional District: 9
Counties/Cities of note: Parts of Brooklyn (Kings County), mainly Brownsville, Crown Heights, Flatbrush, Kensington, Park Slope, Prospect Heights, Prospect Lefferts Gardens, Prospect Park, Grand Army Plaza
Number of Delegates worth: 6
Expected combined spread: Clinton +55%
Clinton Delegates from Spread: 5
Sanders Delegates from Spread: 1
CD9 is majority African American and low income. We already have Kings county going to Clinton in a landslide, but this part of Kings County is even more black than the county as a whole. The district will be a blowout, the only question is by how much - Clinton needs to get above 75% here to secure her 5:1 split otherwise this becomes a 4:2 split district. We think she'll be able to - the entire county is already expected to go to her at high margins. This region though is in effect a concentrated area of her support from Kings County. Still, a little bit of outreach from Bernie here could go a long way. If he could knock Clinton's margins from 5:1 to 4:2, he takes a 4 delegate loss and turns it into a 2 delegate loss here, for only a meager 5% decrease in her vote share here.
Congressional District: 8
Counties/Cities of note: Portions of Brooklyn and Queens
Number of Delegates worth: 6
Expected combined spread: Clinton +60%
Clinton Delegates from Spread: 5
Sanders Delegates from Spread: 1
CD8 is very similar to CD9, but even worse for Sanders. The area is moderately educated, but is also majority female to a large degree (nearly 53%) While CD9 included SOME parts of Brooklyn with white voters, CD8 is made up of almost entirely black voters, with a large contingent of Hispanic voters as well. Many polls have shown Clinton taking a commanding 65-75% lead among Hispanics, and the model picks up on the lack of Bernie demographics in this area to give Clinton a commanding victory. We have her well over the threshold to get 5-1 here, and she is actually quite close to potentially making Sanders not viable in CD8. We'll watch closely.
Congressional District: 7
Counties/Cities of note: Portions of Brooklyn, Queens, and Manhattan
Number of Delegates worth: 7
Expected combined spread: Clinton +20%
Clinton Delegates from Spread: 4
Sanders Delegates from Spread: 3
CD7 is one of the few in the city that is worth 7 delegates. Sanders does make some inroads due to a relatively young median age in this area of the city. It has a large Hispanic presence, but it also has a decent number of Non-Hispanic whites. It does pick up some friendlier margins for Sanders from Manhattan and Queens, but Brooklyn is still overall an unfriendly place to Sanders. CD7 is right on the cusp of being a Hillary 5:2 type CD. When returns come in, watch for close margins in Manhattan for Sanders to probably keep CD7 from being a 5-2 blowout. This will be a contested margin for sure.
Congressional District: 6
Counties/Cities of note: Entirely within Queens
Number of Delegates worth: 6
Expected combined spread: Sanders +10%
Clinton Delegates from Spread: 3
Sanders Delegates from Spread: 3
CD6 is a very friendly area to Sanders. It is extremely Asian, and this population has been voting mainly with the primary demographic in this area. The good news for Sanders is that the area is white, middle class, and fairly ell educated. The bad news is that the area is rather older than the rest of the city. Clinton shouldn't get blown out here, but Sanders has a real shot at a 4:2 split if he can achieve 58%.
Congressional District: 5
Counties/Cities of note: Mostly Queens, some Nassau
Number of Delegates worth: 6
Expected combined spread: Clinton +52%
Clinton Delegates from Spread: 5
Sanders Delegates from Spread: 1
Sanders faces a double whammy in CD5. From the Nassau side, he's got a median household income of $97k, one of the highest in the entire country, and a median age of 41 years old. From the Queens side, he's working with the one of the highest % black CDs in the city area. In 2012, this district voted for Obama 90.6%. That should show you how strong the black vote is here. We expect Clinton to pile on huge margins here. Watch Nassau though - if Nassau looks to be potentially a Sanders county, it may blunt the impact of Queens enough in this district to bring it down to a 4:2 split, which this projection is only 2% away from.
Congressional District: 4
Counties/Cities of note: Central and Southern Nassau
Number of Delegates worth: 6
Expected combined spread: Clinton +10%
Clinton Delegates from Spread: 3
Sanders Delegates from Spread: 3
CD4 is Nassau, extremely wealthy and white. Though the race is expected to be fairly close in Nassau, this section of Nassau has a sizable Hispanic and Black population and is fairly wealthy. We give Clinton a slight edge, but it won't be nearly enough to push her over the 58.4% mark for it to be a 4:2 split. We don't anticipate this going any differntly, but if Nassau surprises and comes in at Clinton +20%, you'll know that this CD and a few others will be heavily affected.
Congressional District: 3
Counties/Cities of note: Central and Northern Nassau, some Suffolk
Number of Delegates worth: 7
Expected combined spread: Sanders +10%
Clinton Delegates from Spread: 3
Sanders Delegates from Spread: 4
CD3 is one of those areas where Sanders gets a nice benefit of an uneven delegate allocation district. Nassau is expected to be a close contest, but one area of Nassau - South Nassau - is much friendlier to Clinton than North Nassau. Sanders is expected to earn the same margin here that Clinton does in South Nassau, but will likely get a delegate out of it because it is uneven allocation. There is a chance Clinton could pull this one off, given that the area is old and high income.
Congressional District: 2
Counties/Cities of note: Suffolk, some Nassau
Number of Delegates worth: 6
Expected combined spread: Clinton +2%
Clinton Delegates from Spread: 3
Sanders Delegates from Spread: 3
CD2 is not expected to be worth anything for either candidate. Neither Clinton nor Sanders is expected to run away with it. The demgraphics are tepid for both. For Sanders, the area is too old and rich to be a runaway and for Clinton, it has far too high a bachelors degree percentage and not nearly enough minorities. Because it is a 6 delegate CD, a candidate needs to get almost 20% on the other candidate to get a delegate out of here. If Suffolk goes 20% in either direction, then you can expect CD2 to net someone a delegate.
Congressional District: 1
Counties/Cities of note: Suffolk
Number of Delegates worth: 6
Expected combined spread: Sanders +1%
Clinton Delegates from Spread: 3
Sanders Delegates from Spread: 3
Just like CD1, CD2 is not expected to be worth anything to either candidate. Its demographics are lukewarm to both. Watch for a candidate to get somewhere around 58.3% in Suffolk for CD1 or CD2 to go either candidate's direction, but the demographics are in direct opposition to any runaway in either of these districts.
I hope everyone got something useful here. We’ll continue to offer a diary or two a week at DailyKos from our website and our analysis of the race, because to be frank, I owe it to all of you. I got my interest in politics as a teenager BECAUSE of DailyKos. It is the reason I am a Democrat to begin with. Think of it as my way to give back to the community.
May the best candidate win this weekend.
You can always follow us for around the clock updates @benchmarkpol on twitter!