Nevada Caucus Data (Final UpDate)

          Grounded in a myriad of polls, sampling, and statistical techniques various individuals, organizations, and pundits have been leaking predictions from the wee hours of the day. As you all know I am a cranky old scientist and damn proud of it. I don't give much credence to such nonsense. I, on the other hand, thrive on data, modeling, and (most of all) clear thinking.

          So, here are some Nevada Democratic Results, reported from actual caucus tally sheets and displayed at www.npr.org (caution link opens in new window). Let us all hope this information is more in touch with reality than the insanity generated by the Iowa democratic (Idiocracy !) establishment.

Captured 16:34 PST 22 Feb 2020

Captured 18:31 PST 22 Feb 2020

Captured 19:33 PST 22 Feb 2020

Captured 20:06 PST 22 Feb 2020

Captured 21:17 PST 22 Feb 2020

Captured 21:49 PST 22 Feb 2020

Captured 05:53 PST 23 Feb 2020

Captured 15:56 PST 23 Feb 2020

Captured 10:38 PST 24 Feb 2020

Captured 12:38 PST 24 Feb 2020

A Plot for your Viewing Pleasure

Share
up
24 users have voted.

Comments

Make no sense?

up
5 users have voted.

@Battle of Blair Mountain @Battle of Blair Mountain
Like the Iowa caucuses, the Nevada caucuses are set up so that precincts elect county delegates. This creates an electoral-college type of thing, where depending on how your votes are distributed, your CCD percent won't map directly to your person-vote percent.

up
12 users have voted.

The earth is a multibillion-year-old sphere.
The Nazis killed millions of Jews.
On 9/11/01 a Boeing 757 (AA77) flew into the Pentagon.
AGCC is happening.
If you cannot accept these facts, I cannot fake an interest in any of your opinions.

@Battle of Blair Mountain
but 47% of the county delegates. The last place I saw an estimate of actual pledged national delegates "so far" (a couple of hours ago, I think), Sanders had 9 and everybody else had 0.

Caucus math, man, it's something else.

up
13 users have voted.

The earth is a multibillion-year-old sphere.
The Nazis killed millions of Jews.
On 9/11/01 a Boeing 757 (AA77) flew into the Pentagon.
AGCC is happening.
If you cannot accept these facts, I cannot fake an interest in any of your opinions.

@UntimelyRippd
that the reported raw votes are from the first round. Delegates are apportioned in each precinct after realignments by those choosing candidates that didn't meet the 15% threshhold for viability. The early voting ballots were ranked choice for three. So, within the presence of one's neighbors, it's a hybrid of ranked-choice voting and not complicated if one has all the data. The delegates are chosen at the caucus -- they are real people and not merely a number.

In 2004 and 2008, IA also published the precinct delegate counts (but not the votes). That was apparently junked in 2016 when they released only the SDE counts, which is a mathematical conversion from the precinct level delegate count, but they wasn't a new practice. However, things can change in-between the caucuses and the state convention; so, it's potentially a distorted picture.

(In 2016 New Hampshire, Sanders received 60.1% of the vote to HRC's 37.7%, but Sanders got 16 delegates to the national convention and HRC got 15.)

Some may recall the conflict (and fake story about chair throwing by Bernie delegates) at the Clark County convention. The background to that is that the Clark Co. DP made it difficult for Sanders' delegates to attend. Everything was made very clear for Clinton's delegates. The known and pre-existing problem for Clinton (in both 2008 and 2016) was that at some caucuses voters were strong-armed into voting for Clinton, and it resulted in delegates not that committed to Clinton. In '08 many of them didn't bother to go the county convention. Whereas Obama's team made sure that his delegates got there. That's how beginning with 45% of the county delegates to Clinton's 50.8% he ended up with 14 delegates to the national convention to her 11 delegates. In '16 the county DP chair and officers were more organized and the Sanders delegation wasn't as organized as Obama's had been. Even then, it was apparent to the Sanders delegates that they were cheating.

up
6 users have voted.

@Marie
vote totals (and you're correct, in the charts above they are first-round, but other sources are showing second-round, and yet other sources show both), the percentage values are the percentages of county delegates, not percentages of the popular votes. As far as I can figure, Sanders currently has less than 40% of the second-round vote (and less than 35% of the first-round vote), but 46% of the county delegates and 75% to 80% of the pledged national delegates.

up
1 user has voted.

The earth is a multibillion-year-old sphere.
The Nazis killed millions of Jews.
On 9/11/01 a Boeing 757 (AA77) flew into the Pentagon.
AGCC is happening.
If you cannot accept these facts, I cannot fake an interest in any of your opinions.

And Buttigieg keeps creeping up toward 15%. That's not good news. I don't know exactly how it's done, but I really want to see him get no delegates.

up
10 users have voted.

This NPR page has result also :

@tle

https://apps.npr.org/liveblogs/20200222-nevada/

up
4 users have voted.

@entrepreneur
continues to decline as Pete's and Warren's increases.

Wouldn't be nearly as curious if Sanders at 46.6% hadn't remained unchanged from 11% reporting to 50% reporting.

I'm probably too suspicious, but something about this looks hinky.

up
4 users have voted.
PriceRip's picture

@Marie

but something about this looks hinky.

          … I don't see anything that "looks hinky".

RIP

up
1 user has voted.

"I know you believe you understand what you think I said, but I am not sure you realize that what you heard is not what I meant."
Robert J. McCloskey, U.S. State Department spokesman. From a press briefing during the Vietnam war.

ovals49's picture

@PriceRip
In fact the nearly straight lines in the first round results reinforces the lack of “hinky”. The later “realignments” are more like the expected results of a ranked choice election. Many of those who were supporting “non viable” candidates had already decided Bernie was to be avoided at all cost and moved to alternate not-Bernie candidates. This would naturally result in the surviving non-Bernie candidate’s rising in the later realignments.

I’m open to many “conspiracy theories” but in this instance the evidence suggests there is no “there” there.

up
4 users have voted.

"An oligarchy of private capital cannot be effectively checked .....because under existing conditions, private capitalists inevitably control, directly or indirectly, the main sources of information."
Albert Einstein

@PriceRip
anything for me. X as equal to precincts reported is flawed because precincts aren't equal. Nor do raw vote totals work as technically it doesn't matter if 10 or 100 voters show up to caucus because the CCD for that precinct are set in advance based on some calculated value for the precinct population. Plus the raw votes reported are before realignments. It's the reported CCD that's at stake, and the reported percentage for each candidate is based on CCD.

up
0 users have voted.

Of the 7 results released so far Bernie has been at 46.6% for 4 of 7. I think their fixer algorithm has been programmed to give him 46.6%. When it deviates slightly it tracks back to 46.6% again.

@Marie

up
4 users have voted.
PriceRip's picture

@entrepreneur

          … have you never seen a linear relationship suggesting an homogeneity?

RIP

up
3 users have voted.

"I know you believe you understand what you think I said, but I am not sure you realize that what you heard is not what I meant."
Robert J. McCloskey, U.S. State Department spokesman. From a press briefing during the Vietnam war.

the latter 5 batches are the tip off that they are a desired and targeted result, but only when viewed in conjunction with the (1st) 4% batch and the (2nd) 11% batch.

The latter 5 batches' individual averages are 46.6%, and the 5 groups' total average is 46.6%. Thus each new incoming batch doesn't change the average more than a tiny bit. Good so far. But let's suppose that there are two outlier batches in the 7 released. It is unlikely that :

1) The only non 46.6% batches happen to be the 1st and 2nd releases, and
2) These two unrelated batches happen to combine to 46.6% (the average of the others).

This looks like a system that had an initial value that was too high (54%. OVER 50%!) and an offset was deliberately applied to move it to a target value of 46.6%.

It has been over 2 hours since the 50% release. I am pretty confident that they are massaging the data. I do think that their end game is to keep Bernie under 50%, which circumstantially fits with them taking the initial 54% down to 46.6%.

Obviously, I am sticking my neck out a bit because only 50% of the data has been released.

@PriceRip

up
11 users have voted.

@entrepreneur In going from 43% to 49.88% of precincts reporting (still under 50%), that 7% of precincts had Biden at 11.6%, vs the first 43% being at 25% (these are all popular vote numbers). Both Sanders and Buttigieg had higher percentages, of ~6.3 and 7% respectively. At that rate, we'll end up with another Iowa.

up
2 users have voted.

I am just about out of steam. The assholes haven't released any more data for 2-1/2 hours. And I think the intermediate data is important.

up
7 users have voted.

@entrepreneur
Now over twenty hours since the caucuses ended and only 60% of the precinct tallies have been reported. Reported results have been slow since yesterday when the first few precincts came in. What's the hold-up?

There was a large flip from 4% reported to 11% with Sanders dropping 7.7% and Biden gaining 6.4%, and Pete jumping from 8.7% to 13.6%. From 11% to 60% reported, Sanders has dropped by a mere 0.6% to 46%. Biden has dropped from 23.8% to 19.6%. And Pete is now up to 15.3%.

There may be a perfectly reasonable explanation for the changes from 4% to 11%. My guess would be a large dump from precincts disproportionately high in AAs and older voters. Both groups favor Joe, but I can't explain why Pete would improve by so much in such precincts. That interpretation seems to be supported by Biden dropping in each of the subsequent reports. Also the precincts reported at 4% were outliers for Steyer and Amy.

The subsequent slight increases for Warren are in-line with the 4% report, and from 11% to 60%, Warren slowly returned to that 10% she had at 4%. The slow and steady slight increases for Steyer and Amy (from 11% to 60% reporting) would also seem to support that data dump between the 4% and 11% reporting. Would say the same for Pete, if not for his significant leap with the date dump.

What seems to be missing are the younger and Latino precincts, groups that favored Sanders by much higher percentages than 46%. Is the NV DP holding those back? Preferably until after the SC primary on Saturday?

up
3 users have voted.

and fraud. They have consistently earned our distrust. And I won't give them the benefit of the doubt.

My guess is that when the first 4% showed Bernie decisively over 50% with a 36% lead they freaked the hell out. The notion that they just accepted it and opted for fairness rather than fudging the results is inconceivable.

@Marie

up
3 users have voted.

@entrepreneur
numbers -- it's really not that easy to do once a caucus is complete. My guess they're selectively releasing precinct results. Holding back on the stronger ones for Sanders.

Reporting appears stuck at 60%, and we have little idea what is being held back and more importantly why.

up
1 user has voted.
ovals49's picture

@Marie
is not much data to draw any conclusions from. Looking at the 8 data sets above, the 4% report seems a bit of an outlier. With additional data the numbers and percentages settle down and the incremental changes become quite consistent.

I worked up a spreadsheet that included percentages of both the raw vote and the CCD allocation percentages and the delta between those two percentages. To my surprise, Bernie consistently outperformed his raw vote percentage in his CCD allocation by plus 1.6% to 7.0%. (If the small 4% initial 2/22 4:34 pm data set is excluded that range tightens to plus 6.5% to 7%). Perhaps something in the process or allocation formula favors the leader, but in any event, Bernie was definitely the beneficiary by a significant margin in the allocation of CCD’s.

With 28% of the returns still out we could still see some hinky monkey business. But at the moment I can not see anything curious other than the Tom and Amy flip on the most recent 72% reporting data. If Amy is the next to drop out we might want to give that anomaly a closer look. Amy only picked up 40 CCD’s while Tom picked up 144! That’s quite a different pattern from the rest of the data releases.

up
1 user has voted.

"An oligarchy of private capital cannot be effectively checked .....because under existing conditions, private capitalists inevitably control, directly or indirectly, the main sources of information."
Albert Einstein

PriceRip's picture

Data Captured 21:49 PST 22 Feb 2020

          compared to

Data Captured 05:53 PST 23 Feb 2020

RIP

up
2 users have voted.

"I know you believe you understand what you think I said, but I am not sure you realize that what you heard is not what I meant."
Robert J. McCloskey, U.S. State Department spokesman. From a press briefing during the Vietnam war.