Iowa Caucus Results Thread (pg 148 - full results)

[jimrtex]

Finally, the 2020 election begins! Aren’t we going to be getting results from Tbilisi at noon? Or will those not be available until the caucus sites close within the state itself? Also, what county/CD will those be counted in?

Each satellite caucus forms a virtual precinct. They allocate county delegates, but don't actually choose county delegates. Participants may volunteer to serve as state delegates (state delegates serve at both CD conventions and the state convention).

The satellite caucuses in each CD form a virtual county. No county convention is held. The number of allocated delegates for each presidential preference can be totaled, viability determined, viability applied, and the number of state delegates calculated. The presidential candidates will choose the actual state delegates from the volunteers.

The out-of-state caucuses will also form a virtual county, but they will only choose delegates to the state convention.

The number of county and state delegates is dynamic based on attendance. More attendees, there will be more county and state delegates.

Logically, there would be reports for 99 actual counties, and 5 virtual counties.

[jimrtex]

How many Iowans are there in Tbilisi anyway?

I'm really not sure either. I almost wonder if it was just so they could say there was a caucus in the Caucuses. Maybe one of the universities have a study abroad program there?

The other two international caucus locations make a bit more sense, Glasgow and Paris, but I also would have expected a few other places like London, Rome, etc.

Originally it was going to be at 7 p.m. Iowa time (UTC-6). Since Georgia is UTC+4, that would be 4 a.m. local time Tuesday. I was encouraging Tender Branson to find an Iowan in Vienna, and carpool to Tblilsi.

The Tbilisi caucus is hosted by a correspondent who covers Turkish, and former USSR. Previously he lived in Istanbul (or maybe Ankara).

The Glasgow caucus is expected to have up to 10 participants. The organizer originally anticipated 6, but that has increased with some driving up from London.

The Paris caucus is expected to have 26. It is held at a dorm for US students in Paris. The organizer is a 20 YO student from Iowa.

Organizers had to apply, and be screened. I suspect there was little publicity of the program. Someone in Iowa might have heard about it, and suggested it to someone elsewhere.

Originally, the IDP wanted to permit absentee voting. The DNC (or members from states jealous of Iowa going first, said that would make it a primary, and vetoed the plan).

[jimrtex]

I think Bill Weld will run around 4% on a good night, and the President will get 93/94%. Trump will obviously get 100% of the delegates.

The Republican caucus is fully proportional. Weld could actually get a delegate or two.

[jimrtex]

Why is there an in-state satellite caucus? Is it just for local people who won't be able to attend tonight? If so, why does this group seem to almost uniquely have this type of allowance?

Most are in Iowa.

14 are work-related (typically for those working the swing shift). Many of these are closed to union members.
24 are college-related. Many of the out-of-state caucuses are so-classified, but there are also some in Iowa. Ordinarily you are supposed to caucus where you are registered to vote. There are several at community colleges, perhaps targeted towards those who have evening classes.
29 are accessibility-related. Many are at nursing homes or assisted living or senior centers. These often have an earlier time.
11 are for language and culture groups (e.g. South Sudanese Center in Des Moines).
9 are for wintering Iowans (4 in Florida, 4 in Arizona, 1 in California).

[jimrtex]

Why is there an in-state satellite caucus? Is it just for local people who won't be able to attend tonight? If so, why does this group seem to almost uniquely have this type of allowance?

What, people who work the night shift at meat packing plants should not be able to vote?

These satellite caucuses are an improvement over the previous system and make it a bit fairer (a bit less biased to wealthy old whites who are more likely to be able to caucus on a weekday evening), although IIRC they still give less delegates than the regular caucus.

People working night shift should be able to vote, and they shouldn't have to go through the archaic Iowa caucus system to do it. But everything I've read about this anti-democratic process has led me to believe that such exceptions weren't really an option. Just from a definitional sense, calling an in-state early caucus a satellite caucus is confusing.

Also, not really the point, but it's debatable that this is an improvement over the old system. If these people truly couldn't have participated otherwise (which seems to be the case here) then it does include more people in the process. However, the reason that the virtual caucus was canceled was because people feared it would have pulled attendees away from the actual events and, in the process, weakened their representation when it came to total delegates. Same applies here.

The main reason the virtual cacuses were scrapped was the Iowa Democratic Party could not prove that they were secure from hacking.

The real reason was that later states weren't willing to approve other forms of voting.

[jimrtex]

Don't mix up your delegate types.

11,402 county convention delegates are awarded proportionally on the basis of the results of the 1678 precinct caucuses and 99 satelite caucuses, and they will go to their local county convention on March 21, 2020, to choose 2,107 District and State Delegates (SDE) for the district conventions on April 25 (selecting the 27 pledged congressional district delegates) and state Democratic convention on June 13 (selecting the remaining 9 pledged at-large delegates and 5 pledged PLEO delegates). In total, 41 pledged delegates will hereby be elected for the 2020 Democratic National Convention on the basis of the State Delegate Equivalents (SDE).

https://en.wikipedia.org/wiki/2020_Iowa_Democratic_caucuses#Procedure

so how do you qualify for a debate via delegates? yang klobuchar buttigieg?

That would be national delegates.

The caucuses tonight are electing delegates to county conventions. The county conventions will choose state delegates to CD and state conventions. But the county delegates are bound to tonight's results, so that the allocation of state delegates will be determined by tonight's results.

So imagine there is a small county with 40 county delegates, that chooses 3 state delegates. There are 3 Yang delegates, 10 Biden, 13 Warren, 7 Sanders, 7 Klobuchar.

Yang gets 3/40 x 3 State Delegate Equivalents (SDE)
Biden gets 10/40 x 3 SDE
Warren gets 13/40 x 3 SDE
Sanders gets 7/40 x 3 SDE
Klobuchar gets 7/40 x 3 SDE

Note that there is no viability test at the county convention, the delegate support is just passed through. The SDE are used to determine the number of national delegates elected at the CD and state conventions.

This small county convention will elect 3 state delegates, one Warren, one Biden, and perhaps a coin flip between Klubuchar and Sanders. They will not be active in determining the number of national delegates, but will decide who the national delegates. So the Warren state delegate would join with the other Warren state delegates to choose who the Warren national delegates are.

[jimrtex]

At the Florida caucus site, all the Warren people have Florida for Warren signs...

But this is supposed to be an IOWA Caucus.

If these people are Floridians with Florida signs rather than Iowa signs, why are they voting? Surely they should have IOWA for Warren signs. Sad.

"In the crafts room we have some butter sculptures by Bessie Shnauphauser"

Caucus goers jump out of their wheel chairs, throw down their crutches, hurdle their walkers, as they rush off to see the butter sculptures.

"But first we have to vote!"

People slump back into their wheel chairs, heads on their chest. Others hold onto some object while their canes or walkers are brought to them.

The imposters were the ones who remained seated throughout.

[jimrtex]

Second round Port Charlotte results:

56 Klobuchar
43 Buttigieg
33 Biden

Biden picked up zero votes

This should result in delegates to the out-of-state virtual county of:

4 Klobuchar
3 Buttigieg
2 Biden

[jimrtex]

https://twitter.com/ndrnicholas/status/1224466464234713094
Wow, big result for Sanders in Des Moines.

A question about the rules: if there are 70 voters here, then the cutoff for staying in the game is 11. Buttigieg is at 8, so under the cut-off, but not so far under that he couldn't survive if lower-ranked candidates' supporters were allowed to switch to him. Do Biden/Klobuchar/Yang have the opportunity to support Pete, or are all four candidates under the threshold immediately eliminated, getting to choose between Warren and Sanders?

Viable groups are locked in. They can even turn in their ballot card and go home.

Members of non-viable groups can:
(1) Join a viable group;
(2) Join a non-viable group to make it viable;
(3) Form a new viable group (this would probably be an uncommitted group).

Tactically, the best move would be to join Buttigieg to make him viable, which would keep Sanders from getting all the delegates.

[jimrtex]

Fun fact:

Here in Austria (9 million people), all votes are counted in 2 hours.

And then it takes another four months to determine who was elected.

[jimrtex]

Why would Warren lose so much (and Biden gain so much) in translating the final vote into SDEs?

The number of caucus participants varies quite a bit across the state. Caucus turnout is particularly high at college campuses, and urban areas, lower in rural areas.

State delegates are awarded to counties based on the Democratic vote for president in 2016, and governor in 2018.

In 2016, caucus turnout was 170K, while Clinton received 650K votes. So roughly 26% of Democratic voters caucused (ignoring for effects of Democratic caucus goers not voting for Clinton).

But in Johnson County (Iowa City/UI) and Story County (Ames/ISU) the ratio was around 40%. It was elevated in more urban counties, and depressed in rural counties (more around 20%).

So imagine two precincts with 1000 Democratic votes in 2016/2018. They would each elect the same number of state delegates (indirectly, which is what SDE are).

One has 30% turnout (300) and the other 20% turnout (200).

Now imagine Warren gets 20% of the vote in Precinct A (60 votes) and 15% of the vote in Precinct B (30 votes). Overall she has 90/500 = 18%.

Now imagine Biden gets 15% of the vote in Precinct A (45 votes) and 20% of the votes in Precinct B (40 votes). Overall she has 85/500 = 17%.

But they would get the same number of delegates.

There could also be selection biases. I'm not saying the Sanders campaign was cherry-picking the data, but they were able to get results from only about 2/5 of precincts. Perhaps they were getting reports from precinct captains where Sanders did well. If your precinct gave him 35% wouldn't you want the Sanders campaign to know? If Sanders was not viable in your precinct, you would be more reluctant to report the result. You might have gone home. If Sanders was not viable, it would likely mean that Warren was.

[jimrtex]

Can someone sum up the latest ridiculous Bernie Bro conspiracy theory for those of us not plugged into our tinfoil hats?

The Buttigieg campaign directly donated to the developers of the Iowa app. Something that's a bit sketchy, but largely overblown. Judging by the past few months, if the DNC is rigging the election for Pete, they aren't doing a great job at it.

Some other Twitter brains on my side think the DMR spiked the Selzer poll to rat**** Bernie, which is utter lunacy. Most of everything that happened in Iowa is better explained by incompetence.

The Buttigieg campaign purchased services from the company. The IDP purchased services from the company.

[jimrtex]

Why should anyone care who won the most “SDEs”?   We might care about who won the most delegates to the national convention, since this actually determines who gets nominated.  And we might care about who won the most votes, since early states are mostly about momentum.  But why should the public pay attention to some abstract, technical, and interstitial step like SDEs?

National delegates are awarded based on SDE's.

If you get 30% of SDE's you would get roughly 30% of the national delegates.

A small county, Adair for example, has 51 delegates to its county convention, and 4 delegates to the state convention. A large county, Polk for example, has 1401 delegates to its county convention, and 392 delegates to the state convention.

The state convention delegates are proportional to the presidential vote in 2016, and gubernatorial vote in 2018.

119804:1133 :: 392:4 (2016 presidential)

120257:1167 :: 392:4 (2018 gubernatorial)

While the state delegates are proportional to the popular votes, the county delegates are not. Small rural counties have relatively larger conventions.

If they were proportional, then Adair might have 14 persons at its county convention. They could insert a couple of leafs in someone's dining room table and meet their, or use a classroom at NVHS and sit at desks in a classroom. Alternatively, Polk might have 5000 persons at its county convention. You are going to have find a facility large enough. You don't want to interfere with the state basketball tournament, or Trump would male West Virginia look like a squeaker ("We are going to cut all the basketball players and basketball coaches (chopping motions), right Tim? (smirk)" And the press would miss the nuance that they would be replaced with apprenticeship programs installing wind farms. 1400 is more manageable, while still getting people to fill like they are an important part of the party. Some might eventually become legislative candidates, etc.

So you could just report the number of county delegates. But then 10 Biden delegates in Adair equals 10 Biden delegates in Polk.

So instead, the county delegates are weighted by the number of state delegates.

Remember Adair had 4 state delegates chosen from 51 county delegates. Alternatively, each county delegate is worth 4/51 of a state delegate.

10 Biden county delegates in Adair are worth 10 x 4/51 = 0.784 SDE.

In Polk 392 state delegates are chosen from 1401 county delegates. Each county delegate is worth 392/1401 of a state delegate. 10 Biden county delegates in Polk are worth 10 x 392 / 1401 = 2.798 SDE.

SDE may have been invented by the press to normalize the results from different counties.

In the distant past (2016), county delegates would choose the state delegates. The 51 delegates in Adair would meet. Based on the party rules those 10 Biden delegates might have been able to elect one Biden delegate to the state convention. Those 0.784 Biden SDE would become 1.000 Biden state delegates.

On the other hand if there were only 10 Biden delegates among 1401 county delegates in Polk, those 2.798 SDE might convert to 0.000 state delegates. And there could be sleeper delegates (BINO's and SINO's, and WINO's).

In 2020, the SDE pass right through the county convention and are used at the CD and state level to allocate the national delegates. The county conventions will choose the state delegates, and the state delegates will choose the national delegates, but not the apportionment of those delegates.

If the SDE were available, we would know the number of pledged delegates from Iowa.

[jimrtex]

https://twitter.com/bhalle87/status/1224558925946793985?s=20
Pete’s Comm Director just casually tweets out the login pin.

We should nullify the results for that reason...

Check the rounding on the sheet as well. He rounded 3.2 up to 4 delegates despite the directions asking them to do the opposite.

There were 70 participants, but only 61 after the realignment. The Bernie Bros may not have been willing to vote for Warren because they didn't believe a woman was electable.

If simple rounding does not add up, you round the largest fractions. In this case there were 8 delegates.

It is unlikely that the "Pin" was a PIN since it was printed on the worksheet. It is more likely an ID number for that precinct.

[jimrtex]

I'd imagine the issue is just the math - they're realizing that a lot of precincts do the math wrong. Which they've probably done in every previous Iowa caucus; we just now have the raw totals to prove it. So now they're scrambling, trying to figure out what they need to do to reconcile the numbers.

Yep.

If I were in charge, I'd treat it like a full recount. Get all records under lock and key.

The IDP probably has a full-time professional staff of under 5, so hire an accounting firm to conduct the audit recount, Form as many teams as possible, and start with the smallest counties. Presidential candidates can appoint observers.

Confiscate all cell phones and sequester everyone. If everyone is satisfied with a precinct move on to the next. If not universal approval, have objections written up, and put back under lock and key. After county is completed, release results for county, except for precincts under dispute.

Move on to the next county.

[jimrtex]

Why does the SDE lead even matter? The popular vote total matters for how candidates will perform in the primary contests, and the delegate total matters for winning the nomination. The SDE is irrelevant except insofar as it is related to the other two.

SDEs have always been the metric used, so that’s what the media is using. The AP will give the checkmark to the SDE winner.

Only because the popular vote total hasn't been available in previous caucases. But now it's available that's clearly the most relevant metric for how well a candidate will do in New Hampshire or Super Tuesday where they don't have SDEs!

Behavior and participation is different in caucuses and primaries. There is in effect a self-selection bias.

[jimrtex]

How do you win a fraction of a state delegate?

Each county gets a whole number of state delegates.

Each precinct elects a whole number of county delegates. For example, in Adair County there are 51 county delegates who will attend the county convention they will choose four state delegates.

Now imagine that those 51 county delegates attended the state convention with a fractional vote of 4/51. Adair County would still have a total of four votes at the state convention.

That it has how it works for purposes of allocating national delegates. Each county delegate in Adair County will count a 4/51 of a vote when determining the delegates chosen at the CD and state level (the CD conventions are chosen as part of the state convention).

So far with 3 of 5 precincts reported, the county convention will have Klobuchar 9, Sanders 9, Buttigieg 7, Warren 2, Steyer 2, Yang 2, and Biden 1 (yes, Biden is 7th in the county), and Klobuchar has 36/51 SDE. Sanders 36/51, Buttigieg 28/51, Warren 8/51, Steyer 8/51, Yang 8/51 and Biden 4/51.

At the county convention 4 state delegate (persons) will be chosen. This will likely be 1 Klobuchar, 1 Sanders, and 1 Buttigieg, with one yet to be determined.

In the distant past (2016) these delegates would be used to allocate national delegates. So in effect, SDE's would be an estimate of state delegates without considering viability at the county level, rounding errors, or treachery.

In 2020 they will caucus with their preference group at the state and CD convention to choose the actual national delegates - and could conceivably advocate for various platform proposals.

Note that it is a DNC rule that the allocation of national delegates occur at the first nominating event. In this case, the precinct caucuses.

[jimrtex]

How do you win a fraction of a state delegate?

SDEs are a media/state party creation from a long time back to make it all simpler to understand. After these caucuses, the elected delegates go to a county convention, where they select delegates to go to the state level. At the state level they select convention nominees. From what I understand, until very recently, counties operated like state legislatures and had total numbers of delegates for the county level vary wildly and often with only a slight accommodation for the variations in size. However, the state level delegates are proportionally distributed. SDEs therefore are a creation designed to cut out the county level step and go straight from the precinct to the state for view comprehension. So,you can have decimal SDEs because say a county could have 50 delegates at the county level, 9 of which came from precinct A, but only 4 delegates are sent to the state convention.

The size of a county conventions in rural counties are typically larger relative to the Democratic popular vote, than in urban counties. While the number of state delegates is proportional to the Democratic popular vote. If county convention sizes were proportional, then those in rural counties, or those in urban counties would be quite large.

For example, the county convention in Adair has 50 delegates, while that in Polk has 1400.

Meanwhile, the state convention delegates from Adair is 4 vs. 392 for Polk. This is a good match for the popular vote for the Democratic candidate in 2016 and 2018 (1K+ in Adair, 100K+ in Polk).

If the county conventions were proportional, then that in Adair would have 14 delegates, or that in Polk would have 5000.

It is similar to activity in a small rural HS, vs, a large urban HS. In the rural HS, any boy might be able to play on the football team - if only to play in practice, and one quarter per season. An exceptional freshman might be able to start. At the large school, the linemen will be huge and do serious weight training or be exceptionally fast. Most male students will be in the stands or in the band, or maybe not at the game at all.

In the community practically all men would be in the volunteer fire department, or at least contribute to buy new (used) equipment.

Historically, conventions worked in a cascaded fashion. Precinct caucuses would choose county delegates, county conventions would choose congressional district delegates, and congressional district conventions would choose state delegates, and the state convention would choose national delegates. The caucuses and conventions are required in state law, and happen every two years.

Iowa held its first primary in 1916. It had low attendance. The governor proposed scrapping the primary, but only the presidential primary was eliminated. During most of the 20th Century, bosses controlled matters. The 1968 Convention/Riot in Chicago resulted in requirement of proportional allotment of delegates. Reform commissions for the 1972 and 1976 conventions were headed by George McGovern and Jimmy Carter who used their knowledge of the rules to help themselves be nominated.

Iowa Democrats with a four-tiered process resulted in an early 1972 caucus. By 1976, Democratic officials in Iowa encouraged candidates to campaign in Iowa; they then told the press they had to come to Iowa to cover the candidates; and the candidates should come because the press was there; and voters should attend the caucuses because of the press and candidates publicizing the effort.

Initially, there would only be county delegates to report, and it was realized that this resulted in more delegates being reported from rural areas. So they (press and party) devised SDE to account for variance in county convention sizes. The actual state delegates would change at the county convention as non-viable candidates would be eliminated, rounding errors, and unfaithful delegates.

[jimrtex]

1520/1678 ordinary precincts counted = 90.3%

87 of the 245 outstanding results are satellites.

Top outstanding:

Polk 18
Scott 13
Woodbury 11
Story 19
Dubuque 9
Pottawatamie 5
Warren 5
Ringgold 4

Notably missing are Dallas complete, Linn 3, and Johnson 2.

Scott, Woodbury, and Dubuque suggest problems with stagecoaches from distant areas.

[jimrtex]

The last dump reduced the number of outstanding regular precincts from 158 to 109. These are in 51 counties. It is possible that they are processing by county. All the materials including ballot cards go to the county chair. By now they might be in Des Moines. Or they could be counting in each county. The counties with the most outstanding before updated the most.

Polk 12 (18 previous dump)
Scott 9 (13)
Story 8 (10)
Woodbury 7 (11)
Pottawattamie 5 (5)
Warren 3 (5)
Clayton 3 (3)
Linn 3 (3)
Tama 3 (3)

OTOH it is possible that they are encountering the most intractable errors, where they don't have ballot cards, or viability was miscalculated.

[jimrtex]

Precincts Reporting---- (1623/1765 Precincts)---  92.0%

Outstanding "Traditional"--- FINAL Non Sat precincts by largest County:

Polk County: (165/177 Precincts)

Des Moines--- Precinct (66) & (83)--- South... Bernie generally leads in surrounding precincts with Pete 2nd

Des Moines--- Precinct (24)--- NE... Sanders 1st, Pete 2nd

Urbandale--- Precinct (3) ---- Warren/Bernie/Pete all pretty strong in surrounding precincts

West DM--- Precinct (212)--- Strong Bernie Country around here, with both Warren & Pete competitive

Clive--- Precinct (4)--- Strong Pete Country to the West with Warren, Klob, & Warren doing well....

Ankeny--- Precinct (---- Bernie/Pete Country in surrounding precincts

Bondurant--- Precinct (1)---- Likely Pete

Sheldahl- Precinct (1)--- Looks to be Pete/Biden/Klob Country

Polk City--- Precinct (1)--- Looks to be Pete/Biden/Klob Country

Allen--- Precinct (1)--- Looks to be Pete/Biden/Klob sort of area

So I'm missing one because of small NYT Map Size (or possible Polk County has it's own Sat???)

I have no idea of the relative total vote within those precincts, but it looks like Bernie has a decent shot at placing 1st in 6/11 listed, and Pete the others (or even possibly a Warren or Klob grab)....

The IDP has precinct results, but they are in Iowabetical order (all mixed up).

DM 66   12
WDM 212   11
Polk City   11
Ankeny 8  10
Clive 4     10
Urb 3       10
DM 83       8
Bond 1      6
DM 24       4
Shel         1
Allen C      1

Des Moines precincts tend to be smaller because they were likely established when cars were less pervasive. OTOH, some may be heavily Democratic, and be delegate rich because delegates are apportioned based on Democratic voting.

I bet DM 62 is very black (19 delegates)

Suburban precincts tend to be larger, but also more Republican. The precincts in Ankeny are among the largest, as a newly developing area (it is a bit more working class) but the newest suburban areas to the west are in Dallas County. Ankeny is a late developer, with the advantage of closeness to Des Moines, but the disadvantage of being built over coal mines.

[jimrtex]

Precincts Reporting---- (1623/1765 Precincts)---  92.0%

Outstanding "Traditional"--- FINAL Non Sat precincts by largest County:

Linn: (83/86 Precincts)----

Cedar Rapids--- Precinct (23).... Looks to be likely Bernie, with Warren as a 2nd... Pete did place 3rd in precinct directly to the Southeast

Cedar Rapids--- Precinct (28).... Solid Bernie Country in surrounding Precincts within the City, with Warren & Pete alternating a few sloppy 2nds . Still, Pete did win a precinct directly to the South with Biden placing 2nd, so.... Lean Sanders (?)

Marion---- Precinct (7).... Tossup/ Tilt Bernie (?) Bernie captured the adjacent excepting the one to the North, but the precinct to the west was won by (1) vote over Pete, and the precinct to the East Warren did a strong 2nd with Pete placing 4th....

CR 23  6 delegates
CR 28  7 delegates
Marion 7  6 delegates

Pretty typical for the two cities. Some of the townships outside the city are smaller.

[jimrtex]

So looks like after the dump a brief survey of the most populated counties:

169/177 Polk
85/86 Linn
59/63 Scott
55/57 Johnson
60/62 Black Hawk
41/44 Woodbury

There are 47 traditional precincts in 29 counties still out. It appears that they are working all counties simultaneously. Maybe they have brought the county chairs in, along with the paper records? This last dump finished up 21 counties.

[jimrtex]

Those satellite caucuses are coming in quite strong for Bernie.

Probably Pete just wins by 1-3 delegate equivalents in the end.

He still overperformed the polls by about 4% in the initial vote count though.

He seems also on good footing for NH, with the new tracking polls out.

PETEMENTUM !

The allocation of delegates for the satellite precincts is strange.

Any satellite precinct gets 4 delegates, 3 for existing and 1 for 1-20 in attendance. There is an additional delegate for every 20 extra participants, but this is capped at 9.

1-20 4
21-40 5
41-60 6
61-80 7
81-100 8
100+ 9

The number of state delegates is also dynamic, in steps of 600. Does anyone have the state delegate counts for the traditional precincts.

In one nursing home in Mason City, there was a three-way tie for first (with one vote each). Bloomberg appears to have won the coin flip to gain 2 "county" delegates.

There was a satellite caucus in Storm Lake, apparently for workers at two Tyson Plants, either for swing shift workers, or workers uncomfortable with English language or American politics. Storm Lake schools are 60% Hispanic, and 10% Asian. These pushed Sanders popular support in the satellite caucuses for CD 4 to 75%, but he might not have a majority for the virtual county due to all the micro-caucuses.

In the out-of-state caucuses, Sanders finished fifth in the popular vote, due to a weak performance in the winter sites, but had strong support at the college sites, and picked up support there.

[jimrtex]

I'm looking at the satellite results and they don't make sense.

Either I don't understand how they are calculated, or the IDP doesn't understand.

I think the NYT needle is at 50/50.