Biden vs. Trump: The view from six months out

A note to readers: I have temporarily stopped writing “dispatches” about how my wife Nell, our two daughters and I cope with social distancing and the closure of Massachusetts schools through the end of the 2019-20 school year because they started to feel repetitive. When and if that changes, I will resume dispatching.

**********

As I write this, it is exactly six months until the 2020 United States (U.S.) presidential election, which will conclude on November 3, 2020. On April 8, 2020, U.S. Senator from Vermont Bernie Sanders announced he was suspending his campaign for the 2020 Democratic presidential nomination, making former Vice President Joseph R. Biden, Jr. the presumptive nominee against incumbent Republican president Donald J. Trump.

Using all publicly-available polls of the presidential election—both nationally and at the state level, recognizing presidential elections are determined by the Electoral College—conducted since January 1, 2019, I have been tracking the relative performance of contenders for the 2020 Democratic nomination against Trump. When given the choice, I used polls of likely voters over those of registered voters, and the latter over polls of adults only; I also used polls including such possible third-party candidates as former Starbucks CEO Howard Schultz and U.S. House of Representatives Member Justin Amash of Michigan. Table 1 lists the number of national polls conducted each month for both candidates based upon the midpoint of the poll’s field dates; some polls were actually conducted in two months.

Table 1: Number of National Polls Assessing Hypothetical 2020 Match-ups Between Biden/Sanders and Trump by Month

Month Biden Sanders
January 2019 1 1
February 2019 4 3
March 2019 7 6
April 2019 6 6
May2019 7 5
June 2019 10 9
July 2019 8 7
August 2019 8 8
September 2019 15 11
October 2019 18 13
November 2019 8 4
December 2019 14 9
January 2020 20 17
February 2020 23 21
March 2020 33 23
April 2020 41 3
TOTAL 223 146

Just seven of 41 total pollsters (average grade: B-/B) account for 54% of Biden versus Trump polls; the values are similar for Sanders:

  • IBD/TIPP (A/B), 10 polls
  • Fox News (A-), 13 polls
  • Harris X (C+), 13 polls
  • Emerson College (B+), 18 polls
  • Ipsos (B-), 18 polls
  • Morning Consult (B/C), 22 polls
  • YouGov (B-), 36 polls

Figure 1, meanwhile, shows how Biden and Sanders fared monthly against the president, using my weighted-adjusted polling averages, or WAPA. Basically, I use data published by FiveThirtyEight.com to adjust each poll for partisan lean (tendency of a pollster to err more Democratic or Republican than other pollsters in analogous races) and overall quality (using the letter grade assigned by FiveThirtyEight.com). I also weight more recent polls—again using field midpoint—higher, using the ratio of the number of days since January 1, 2019 and the total number of days between January 1, 2019 and November 3, 2020. Finally, I average two different versions of WAPA: one treating polls by the same pollsters as statistically independent values, and one which treats all polls by the same pollster as a single value; differences between estimates are generally negligible.

Figure 1: Monthly weighted-adjusted average margins for Biden and Sanders versus Trump since January 2019Biden and Sanders v Trump since Jan 2019

Only one national poll assessing hypothetical matchups between Biden or Sanders and Trump was conducted in January 2019, so I combined them with the four and three, respectively, from February 2019 to generate Figure 1. Biden and Sanders have consistently led Trump in head-to-head matchups, never dropping below Sanders’ 2.0 percentage point (“points”) lead in December 2019. Through September 2019, Biden’s margin was typically three-to-four points higher, though Sanders still led Trump by 4.3 points on average, versus 7.8 points for Biden. From October 2019 through February 2020, though, the two men fared equally well versus Trump, with Biden ahead an average 5.4 points and Sanders ahead 4.9 points. Once Biden’s nomination began to become clear in March 2020, however, Biden again began to fare better versus Trump than Sanders, averaging a 5.7-point-lead to Sanders’ 3.4-point lead. Overall, Biden has a 6.1-point lead over Trump, not meaningfully different than his lead over the last two months; Sanders exited the race with an overall national lead of 4.3 points versus Trump, though that lead had begun to drop slightly over the last two months.

**********

Again, however, presidential elections are actually fought across all 50 states and the District of Columbia (“DC”), with the plurality winner in each state/DC winning every electoral vote (“EV”) from that state.

To that end, Table 2 lists the number of polls conducted within each state since January 1, 2019 of hypothetical matchups between Biden/Sanders and Trump, plus that state’s 3W-RDM, an estimate of much more or less Democratic than the nation a state tends to vote; 11 states[1] and DC have not yet been polled. 

Table 2: Number of state-level polls assessing hypothetical 2020 matchups between Biden/Sanders and Trump since January 1, 2019

State 3W-RDM Biden Sanders
Michigan 2.2 33 23
Wisconsin 0.7 30 26
Texas -15.3 27 21
North Carolina -6.0 23 16
Pennsylvania -0.4 23 17
Florida -3.4 19 11
Arizona -9.7 17 14
California 23.2 14 13
New Hampshire 0.1 10 10
Iowa -4.7 9 8
Georgia -9.6 8 6
Ohio -5.8 7 6
Virginia 1.5 7 6
Nevada 2.0 6 6
Utah -33.1 5 3
South Carolina -15.7 4 4
Maine 5.9 4 3
North Dakota -29.4 4 2
Washington 12.1 4 3
Missouri -15.9 4 3
Connecticut 12.8 4 4
New York 21.6 3 1
Colorado 2.2 3 2
Kentucky -28.7 2 1
Montana -18.6 2 2
New Mexico 6.5 2 1
Alabama -28.4 2 2
Kansas -23.4 2 2
Oklahoma -38.1 2 2
New Jersey 12.0 2 1
Mississippi -18.5 2 1
Minnesota 1.5 1 1
Massachusetts 22.1 1 1
Alaska -19.2 1 1
West Virginia -35.5 1 1
Delaware 12.5 1 1
Tennessee -25.8 1 1
Maryland 22.6 1 1
Indiana -16.3 1 0
TOTAL D-6.2 292 227

It is not surprising that eight of the 14 most-polled states thus far are “swing” states, those with 3W-RDM between -5.0 and +5.0, including the four closest states won by Trump in  2016: Florida (19 Biden, 11 Sanders), Pennsylvania (23, 17), Wisconsin (30,26) and Michigan (33,23). In fact, the Pearson correlation between the absolute value of a state’s 3W-RDM and the number of times it has been polled for the 2020 presidential election is -0.47 for Biden and -0.48 for Sanders, meaning the closer a state is to the national average (i.e., a pure toss-up in a dead-even national race), the more often it has been polled. Also highly-polled are large states like California and Texas, red-drifting states like Ohio and Iowa, and emerging Democratic opportunities like Arizona, Georgia and North Carolina. 

While U.S. presidential elections are decided on a state-by-state basis, though, national averages are still important. Combined with 3W-RDM, they provide the “expected Democratic-minus-Republican margin” in each state in 2020, all else being equal. Comparing polling averages to this expected value tells us where Biden may currently be under- or over-performing, or which states have drifted Democratic or Republican since 2016.

For example, Biden leads Trump overall by 6.1 points. North Carolina has recently been about 6.0 points less Democratic than the nation as a whole. Adding those two values together (6.1 – 6.0 = +0.1) yields an expected photo-finish in North Carolina in 2020. However, Biden leads Trump by a mean 2.2 points in 23 polls thus far in North Carolina, meaning Biden is “outperforming” expectations there by about 2.1 points.

This could mean any or all of three things:

  1. WAPA is the more accurate reflection of the November election and either
    1. North Carolina has drifted about two points toward the Democrats since 2016, or
    2. The true “expected value” is somewhere between Trump winning by 5.3 points and Biden winning by 5.5 points, based upon an average 3W-RDM error margin of 5.4 points in recent elections.
  2. The “expected” value is the more accurate reflection, and Republican-leaning voters will drift back toward Trump over the next six months, making North Carolina nail-bitingly close on election day.

Table 3 lists every state’s expected value and WAPA; for ease of presentation, I include Biden-Trump values only.

Table 3: Expected and actual polling margins for Biden over Trump in each state in November 2020

State 3W-RDM Expected WAPA WAPA-Expected
DC 82.0 88.2    
Hawaii 34.3 40.4    
Vermont 27.7 33.8    
California 23.2 29.3 27.1 -2.2
Maryland 22.6 28.7 25.0 -3.7
Massachusetts 22.1 28.2 38.0 9.8
New York 21.6 27.7 27.9 0.2
Rhode Island 18.0 24.1    
Illinois 14.7 20.8    
Connecticut 12.8 18.9 16.9 -2.0
Delaware 12.5 18.6 16.4 -2.2
Washington 12.1 18.2 19.8 1.6
New Jersey 12.0 18.1 16.1 -2.0
Oregon 8.7 14.8    
New Mexico 6.5 12.6 10.4 -2.2
Maine 5.9 12.0 9.2 -2.8
Michigan 2.2 8.4 5.9 -2.5
Colorado 2.2 8.3 6.9 -1.4
Nevada 2.0 8.1 3.5 -4.6
Minnesota 1.5 7.6 12.7 5.1
Virginia 1.5 7.6 7.8 0.2
Wisconsin 0.7 6.8 1.7 -5.1
New Hampshire 0.1 6.2 4.5 -1.7
Pennsylvania -0.4 5.7 4.2 -1.5
Florida -3.4 2.7 1.9 -0.9
Iowa -4.7 1.4 -3.5 -4.9
Ohio -5.8 0.3 3.0 2.7
North Carolina -6.0 0.1 2.2 2.1
Georgia -9.6 -3.4 -0.3 3.2
Arizona -9.7 -3.6 2.0 5.6
Texas -15.3 -9.1 -2.0 7.2
South Carolina -15.7 -9.6 -9.6 0.0
Missouri -15.9 -9.8 -8.6 1.3
Indiana -16.3 -10.2 -14.1 -3.9
Mississippi -18.5 -12.4 -12.9 -0.5
Montana -18.6 -12.5 -16.0 -3.5
Alaska -19.2 -13.0 -4.2 8.8
Louisiana -22.2 -16.1    
Kansas -23.4 -17.3 -11.2 6.1
Nebraska -25.8 -19.7    
South Dakota -25.8 -19.7    
Tennessee -25.8 -19.7 -15.3 4.4
Arkansas -28.2 -22.1    
Alabama -28.4 -22.3 -19.6 2.7
Kentucky -28.7 -22.6 -15.9 6.7
North Dakota -29.4 -23.3 -20.6 2.7
Utah -33.1 -27.0 -12.3 14.7
Idaho -34.2 -28.1    
West Virginia -35.5 -29.3 -34.0 -4.7
Oklahoma -38.1 -32.0 -26.1 5.9
Wyoming -45.7 -39.6    
Average D-6.4 Trump+0.05* Biden+0.9 +1.0

        * Only for the 39 states with both measures

The correlation between the expected margin and WAPA is a very-reassuring +0.96, meaning the polling is broadly in line with the underlying “fundamentals” of the election. Still, Biden is polling ahead of those fundamentals by an average of about one percentage point, meaning the state-level polling as a whole is even better for Biden than his already-solid national polling.

Nonetheless, there are clearly states where Biden is underperforming expectations, including the vital and heavily-polled state of Wisconsin. While Biden leads there by about 1.7 points overall, he “should” be ahead there by about 6.8 points. Moreover, he is trailing by about 3.5 points in nearby Iowa, even though Biden “should” be ahead by about 1.4 points. And while Biden leads Trump by about 3.5 points in Nevada, that is 4.6 points below what the fundamentals suggest.

The story is similar, but more narrowly so, in the key states of Michigan, Pennsylvania, New Hampshire and Florida: Biden leads Trump in these states by an average of 4.1 points, though he “should” lead by an average of 5.8 points, a mean “underperformance” of 1.7 points.

Moreover, there appears to be something of a partisan split in Biden’s over-and under-performance: in the 10 states with both measures and 3W-RDM≥5.0, Biden is underperforming by 0.3 points, on average, though once you remove the single poll of Massachusetts, that jumps to -1.6 points. At the same time, in the analogous 20 Republican states with 3W-RDM≤5.0, Biden is overperforming by 3.2 points, though that drops to 2.6 with the massive outlier of Utah removed.

Let me again stress, however, that there is a lot of “wobble” in the “expected margins,” as well as in the polling averages—especially given that most states have seen very little recent polling. All of this “over- and underperforming” may simply be statistical noise, as we try to read too much into highly stochastic data.

Still, the two values are sufficiently closely aligned to combine them into a single, six-months-out estimate of Biden’s margin over Trump on November 3, 2020, based upon the assumption polls become more predictive as an election gets closer:

  1. Arbitrarily assign expected value and WAPA equal weight as of January 1, 2020.
  2. If the most recent poll in a state was conducted more than 100 days prior to January 1, 2020, WAPA is weighted just 10%. This only applies to Massachusetts, Alaska and Kentucky, with Minnesota the only other state whose most recent poll was conducted in 2019.
  3. WAPA weight increases, by day, with proximity to November 3, 2020.

At the same time, I introduced a probabilistic element into these estimates—rough calculations of how likely Biden is to win the EV from each state, assuming such likelihood is distributed normally:

  1. For expected margins, I used a mean of estimate-0.8 and a standard error of 7.1[2]
  2. For WAPA, I used a standard error of 3.0, roughly the margin of error in most quality polls.
  3. Overall probability Biden wins a state’s EV calculated the same as for predicted final margin

While the means and standard errors are somewhat arbitrary, albeit broadly defensible, the final EV probabilities shown in Table 4 are in line with what other forecasters are saying.

Table 4: Estimated final state margins and probability of winning EV, Biden vs. Trump, November 2020

State EV P(EV): Expected P(EV):

WAPA

P(EV):

Overall

Predicted Margin
DC 3 100.0%   100.0% 88.2
Hawaii 4 100.0%   100.0% 40.4
Vermont 3 100.0%   100.0% 33.8
California 55 100.0% 100.0% 100.0% 27.9
Maryland 10 100.0% 100.0% 100.0% 26.6
Massachusetts 11 100.0% 100.0% 100.0% 29.2
New York 29 100.0% 100.0% 100.0% 27.8
Rhode Island 4 99.9%   99.9% 24.1
Illinois 20 99.8%   99.8% 20.8
Connecticut 7 99.5% 100.0% 99.8% 17.9
Delaware 3 99.4% 100.0% 99.7% 17.5
Washington 12 99.3% 100.0% 99.8% 19.0
New Jersey 14 99.2% 100.0% 99.7% 17.1
Oregon 7 97.6%   97.6% 14.8
New Mexico 5 95.2% 100.0% 97.6% 11.5
Maine 4 94.3% 99.9% 97.7% 10.3
Michigan 16 85.6% 97.5% 93.9% 6.6
Colorado 9 85.5% 99.0% 93.3% 7.5
Nevada 6 84.8% 88.0% 86.7% 5.4
Minnesota 10 83.1% 100.0% 89.4% 9.5
Virginia 13 83.0% 99.5% 93.7% 7.7
Wisconsin 10 80.2% 71.5% 74.3% 3.3
New Hampshire 4 77.7% 93.2% 88.4% 5.0
Pennsylvania 20 75.6% 92.0% 86.9% 4.7
Florida 29 60.7% 73.5% 69.4% 2.2
Iowa 6 53.3% 12.0% 28.4% -1.6
Ohio 18 47.1% 84.1% 72.5% 2.1
North Carolina 15 46.1% 76.5% 67.2% 1.5
Georgia 16 27.5% 46.3% 40.5% -1.3
Arizona 11 26.8% 75.1% 58.7% 0.1
Texas 38 8.1% 25.5% 20.1% -4.2
South Carolina 9 7.2% 0.1% 3.0% -9.6
Missouri 10 6.7% 0.2% 2.9% -9.1
Indiana 11 6.1% 0.0% 2.0% -12.8
Mississippi 6 3.2% 0.0% 1.3% -12.7
Montana 3 3.1% 0.0% 1.3% -14.5
Alaska 3 2.6% 8.1% 3.1% -12.2
Louisiana 8 0.9%   0.9% -16.1
Kansas 6 0.5% 0.0% 0.2% -14.3
Nebraska 5 0.2%   0.2% -19.7
South Dakota 3 0.2%   0.2% -19.7
Tennessee 11 0.2% 0.0% 0.1% -17.5
Arkansas 6 0.1%   0.1% -22.1
Alabama 9 0.1% 0.0% 0.0% -20.9
Kentucky 8 0.0% 0.0% 0.0% -21.9
North Dakota 3 0.0% 0.0% 0.0% -21.6
Utah 6 0.0% 0.0% 0.0% -17.1
Idaho 4 0.0%   0.0% -28.1
West Virginia 5 0.0% 0.0% 0.0% -31.7
Oklahoma 7 0.0% 0.0% 0.0% -29.0
Wyoming 3 0.0% -39.6 0.0% -39.6

Six months before election day 2020, and with all of the caveats about what voting will even look like during a pandemic, Biden is clearly in a commanding position to be elected the 46th president of the United States.

  • He is projected to win by at least 3.3 points in enough states to get him to 279 EV, or 278 depending on what happens in Maine, which, along with Nebraska, allocates two EV to the statewide winner and one each to the winner of its Congressional districts.
    • He has narrower leads in Florida, Ohio and North Carolina, which combine for 62 EV, increasing his total to 340 or 341.
    • Arizona’s 11 EV are balanced on a knife’s edge.
  • He is favored at least 86% in enough states to get him to 268 or 269 EV
    • He would then need to win ONLY ONE of Wisconsin (74.3%), Ohio (72.5%), Florida (69.4%) or North Carolina (67.2%) to win the presidency. Assuming Biden’s chances of winning each state are statistically independent from each other (a lousy assumption), he has about a 99% chance of winning AT LEAST one of these states.
  • He has at least a 58% chance in enough states to earn him 351 or 352 EV, at least 81 more than required.
  • And if things truly break Biden’s way, he has a 40.5% chance to win the 16 EV in Georgia, a 28.4% to win the 6 EV in Iowa, and a 20.1% chance to win the 38 EV of Texas, upping his total to 411-413 EV, depending on what happens in the 2nd Congressional district of Nebraska, which allocates its EV the same as Maine.

Using the simplistic—perhaps even simple-minded—method of multiplying Biden’s probability of winning each state by its EV and summing yields a “projected” EV total of 335.2, fairly close to the 341 generated by taking the 232 EV won by Hillary Clinton in 2016, adding Michigan and Pennsylvania to get to 268, then adding Wisconsin, Florida, Ohio and North Carolina (and the last EV in Maine).

This lead looks even more robust when you make either of two reasonable assumptions:

All polls are overestimating Biden’s margins by 3.0 points.

In this scenario, Biden’s projected EV drops to 286, still 16 more than required. He would be favored at least 80% to win in enough states to win 239 EV, though he would be favored by at least 64% in three states totaling 30 EV, putting him on the doorstep. He would then have to win one of Wisconsin or Ohio, at 44% each; he would have about a 69% chance to do so.

The point is, even if the polls are consistently off by this much, Biden would still be roughly even money to win the presidency. That said, Biden would still be winning by 3.1 points nationally, demonstrating the current Republican bias in the Electoral College.

All polls are underestimating Biden’s margins by 3.0 points.

In this scenario, Biden’s projected EV are a landslide-level 373.7, more than 100 more than necessary. He would be favored at least 80% to win enough states to earn 341 EV, while being a 77.3% favorite in Arizona and a 69.8% favorite in Georgia, for a total of 368 EV. Adding in the states where Biden would be roughly even money—Iowa and Texas—gets us once again to 412.

This appears to be Biden’s upper limit, as even in this scenario where he is wining nationally by 9.1 points, he is no more than 9% favored to win any additional states.

Now, none of this is to say Biden is guaranteed to be the next president of the United States; it would be monumentally foolish for me to conclude that this far from the election, particularly if Amash earns more than, say, three points in the national popular vote. I am simply noting that all indications point very strongly in that direction, based on the data we have right now.

Until next time…please stay safe and healthy…

[1] Hawaii, Vermont, Rhode Island, Illinois, Oregon, Louisiana, Nebraska, South Dakota, Arkansas, Idaho, Wyoming

[2] The former value is the mean arithmetic difference between “expected” and actual D-R margins across 153 state-level contests in 2008, 2012 and 2016, while the latter value is the standard deviation of these values. I recognize this is not a standard error. However, using the value 13.6—the range of values covering 95% of all values divided by 1.96, the final EV projection changes by only 1.0

Dispatches from Brookline: Home Schooling and Social Distancing V

On Wednesday, March 25, 2020, Massachusetts Governor Charlie Baker issued an executive order extending the closure of all public schools in the Commonwealth until at least May 4, 2020.

In four previous posts (I, II, III, IV), I described how my wife Nell, our two daughters—one in 4th grade and one in 6th grade—and I were already coping with social distancing and the closure of the public schools in Brookline, Massachusetts until at least April 7, 2020. Besides staying inside as much as possible, we converted our dining room into a functioning classroom complete with workbooks, flip charts and a very popular white board.

**********

To give our daughters something of a break during the week—especially our younger daughter, who has a yet-to-be-formally-diagnosed learning disability and attention deficit disorder—there is no “school” on Wednesday mornings. This means that when I came downstairs on March 25, 2020, Nell had not written a daily schedule on the flip chart. This likely saddened our younger daughter who was apparently going to have free reign over what the afternoon classes would be called.

To be fair, the girls had done something broadly educational that morning. With Nell, they had watched and discussed two episodes of The Blue Planet.

And they are continuing to produce drawings at a solid clip.

Wall of art March 25

The framed painting in the middle is one of two I bought when I first moved back to the Boston area—Waltham, to be precise—from my native Philadelphia in early September 2005. I do not recall why I entered the Martin Lawrence Galleries on Newbury Street (which appears recently to have closed), but once inside I was quite taken with a collection of paintings by Liudmila Kondakova. Using funds from a recent inheritance, I bought this painting and a smaller one. Both depict Paris street scenes, and both have my last name written somewhere in them.

**********

The break from school work does not extend to the afternoons, so we convened just after 2:45 pm to discuss the history of the American presidential nominating system. My attached notes for this class were a bit more scattershot than usual, but they worked well enough to tell a series of what I hoped would be interesting stories.

March 25.docx

I noted in “Dispatch IV” our daughters’ penchant for assigning monikers to historical figures. Well, they came close to doing that when I came to the 1960 Democratic nomination process, and I explained one of the primary contenders that year was United States Senator (“Senator”) Hubert Humphrey of Minnesota.

“Who names a kid ‘Hubert?” asked our older daughter. “Did his parents want him to get teased his whole life?”

After observing his middle name was Horatio—he was once erroneously referred to as Hubert Horatio Hornblower—I defended the late Vice President as a good and honorable man, though I never did get around to discussing his groundbreaking speech on civil rights at the 1948 Democratic national convention.

We concluded with a rapid-fire discussion of how Democrats—proportionally, with a minimum of 15% statewide or in a Congressional district—and Republicans—mostly winner-take-all—differ in the way they apportion nominating convention delegates.

This was followed by easily the most cringeworthy moment I have thus far endured as a parent.

I had been talking about the role “expectations” play in the modern primary and caucus system, One example I used was the way then-Arkansas Governor Bill Clinton used a 2nd place finish in the 1992 New Hampshire Democratic Primary to label himself “The Comeback Kid.”

They had been vaguely aware of Clinton’s marital indiscretions, and they understood he had been impeached for lying under oath about cheating on his wife while he was president of the United States. What they did not know, though, were the sordid details.

And they very much wanted to know what they are; they essentially promised to hear the end of my spiel in exchange.

So…after pouring myself a fresh cup of hot black coffee, half-decaffeinated to brace myself…I told them.

I did not use the words “blow job” or “fellatio,” but I described how a government shutdown in 1995 had allowed Clinton to spend time alone in the Oval Office with a young White House intern named Monica Lewinsky. And how one time she had worn a blue dress. And how she kept that dress after it came to have Clinton’s semen on it after a certain action I described…

…at which our older daughter interjected, “Oooo, gross! He peed through that! Why would anyone ever want to do that?!?”—or words broadly to that effect. Our younger daughter, meanwhile, just sat quietly, listening.

They particularly wanted to know why Ms. Lewinsky had kept that dress.

“Well, Clinton kept lying about what they had done. So she kept it as proof.”

And that was that.

Oy.

**********

At just after 4:45 pm, we reconvened for what I had thought would be the most fun part of the afternoon.

I wanted to talk about random sampling—the idea that you could get, for example, a fairly accurate impression of the distribution of attitudes in a very large population by randomly identifying a much smaller proportion of that population. However, I should have known that things would go awry when I used this example: a group of one million people includes 750,000 (75%) who prefer chocolate ice cream and 250,000 (25%) who prefer vanilla ice cream. Rather than ask every one of those people which flavor they prefer one could simply randomly select 1,000 of them to ask. Most of the time, if you sample properly, you will come within a few percentages either way of 75 and 25.

Well, our younger daughter simply wasn’t having it.

“What if someone doesn’t like either?” she began.

I explained this was merely an example, but that did not work.

“What if you like some other flavor?”

“It is a forced choice,” I weakly noted.

At this point, her sister chimed in.

“Well, which one do you prefer?”

This led to a long pause which ended in a non-answer.

At this point, I simply began talking about the activity we were about to do, one that involved 100 carefully selected cards from an UNO deck.

What I wanted to do was illustrate how queried multiple random samples from an identical population will center around “true” values within that population. My original conception was to put something like 60 blue and 40 red of the same small objects into a hat—Nell’s grandfather’s top hat lives in my home office—and have them draw 15 balls from that hat 10 times. We would record those draws to see how close they came to 60% blue and 40% red in the aggregate.

Of course, we did not have quite the objects I was envisioning, and I did not really want to cut up small bits of blue- and red-colored paper. That was when I remembered our bedraggled deck of UNO cards. There were enough cards remaining for me to compile a deck of 100 cards:

  • 50 blue and green cards, with the former “definitely voting Democratic” and the latter “leaning toward voting Democratic”
  • 43 red and yellow cards, with the former “definitely voting Republican” and the latter “leaning toward voting Republican”
  • 7 wild cards, for undecided voters

What I had not counted on was just how hard it is to shuffle—and I mean really, properly, thoroughly shuffle—a deck of 100 cards. Thus, what I thought would be a fun exercise where the girls alternated which one drew 15 cards and which one tallied the colors on the white board quickly devolved into a “why is this taking so long?” battle of long stretches of card shuffling, slow drawing and slower tallying.

Perhaps I was still reacting to the news we would be home schooling five weeks longer than we had anticipated. Perhaps I was overtired—this is more exhausting than I had expected. Or perhaps I was mad at myself for choosing an overly-thick deck of cards I could not properly “randomize.”

Whatever the reason, I snapped multiple times at both daughters, making the older one huffy and the younger one teary. I apologized—again; Nell, who taught elementary school for more than a decade, gently pointed out this is why you do not teach children “at 5:30…they are toast.”

For all the drama, however, we managed to draw 15 sets of cards. As you see, the results were not what I had anticipated. The 21 yellow cards kept making a disproportionate appearance.

Sampling results March 25

Here is a graphical representation of the results. Had I not counted the cards very carefully, I would almost think I simply had the “true” totals reversed; it is more likely simply very difficult properly to shuffle a double-deck of cards…and that randomness does not guarantee anything.

Biased sampling March 25

Even teachers have things to learn from their own lessons.

Until next time…please stay safe and healthy…

2020 Super Tuesday contests: My final polling update

[Ed. note: A few hours after I published this, a complete set of Super Tuesday polls was released by Swayable (C+) and Data for Progress (B-/C+), as well as a Spry Strategies poll (C-/D+) of North Carolina. I did not update each state’s final WAPA, though I did update the projected distribution of pledged delegates.]

On March 3, 2020, for the first time during the 2020 Democratic presidential nominating process, multiple states—as well as American Samoa and Democrats Abroad—will hold contests on the same day; there is a reason this day is called “Super Tuesday.” Before examining those contests, you may review results from the Iowa Caucuses, New Hampshire Primary, Nevada Caucuses and South Carolina Primary.

Table 1 lists the 16 jurisdictions holding primaries on Super Tuesday, along with poll closing times and the number of pledged delegates each state will provide to the Democratic National Convention, which will be held July 13-16, 2020 in Milwaukee, Wisconsin.

Table 1: Democratic presidential nominating contests, March 3, 2020 by Poll Closing Times

Jurisidiction Poll Closing (EST) Pledged Delegates
Vermont 7 pm 16
Virginia 7 pm 99
North Carolina 7:30 pm 110
Alabama 8 pm 52
Maine 8 pm 24
Massachusetts 8 pm 91
Oklahoma 8 pm 37
Tennessee 8 pm 64
Texas 8 pm 228
Arkansas 8:30 pm 31
Colorado 9 pm 67
Minnesota 9 pm 75
Utah 9 pm 29
California 11 pm 415
American Samoa n/a 6
Democrats Abroad March 10 13
TOTAL PLEDGED DELEGATES 1,357

This Tuesday will be the closest the United States has ever come to a national presidential primary, with three contests in New England, five in the south, three in the center of the nation and three in the west; fully 34.1% of the 3,979 total pledged delegates will be awarded.[1] Just five states—California, Texas, North Carolina, Virginia and Massachusetts—will provide 943 (69.5%) of pledged delegates awarded on Super Tuesday. And as with the four previous Democratic presidential nominating contests, a candidate must win more than 15% of the vote to be awarded any delegates either statewide or within a Congressional district.

As of this writing, there were five declared candidates for the 2020 Democratic presidential nomination:

  • Former Vice President Joe Biden
  • Former New York City Mayor Michael Bloomberg
  • United States House of Representatives Member from Hawaii Tulsi Gabbard
  • United States Senator (“Senator”) from Vermont Bernie Sanders
  • Massachusetts Senator Elizabeth Warren

Bloomberg will actually be appearing on a ballot for the first time. And as I was writing this, Minnesota Senator Amy Klobuchar ended her campaign—and, like former South Bend, Indiana Mayor Pete Buttigieg, endorsed Biden.

In the remainder of this post, I present final WAPA (weighted-adjusted polling average), calculated multiple ways depending upon available data, for each candidate in each state, sorted by poll closing time; polls are up-to-date as of 2 am EST March 3, 2020. All publicly-available polls conducted since January 1, 2019 may be found here.

And here is my updated weighting scheme:

  • Polls conducted entirely or partially after February 3, 2020, but before February 12, 2020 are weighted 2.00 or 1.00+fraction[2] times, respectively, higher than polls conducted entirely before February 4, 2020.
  • Polls conducted entirely or partially after February 11, 2020, but before February 23, 2020 are weighted 3.00 or 2.00+fraction times, respectively, higher than polls conducted entirely before February 12, 2020.
  • Polls conducted entirely or partially after February 22, 2020, but before March 1, 2020 are weighted 4.00 or 3.00+fraction times, respectively, higher than polls conducted entirely before February 23, 2020.
  • Polls conducted entirely or partially after February 29, 2020, but before March 4, 2020 are weighted 5.00 or 4.00+fraction times, respectively, higher than polls conducted entirely before February 29, 2020.

To provide context for the percentage either truly undecided or selecting a different candidate (“DK/Other”), I also include the final state WAPA for Buttigieg, Klobuchar and billionaire activist Tom Steyer combined (“B/K/S”); their supporters consistently comprise a sizeable proportion of the “DK/Other” group adding a great deal of additional uncertainty to the outcomes of Tuesday’s races.

7 pm EST

Vermont

IMG_2671 (2)

Only one poll was conducted of Sanders’ home state, by Braun Research between February 4 and February 10, 2020; according to FiveThirtyEight.com’s pollster ratings, they have a B-/C+ rating.  Sanders led with 51%; no other candidate reached the 15% delegate threshold. It is very likely Sanders will accrue most if not all of the 16 available pledged delegates.

Virginia

Here is the breakdown of publicly-available polls of the 2020 Virginia Democratic Primary:

  • 8 since January 1, 2019
  • 5 since the Iowa Caucuses on February 3, 2020

Table 2: Final Virginia Primary WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls February 2020
Biden 28.1 29.4
Sanders 23.4 24.2
Bloomberg 14.2 14.7
Warren 10.1 10.5
Gabbard 0.7 0.6
DK/Other 23.5 20.5
B/K/S 12.0 11.5

Both Biden and Sanders appear to have some momentum in Virginia going into Super Tuesday. In fact, in two polls conducted entirely after the South Carolina Primary, albeit with a C/C+ average, Biden averages 43.5% to Sanders’ 26.5%; neither Warren nor Bloomberg top 12%. If this holds, Biden could easily win well over half of Virginia’s 99 pledged delegates, with Sanders winning most of the rest, and allowing for the possibility Bloomberg and/or Warren top the 15% threshold in at least one of the Commonwealth’s 11 Congressional districts.

7:30 pm EST

North Carolina

Here is the breakdown of publicly-available polls of the 2020 North Carolina Democratic Primary:

  • 22 since January 1, 2019
  • 13 since the Iowa Caucuses on February 3, 2020 (one poll overlapped)

 Table 3: Final North Carolina Primary WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls Post Iowa Caucuses
Biden 23.6 22.0
Sanders 22.8 23.8
Bloomberg 14.9 17.1
Warren 11.7 11.2
Gabbard 1.0 1.0
DK/Other 21.4 19.7
B/K/S 13.9 15.0

Similar to neighboring Virginia, it is difficult to discern momentum in these data, especially with just more than one in five potential voters either genuinely undecided or choosing a different candidate. Nonetheless, if these percentages are predictive, Sanders, Biden and Bloomberg will most likely divide the state’s 99 pledged delegates between them, though Warren could top the 15% threshold in at least one of the state’s 13 Congressional districts.

What these two Eastern seaboard southern states have in common is they will provide the first tests of two propositions:

  1. Biden has significant momentum, at least in southern states, from his landslide win in South Carolina.
  2. Bloomberg will fade following two poor debate performances.

8 pm EST

Alabama

Only two polls were conducted here, both in 2019: by Change Research between March 20 and March 23 and by Survey Monkey between July 2 and July 16; these pollsters have an average rating of C-/D+, so caution is urged. Still, Alabama resembles South Carolina in many ways, so Biden’s weighted average of 38.3%–with no other candidate topping 15%–is likely highly predictive; Biden could easily win the vast majority of the 52 available pledged delegates.

Maine

IMG_2771

Only five polls were conducted of the 2020 Maine Democratic Presidential Primary, of which two were conducted after the Iowa Caucuses: one by SocialSphere (B-/C+) between February 10 and 13, 2020, and one by Change Research (C) between March 1 and 2, 2020.

Table 4: Final Maine Primary WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls February 2020
Sanders 31.2 36.3
Biden 20.3 19.5
Warren 16.0 13.4
Bloomberg 8.9 11.5
Gabbard 0.9 1.1
DK/Other 21.6 16.7
B/K/S 8.4 8.2

Fellow New Englander Sanders would appear to have momentum in Maine, and could easily more than half of the state’s 24 pledged delegates, with Biden and, possibly, Warren splitting the remainder.

Massachusetts

In the interest of full disclosure, here is the relevant portion of the ballot I cast last Thursday in the 2020 Massachusetts Democratic Presidential Primary:

Voting for Warren 2020

Here is the breakdown of publicly-available polls of the 2020 Massachusettts Democratic Primary:

  • 11 since January 1, 2019
  • 5 since the Iowa Caucuses on February 3, 2020

Table 5: Final Massachusetts Primary WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls Post Iowa Caucuses
Sanders 21.5 22.3
Warren 21.1 20.5
Biden 13.3 12.1
Bloomberg 10.4 11.5
Gabbard 2.4 2.5
DK/Other 31.4 31.1
B/K/S 21.8 23.8

It is very likely Sanders and Warren—who may not win her home state on Super Tuesday—will divide the Commonwealth’s 91 pledged delegates roughly evenly between them, though I would not discount the possibility Biden and/or Bloomberg winning at least 15% of the vote in one or more of Massachusetts’ nine Congressional districts.

 Oklahoma

Only three polls have been conducted of the 2020 Oklahoma Democratic Presidential Primary, though two of those were conducted after the Iowa Caucuses: one by SoonerPoll.com (B-/C+) from February 17 to 21 and one by Cole Hargrave Snodgrass & Associates (B-/C+) from February 10 to 13.

Table 6: Final Oklahoma Primary WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls Post Iowa Caucuses
Bloomberg 18.1 19.2
Biden 17.3 17.3
Sanders 11.6 12.1
Warren 8.7 8.6
Gabbard 0.0 0.0
DK/Other 44.3 42.8
B/K/S 16.3 17.2

While this appears to be a battle between Bloomberg, Biden and Sanders, I think Biden is very likely to win a large majority of the state’s 37 pledged delegates. Still, the more than 40% not choosing one of the five remaining declared candidates makes the outcome of this contest especially uncertain.

Tennessee

Only one poll has been conducted of the 2020 Tennessee Democratic Primary, by SurveyMonkey (D-) between July 2 and July 16, 2019. Biden led the poll with 33%, however, signaling the possibility he will win the vast majority of the state’s 64 pledged delegates.

Texas

Here is the breakdown of publicly-available polls of the 2020 Texas Democratic Primary:

  • 35 since January 1, 2019
  • 12 since the Iowa Caucuses on February 3, 2020

Table 7: Final Texas Primary WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls Post Iowa Caucuses
Sanders 26.7 28.7
Biden 22.4 21.9
Bloomberg 13.8 16.8
Warren 12.7 12.6
Gabbard 1.3 1.4
DK/Other 23.1 18.6
B/K/S 11.9 13.1

Sanders and Bloomberg would appear to have momentum in Texas heading into Super Tuesday, though it is a much closer race for delegates between Sanders, Biden and Bloomberg, with Warren likely accumulating some pledged delegates in around one-third of the state’s 36 Congressional districts. Still, it would not be a surprise to see Sanders win a plurality of the state’s 228 pledged delegates, with Biden not too far behind him.

8:30 pm EST

Arkansas

Only one poll was conducted here, by Hendrix College (B-/C+) between February 6 and 7, 2020. Bloomberg “led” with 20%, followed by Biden (19%), Sanders and Buttigieg each with 16%, and Warren at 9%; Gabbard was not included in the poll. A total of 36% were either undecided or chose a different candidate, with 21% choosing either Buttigieg or Klobuchar (Steyer was not included). Assuming a Biden surge and a Bloomberg collapse, however, I would expect Biden to win at least a plurality, if not an outright majority of the state’s 31 pledged delegates, with most of the remainder going to Sanders.

9 pm EST

Colorado

Only five polls have been conducted of the 2020 Colorado Democratic Presidential Primary, though three of those were conducted after the Iowa Caucuses.

Table 8: Final Colorado Primary WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls Post Iowa Caucuses
Sanders 30.8 31.6
Warren 16.5 16.4
Biden 11.7 10.3
Bloomberg 10.2 11.4
Gabbard 0.7 0.7
DK/Other 30.1 29.6
B/K/S 16.8 18.0

Sanders is very likely to win Colorado, though perhaps “only” by a high-single-digit margin, thus accruing a majority of the state’s 67 pledged delegates. Still, Warren could potentially finish a close second, winning a few dozen delegates herself. Biden might also accrue a handful of pledged delegates here.

Minnesota

This, of course, is Klobuchar’s home state, so extrapolating from existing polls is extremely tricky. Moreover, only four polls have been conducted of the 2020 Minnesota Democratic Presidential Primary, though two of those were conducted after the Iowa Caucuses: one by Mason-Dixon Research & Polling Inc. (B+) from February 17 to 20 and one by University of Massachusetts, Lowell (A-/B+) from February 13 to 19.

Table 9: Final Minnesota Primary WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls Post Iowa Caucuses
Sanders 19.0 22.0
Warren 18.3 13.6
Biden 12.8 8.5
Bloomberg 3.0 6.1
Gabbard 1.3 2.5
DK/Other 46.7 47.3
B/K/S 29.9 (Klobuchar 21.8) 35.1 (Klobuchar 28.0)

As with Colorado, Sanders now seems poised to win Minnesota, though perhaps “only” by a high-single-digit margin over Warren and Biden. Those three are likely to win nearly all of the state’s 75 pledged delegates, barring Klobuchar still receiving a significant share of the vote, particularly among those who voted early.

Utah

Only two polls were conducted here, by Suffolk University (A-) between January 18 and 22, 2020 and by HarrisX (C+) between February 22 and 26, 2020. Sanders leads across these two polls with 27.7%, followed by Bloomberg at 16.5%, Warren at 14.7%, Biden at 7.7% and Gabbard at 0.3%; fully one-third of voters (33.1%) were either truly undecided or chose a different candidate, of whom more than half (18.4%) chose Buttigieg, Klobuchar or Steyer. While, as with the other states whose polls close at 9 pm, Sanders is the favorite to win at least a plurality of Utah’s 29 pledged delegates, Warren could once again surprise here with a strong second place showing, with Biden the wildcard.

11 pm EST

California

IMG_3510

Given the Golden State’s propensity for counting votes slowly, and the fact their 415 pledged delegates represent nearly one-third (31%) of those available on Super Tuesday, the full impact of these 16 contests on the race for the 2020 Democratic presidential nomination may not come fully into focus until early Wednesday morning, at the absolute earliest.

Here is the breakdown of publicly-available polls of the 2020 California Democratic Primary:

  • 53 since January 1, 2019
  • 16 since the Iowa Caucuses on February 3, 2020 (one conducted almost entirely afterward)

Table 10: Final California Primary WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls Post Iowa Caucuses
Sanders 28.7 31.5
Biden 17.8 16.3
Warren 15.2 14.1
Bloomberg 10.4 12.9
Gabbard 1.8 1.9
DK/Other 26.1 23.2
B/K/S 14.2 15.3

The only question about the 2020 California Democratic Primary is how large Sanders’ margin of victory will be. The corollary question is whether Warren and/or Bloomberg reach 15% statewide and/or in any of the state’s 53 Congressional districts. If not, Sanders could possibly win two-thirds of the state’s 415 pledged delegates, netting a minimum of 138 delegates over Biden in just one state.

Later

American Samoa, Democrats Abroad

There is no polling to indicate how the 19 total pledged delegates available from these two jurisdictions will be divided among the candidates.

**********

Perhaps the only two indisputable things we know about Super Tuesday are that a) Sanders appears to be the only candidate who is competitive everywhere, and b) there are an unusually high number of voters either truly undecided or choosing a different candidate in pre-primary polling.

To get a sense of just how good a day Sanders could have—or not–here is a very back-of-the-envelope pseudo-prediction for the number of pledged delegates each candidate will receive on Super Tuesday, based on the following extremely arbitrary assumptions:

  1. Excluding a slight uptick from “DK/Other” voters, Sanders’s Super Tuesday percentages will closely match his final WAPA
  2. Biden will see an uptick in this final WAPA equivalent to 20% of the final WAPA for Bloomberg.
  3. Bloomberg’s results will be 80% of his final WAPA, with no uptick from “DK/Other” voters
  4. B/K/S will split their votes this way:
    1. 50% for B/K/S
    2. 20% each for Biden and Warren
    3. 10% for Sanders
  5. The remaining “DK/Other” voters will split:
    1. 45% each for Biden and Warren
    2. 10% for Sanders
  6. The 19 pledged delegates from American Samoa and Democrats Abroad will split 7 Sanders, 7 Biden, 3 Warren, 2 Bloomberg
  7. There is no substantive difference between statewide and Congressional-district allocation of pledged delegates. This is by far the least-defensible assumption.

Based upon those quite rosy assumptions for Biden and Warren, here is my extremely timey-wimey, wibbly-wobbly not-quite prediction for the distribution of pledged delegates awarded on Super Tuesday:

Biden              470

Sanders          464

Warren           284

Bloomberg     139

I have to say, this rather surprised me—until I realized just how well Biden could do in Virginia, Tennessee, Alabama, Arkansas and Oklahoma, and that he is likely to make the 15% threshold in every state. However, if I distribute the B/K/S and “pure” DK/Other votes evenly between Biden, Sanders and Warren, the delegate allocation looks like this:

Sanders          501

Biden              451

Warren           263

Bloomberg     139

This small tweak in assumptions is the difference between Biden and Sanders being effectively tied in pledged delegates after Super Tuesday and Sanders having a nearly-70 delegate lead.

And that is a huge difference.

We shall see.

Until next time…

[1] An additional 764, at least, “automatic delegates” (also known as “superdelegates”)—mostly elected Democrats—would vote on a second ballot if not candidate clears the 1,991 vote threshold on the first ballot at the Democratic National Convention.

[2] Percentage of days the poll was being conducted were after the most recent primary or caucuses

2020 South Carolina Primary: How did my final polling averages fare?

Given the extremely volatile polling for the 2020 Democratic presidential nomination now that voting has commenced, I will not provide global monthly updates for the next few months. Instead, I will focus on the first handful of primaries and caucuses: Iowa on February 3, New Hampshire on February 11, Nevada on February 22, South Carolina on February 20, the 14 Super Tuesday contests on March 3, and so forth.

Here is my updated weighting scheme:

  • Polls conducted entirely or partially after February 3, 2020, but before February 12, 2020 are weighted 2.00 or 1.00+fraction[1] times, respectively, higher than polls conducted entirely before February 4, 2020.
  • Polls conducted entirely or partially after February 11, 2020, but before February 23, 2020 are weighted 3.00 or 2.00+fraction times, respectively, higher than polls conducted entirely before February 12, 2020.
  • Polls conducted entirely or partially after February 22, 2020, but before March 1, 2020 are weighted 4.00 or 3.00+fraction times, respectively, higher than polls conducted entirely before February 23, 2020.
  • Polls conducted entirely or partially after February 29, 2020, but before March 4, 2020 are weighted 5.00 or 4.00+fraction times, respectively, higher than polls conducted entirely before March 1, 2020.

And then there were six—though it is somewhat unclear whether United States House of Representatives Member (“Representative”) from Hawaii Tulsi Gabbard is still actively campaigning. Within 24 hours of the announcement of results from the 2020 South Carolina Democratic Presidential Primary, two candidates ended their bid to be the 2020 Democratic nominee for president: billionaire activist Tom Steyer and former South Bend, Indiana Mayor Pete Buttigieg. Both men ran optimistic, forward-looking and generally positive races, and they are to be commended for ending their campaigns with class and dignity.

Traffic light trees 2 10-12-2008

I also praise the South Carolina Democratic Party for disseminating results from >99% of precincts within five hours of polls closing in their state at 7 pm EST on February 29, 2020. Earlier that day, I published my final 2020 South Carolina Democratic Primary WAPA (weighted-adjusted polling average) for the eight then-declared Democratic presidential candidates, calculated seven different ways (Tables 1 and 2):

  • 53 since January 1, 2019
  • 41 since the 1st Democratic debate on June 26, 2019
  • 22 since the 5th Democratic debate on November 19, 2019
  • 18 since the 7th Democratic debate on January 14, 2020
  • 15 since the Iowa Caucuses on February 3, 2020
    • 7 between February 12 and February 22
    • 8 beginning February 23

Table 1: Final South Carolina WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate Since

Iowa

Biden 33.2 33.0 32.4 32.2 32.3
Sanders 18.6 18.7 20.3 20.7 20.7
Steyer 11.4 11.7 13.4 13.5 13.5
Warren 9.1 9.1 8.4 8.1 8.1
Buttigieg 7.7 7.7 8.5 8.7 8.8
Klobuchar 3.9 4.0 4.6 4.8 4.9
Gabbard 2.1 2.2 2.5 2.6 2.4
Bloomberg 0.3 0.3 0.3 0.2 0.2
DK/Other 13.7 13.3 9.6 9.1 9.1

Table 2: South Carolina WAPA for declared 2020 Democratic presidential nomination candidates following the Iowa Democratic Caucuses

Candidate Before Nevada Caucuses After Nevada Caucuses Change
Biden 26.0 36.5 +10.5
Sanders 21.2 20.6 -0.6
Steyer 15.7 12.9 -2.8
Warren 8.7 7.7 -1.0
Buttigieg 9.5 8.3 -1.2
Klobuchar 5.7 4.0 -1.8
Gabbard 2.1 2.2 +0.1
Bloomberg 0.7 0.0 -0.7
DK/Other 10.5 7.8 -2.7

Based solely on these numbers, one could reasonably draw the following conclusions:

  • Former Vice President Joe Biden was rapidly rising in the polls.
  • No other candidate was moving in the polls more than a few points one way or the other.

Comparing WAPA to results. Table 3 lists the results of the 2020 South Carolina Democratic Primary as of 7:30 pm EST on March 1, 2020:

Table 3: Percentage of vote received in 2020 New Hampshire Democratic Primary

Candidate % Vote
Biden 48.4
Sanders 19.9
Steyer 11.3
Buttigieg 8.2
Warren 7.1
Klobuchar 3.2
Gabbard 1.3
Bloomberg (not on ballot) 0.0
All others 0.6

Table 4 lists the arithmetic differences between each candidate’s final South Carolina Primary WAPA and each of the three reported measures; positive values indicate better performance in the primary than in the polls.

Table 4: Arithmetic difference between Vote % and WAPA, 2020 New Hampshire Democratic Primary

Candidate All

Polls

Since 1st Debate Since 5th Debate Since 7th Debate Since

 Iowa

Mean

Difference

Biden 15.2 15.4 16.0 16.2 16.1 15.8
Sanders 1.3 1.2 -0.4 -0.8 -0.8 0.1
Steyer -0.1 -0.4 -2.1 -2.2 -2.2 -1.4
Warren -0.9 -0.9 -0.2 0.1 0.1 -0.4
Buttigieg -0.6 -0.6 -1.4 -1.6 -1.7 -1.2
Klobuchar -0.7 -0.8 -1.4 -1.6 -1.7 -1.2
Gabbard -0.8 -0.9 -1.2 -1.3 -1.1 -1.1
Bloomberg -0.3 -0.3 -0.3 -0.2 -0.2 -0.3

With one glaring exception, these polling averages were remarkably accurate; only one candidate did not finish within 1.4 percentage points (“points”) in either direction of her/his final WAPA. But this was a substantial exception: Biden outperformed his final WAPA by an average of 15.8 points. Even then, however, the average “miss”—regardless of direction—was only 2.7 points. And limiting the comparison only to the eight polls released in the preceding week, Biden still outperformed by 11.9 points, with only Steyer missing by >1.0 point (-1.6).

Bottom line. To evaluate these comparisons globally, I calculated two difference measures for each of the five WAPA, excluding “DK/Other” (Table 5):

  1. Means of the absolute value of each candidate’s value in Table 4
  2. Sums of the squared differences (“SSE”) between each WAPA value and final results

Table 5: Global differences between WAPA and results, 2020 New Hampshire Democratic Primary

Polling period Mean AV Difference SSE
All Polls 2.5 235.1
Since 1st Debate only 2.6 241.5
Since 5th Debate only 2.9 266.1
Since 7th Debate only 3.0 274.8
Since Iowa Caucuses 3.0 271.7
Iowa-Nevada 4.4 536.2
Post-Nevada 2.2 147.8
Average 2.7 257.8

The five primary versions of WAPA were very accurate, even with Biden’s large overperformance, missing by between 2.3 and 3.0 points in either direction, on average, and with a fairly low and narrow range of SSE. For the first time, however, it was the older set of polls that was (barely) more accurate.

There is one glaring exception to this pattern, though, that demonstrates how difficult it is to poll voters during a fast-moving presidential nomination process. Polling conducted between the Iowa Caucuses on February 3 and the Nevada Caucuses on February 22 was far less predictive of the final results (though, to be fair, still reasonably accurate) than polling conducted in the week after the Nevada Caucuses; the latter was by far the most predictive of all, even with Biden’s overperformance.

Given the variation in patterns over the first four Democratic presidential nominating contests, I will continue to use this template to assess WAPA.

Now, on to the 16 Super Tuesday contests, with 1,344 pledged delegates at stake, on March 3, 2020!

Until next time…

[1] Percentage of days the poll was being conducted were after the most recent primary or caucuses

2020 South Carolina Primary: Final Polling Update

From 7 am to 7 pm EST on February 29, 2020, polls will be open in South Carolina for that state’s Democratic presidential primary. This election follows similar contests in Iowa and New Hampshire in which United States Senator (“Senator”) from Vermont Bernie Sanders and former South Bend, Indiana Mayor Pete Buttigieg fared very well, and a strong win by Sanders in the Nevada Caucuses–where former Vice President Joe Biden finished a strong second. Unlike those three states, though, the Democratic primary electorate in South Carolina is expected to be majority-black; three in five (61%) 2016 South Carolina Democratic Primary voters were black, as were 55% in 2008. This is also the final early nomination-related election in which former New York City Mayor Michael Bloomberg will not appear on the ballot.

To learn how I calculate candidate WAPA (weighted-adjusted polling average), please see here. Here is my updated weighting scheme:

  • Polls conducted entirely or partially after February 3, 2020, but before February 12, 2020 are weighted 2.00 or 1.00+fraction[1] times, respectively, higher than polls conducted entirely before February 4, 2020.
  • Polls conducted entirely or partially after February 11, 2020, but before February 23, 2020 are weighted 3.00 or 2.00+fraction times, respectively, higher than polls conducted entirely before February 12, 2020.
  • Polls conducted entirely or partially after February 22, 2020, but before March 1, 2020 are weighted 4.00 or 3.00+fraction times, respectively, higher than polls conducted entirely before February 23, 2020.

I have only been to South Carolina once, driving through it one March 1990 day as part of an epic round-trip road trip between Boston, Massachusetts and Atlanta, Georgia. I have no photographs from that trip, however, so here is one I took in Concord, Massachusetts in October 2008 of trees mimicking a traffic light.

Traffic light trees 2 10-12-2008

**********

Here is a breakdown of publicly-available South Carolina Primary polls as of midnight EST on February 29, 2020:

  • 53 since January 1, 2019
  • 41 since the 1st Democratic debate on June 26, 2019
  • 22 since the 5th Democratic debate on November 19, 2019
  • 18 since the 7th Democratic debate on January 14, 2020
  • 15 since the Iowa Caucuses on February 3, 2020
    • 7 between February 12 and February 22
    • 8 beginning February 23

Table 1: Final South Carolina WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate Since

Iowa

Biden 33.2 33.0 32.4 32.2 32.3
Sanders 18.6 18.7 20.3 20.7 20.7
Steyer 11.4 11.7 13.4 13.5 13.5
Warren 9.1 9.1 8.4 8.1 8.1
Buttigieg 7.7 7.7 8.5 8.7 8.8
Klobuchar 3.9 4.0 4.6 4.8 4.9
Gabbard 2.1 2.2 2.5 2.6 2.4
Bloomberg 0.3 0.3 0.3 0.2 0.2
DK/Other 13.7 13.3 9.6 9.1 9.1

Other than a slight uptick for Sanders and billionaire activist Tom Steyer—and an unsurprising reduction in the percentage either undecided or choosing an unlisted candidate—it is difficult to discern momentum in either direction for any candidate in Table 1. Biden has consistently led polls of the South Carolina Primary, hovering in the 32-34% range on average, with Sanders in second place at 18-21%. If these numbers are predictive, only Biden, Sanders and (possibly) Steyer seem poised to crack the 15% statewide threshold to be awarded pledged delegates to the Democratic National Convention this July; depending on how their support is distributed geographically, Massachusetts Senator Elizabeth Warren and Buttigieg could accrue some delegates in one or more Congressional districts.

However, the percentages calculated since the Iowa Democratic Caucuses on February 3 mask a substantial shift in support following the Nevada Democratic Caucuses on February 22 (Table 2).

Table 2: South Carolina WAPA for declared 2020 Democratic presidential nomination candidates following the Iowa Democratic Caucuses

Candidate Before Nevada Caucuses After Nevada Caucuses Change
Biden 26.0 36.5 +10.5
Sanders 21.2 20.6 -0.6
Steyer 15.7 12.9 -2.8
Warren 8.7 7.7 -1.0
Buttigieg 9.5 8.3 -1.2
Klobuchar 5.7 4.0 -1.8
Gabbard 2.1 2.2 +0.1
Bloomberg 0.7 0.0 -0.7
DK/Other 10.5 7.8 -2.7

After finishing 4th and 5th in the Iowa Caucuses and New Hampshire Primary, respectively, Biden appeared to be teetering on the edge of irrelevance, falling behind Sanders in my overall WAPA for the first time (data not shown). But his strong second-place finish in the Nevada Caucuses may have been just what Biden needed: his standing has soared more than 10 percentage points (“points”) in the South Carolina Primary since then. The percentage undecided or naming other candidates has dropped nearly three points over the same time span; Biden could well be picking up support from these late-deciders. At the same time, drops in support of between 1.2 and 2.8 points were registered by Steyer, Minnesota Senator Amy Klobuchar and Buttigieg .

One slight note of caution: using FiveThirtyEight.com’s pollster ratings, the eight pollsters assessing South Carolina Democratic voters before the Nevada Caucuses had a B average, while the seven pollsters in the field afterward had a B- average. Still, if this late shift in support in Biden is real—and it continues as voters make up their minds—Biden could earn 40% of the vote on Saturday, with Sanders the only other candidate to top 20%. The question then would be whether any other candidates accrue any delegates.

We shall see.

Until next time…

[1] Percentage of days the poll was being conducted were after the most recent primary or caucuses

2020 Nevada Caucuses: How did my final polling averages fare?

Given the extremely volatile polling for the 2020 Democratic presidential nomination now that voting has commenced, I will not provide global monthly updates for the next few months. Instead, I will focus on the first handful of primaries and caucuses: Iowa on February 3, New Hampshire on February 11, Nevada on February 22, South Carolina on February 20, the 14 Super Tuesday contests on March 3, and so forth.

Here is my updated weighting scheme:

  • Polls conducted entirely or partially after February 3, 2020, but before February 12, 2020 are weighted 2.00 or 1.00+fraction[1] times, respectively, higher than polls conducted entirely before February 4, 2020.
  • Polls conducted entirely or partially after February 11, 2020, but before February 23, 2020 are weighted 3.00 or 2.00+fraction times, respectively, higher than polls conducted entirely before February 12, 2020.
  • Polls conducted entirely or partially after February 22, 2020, but before March 1, 2020 are weighted 4.00 or 3.00+fraction times, respectively, higher than polls conducted entirely before February 23, 2020.

Table 1 below lists my final Nevada Caucuses WAPA (weighted-adjusted polling average) for the eight declared Democratic presidential candidates, calculated four different ways:

  • Using all 22 polls conducted since January 1, 2019
  • Using only the 19 polls released since the 1st Democratic debate on June 26, 2019
  • Using only the 10 polls released since the 5th Democratic debate on November 19, 2019
  • Using only the 7 polls released since the 7th Democratic debate on January 14, 2020

Feb 2020 lighthouse

Table 1: Final Nevada Caucuses WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate
Sanders 24.6 24.8 26.4 27.8
Biden 19.6 19.0 16.8 15.5
Warren 13.4 13.4 11.4 11.2
Buttigieg 10.9 11.0 12.6 13.7
Steyer 9.8 9.8 12.2 12.9
Klobuchar 6.5 6.3 8.3 9.4
Gabbard 1.4 1.3 1.5 1.5
Bloomberg 0.1 0.1 0.1 0.0
DK/Other 13.7 14.4 10.6 7.9

Based solely on these numbers, one could reasonably draw the following conclusions:

  • United States Senator (“Senator”) from Vermont Bernie Sanders; former South Bend, Indiana Mayor Pete Buttigieg; Minnesota Senator Amy Klobuchar and billionaire activist Tom Steyer were rising in the polls heading into the Nevada Caucuses.
  • Former Vice President Joe Biden and Massachusetts Senator Elizabeth Warren were declining in the polls.
  • No other candidate was moving in the polls in either direction.

Unlike results from the Iowa Democratic Caucuses conducted 19 days earlier, the following three unofficial, final caucus tabulations were reported by the Nevada Democratic Party within two days (Table 2), though not without some griping:

  1. Initial headcount of support for each Democratic candidate (“Initial tally”)
  2. Post-realignment headcount of support for each Democratic candidate (“Final tally”)
  3. Allocation of “county convention delegates,” or CCD’s, the only measure previously reported [ Note: I mistakenly referred to these as “state delegate equivalents” in the earlier post.]

Table 2: Final Nevada Democratic Caucuses results, February 3, 2020

Candidate Initial Tally Final Tally SDE’s
Sanders 34.0 40.5 46.8
Biden 17.6 18.9 20.2
Warren 12.8 11.5 9.7
Buttigieg 15.4 17.3 14.3
Steyer 9.1 4.1 4.7
Klobuchar 9.6 7.3 4.2
Gabbard 0.3 0.0 0.1
Bloomberg 0.0 0.0 0.0

The following three tables list the arithmetic differences between each candidate’s final Nevada Caucuses WAPA and each of the three reported measures; positive values indicate better performance in the Caucuses than in the polls.

Table 3: Arithmetic difference between Initial Nevada Caucuses % of vote and Nevada Caucuses WAPA

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate Mean

Difference

Sanders 9.4 9.2 7.6 6.2 8.1
Biden -2.0 -1.4 0.8 2.1 -0.2
Warren -0.6 -0.6 1.4 1.6 0.5
Buttigieg 4.5 4.4 2.8 1.7 3.3
Steyer -0.7 -0.7 -3.1 -3.8 -2.1
Klobuchar 3.1 3.3 1.3 0.2 2.0
Gabbard -1.1 -1.0 -1.2 -1.2 -1.1
Bloomberg -0.1 -0.1 -0.1 0.0 -0.1

Initial tally. If the Nevada Caucuses were instead the Nevada Primary, this would have been the only vote reported. On this measure Sanders, Buttigieg and Klobuchar outperformed their final WAPA percentages by 8.1, 3.3 and 2.0 percentage points (“points”), respectively. And for these three candidates, the closer in time the polls were to the Caucuses, the more “accurate” the WAPA. By contrast, Steyer performed an average of 2.1 points worse in the initial tally than his WAPA, with his percentages becoming less accurate over time. United House of Representatives Member from Hawaii Tulsi Gabbard also performed worse than her WAPA, with no noticeable effect of polling period. Finally, WAPA for Biden and Warren were very close, on average, to their initial tallies, with a minimal impact of polling period.

The overperformance of Sanders and Buttigieg aside, these polls were remarkably accurate. This is reassuring given the dearth and quality of recent Nevada Caucuses polling I detailed in the previous post.

Table 4: Arithmetic difference between Final Nevada Caucuses % of vote and Nevada Caucuses WAPA

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate Mean

Difference

Sanders 15.9 15.7 14.1 12.7 14.6
Biden -0.7 -0.1 2.1 3.4 1.2
Warren -1.9 -1.9 0.1 0.3 -0.8
Buttigieg 6.4 6.3 4.7 3.6 5.3
Steyer -5.7 -5.7 -8.1 -8.8 -7.1
Klobuchar 0.8 1.0 -1.0 -2.1 -0.4
Gabbard -1.4 -1.3 -1.5 -1.5 -1.4
Bloomberg -0.1 -0.1 -0.1 0.0 -0.1

Final tally. Only three candidates improved their vote totals after supporters of non-viable candidates shifted to a viable candidate (15% of attendees at a precinct caucus):

  • Sanders (+5,423 supporters; +6.5 points)
  • Buttigieg (+1,496; +1.9)
  • Biden (755; +1.3)

These three candidates also performed better in the final tally than their WAPA, on average; Sanders, in particular, outperformed these WAPA by a remarkable 14.6 points! As with the initial tally, WAPA using more recent polls was most predictive for Sanders and Buttigieg, with the opposite true for Biden. At the same time, Steyer lost 5,383 supporters—and 5.0 points—between initial and final tallies, as did Klobuchar (-2,724; 2.3) and Warren (-1,735, 1.3). For Steyer and Klobuchar, older polls were more predictive, with the opposite true for Warren.

Table 5: Arithmetic difference between Nevada Caucuses CCD % and Nevada Caucuses WAPA

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate Mean

Difference

Sanders 22.2 22.0 20.4 19.0 20.9
Biden 0.6 1.2 3.4 4.7 2.5
Warren -3.7 -3.7 -1.7 -1.5 -2.7
Buttigieg 3.4 3.3 1.7 0.6 2.3
Steyer -5.1 -5.1 -7.5 -8.2 -6.5
Klobuchar -2.3 -2.1 -4.1 -5.2 -3.4
Gabbard -1.3 -1.2 -1.4 -1.4 -1.3
Bloomberg -0.1 -0.1 -0.1 0.0 -0.1

CCDs. The same pattern holds for CCDs as for final vote tally.

  • Outperforming their WAPA were Sanders, by an extraordinary 20.9 points, along with Biden and Buttigieg
  • More recent WAPA were more predictive for Sanders and Buttigieg, and less so for Biden
  • Underperforming their WAPA were Steyer, by fully 6.5 points, along with Klobuchar, Warren and Gabbard, with the difference increasing with more recent polls for Steyer and Klobuchar
  • More recent WAPA were more predictive fore Warren; there was no polling period effect for Gabbard.

The bottom line. To evaluate these comparisons globally, I used the sum of the squared differences (“SSE”) between each WAPA value and the results value. Table 6 lists the SSE for each comparison; higher values indicate lower predictive power.

Table 6: SSE of differences between Nevada Caucuses Resuults and Nevada Caucuses WAPA

Polling period Initial Tally Final Tally CCDs
All Polls 124.0 332.1 551.5
Since 1st Debate 118.4 324.1 541.9
Since 5th Debate 80.8 293.7 508.5
Since 7th Debate 64.1 269.7 481.9

WAPA was most predictive of the initial tally, not surprising given that poll respondents are asked which candidate they planned to support upon arriving at the caucus site, and not about second or third choices. WAPA was more predictive of the final raw tally of supporters than of the distribution of CCDs, though neither was especially predictive; I attribute this to Sanders’ astonishing ability both to add supporters during realignment and to acquire at least 15% of the final tally in nearly every precinct caucus.

For each reported measure, WAPA was more predictive the closer the polls were to the Caucuses; this surprised me, given the candidate-specific differences detailed above. One explanation is that including older polls, however low-weighted, masks late polling movement of the kind that occurred to Sanders and Buttigieg.

For now, however, I will continue to report multiple versions of WAPA, if only to see if this pattern holds for later contests.

Now, on to South Carolina!

Until next time…

[1] Percentage of days the poll was being conducted were after the most recent primary or caucuses

2020 Nevada Caucuses: Final Polling Update

At 10 am Nevada time (1 pm EST) on February 22, 2020, Nevadans who did not already vote between February 15 and 18 by submitting a ranking of their three top choices to be the 2020 Democratic nominee will gather in nearly 2,100 meeting places to support their preferred candidate(s). While about 84,000 Nevadans caucused for Democrats in total in 2016, the number of Nevadans who voted early in 2020 is estimated to be at least 70,000.

This post presents four variations on my final pre-Caucuses WAPA, calculated using the four different timeframes detailed below. To learn how I calculate candidate WAPA (weighted-adjusted polling averages), please see here. Polls conducted partially or entirely after the Iowa Caucuses, but before the New Hampshire Primary, are weighted up to 1.33 to 2.00 times higher than polls conducted before the Iowa Caucuses. Polls conducted partially or entirely after the New Hampshire Primary are weighted 2.33 to 3.00 times higher than polls conducted before the New Hampshire Primary.

I have never been to Nevada, so here is the February 2020 lighthouse photograph in my Down East 2020 Maine Lighthouses wall calendar.

Feb 2020 lighthouse

**********

As of midnight EST on February 22, 2020, here is the breakdown of publicly-available Nevada Caucuses polls.

  • 22 conducted since January 1, 2019
  • 19 conducted since the 1st Democratic debate on June 26, 2019
  • 10 conducted since the 5th Democratic debate on November 19, 2019
  • 7 conducted since the 7th Democratic debate on January 14, 2020

First, however, here are some words of caution.

1. Very few high-quality polls

Only 22 polls of the Nevada Caucuses have been conducted since January 1, 2019, considerably fewer than the 59 polls of the Iowa Caucuses and the 72 polls of the New Hampshire Primary conducted over the same period of time. Overall, the 15 pollsters operating in Nevada, as a group, compare favorably to the 19 pollsters who operated in Iowa and the 21 who operated in New Hampshire: all three sets of pollsters averaged a B/B-, using FiveThirtyEight.com pollster ratings. However, only one of the seven polls conducted since the January 14, 2020 Democratic debate was conducted by a pollster who do have either a B/C (three) or a C+ (three) rating: an Emerson College poll (A-) conducted February 19-20, 2020. This poll is weighted 2.57; no other poll is weighted higher than 1.76.

The combination of many fewer polls, a dearth of recent high-quality polls and the fact polling in Nevada is extremely challenging in the best of circumstances makes these final Nevada Caucuses WAPA even more “wobbly” than those for the unpredictable Iowa Caucuses.

2. Reporting snafus 2.0?

Recall that in a caucus setting, voters literally declare their preferences in a public setting. If, after an initial tabulation, a candidate does not have the support of at least 15% of that site’s caucus attendees, supporters of that candidate either join a group that does (i.e., is “viable”), the “uncommitted” group or call it a day. This realigning continues until only viable candidates remain, at which time a complex formula is used to calculate “state delegate equivalents,” or SDEs. Prior to 2020, only the percentage of SDEs won were reported.

However, as was the case in Iowa, this year the Nevada Democratic Party plans to report three sets of results:

  • The initial statewide tally for each candidate
  • The post-viability tally for each candidate
  • SDE’s for each candidate

Having to report all three tallies in Iowa proved so difficult it ultimately cost the state’s Democratic Party Chair, Troy Price, his job. The Nevada Democratic Party has reassured us it is doing everything it can to report the results of its caucuses in a timely manner, though accuracy should always be prioritized over speed. There is a vast difference between conducting an election and reporting the results of that election; the former should take precedence over the latter, whatever the discomfiture level of journalists waiting to report results.

With those two caveats, Table 1 presents the final pre-Caucuses WAPA for the eight remaining candidates for the 2020 Democratic presidential nomination; even though former New York City Mayor Michael Bloomberg is not actively campaigning in Nevada, he is included as Nevadans could theoretically caucus for him.

Table 1: Final Nevada Caucuses WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate
Sanders 24.6 24.8 26.4 27.8
Biden 19.6 19.0 16.8 15.5
Warren 13.4 13.4 11.4 11.2
Buttigieg 10.9 11.0 12.6 13.7
Steyer 9.8 9.8 12.2 12.9
Klobuchar 6.5 6.3 8.3 9.4
Gabbard 1.4 1.3 1.5 1.5
Bloomberg 0.1 0.1 0.1 0.0
DK/Other 13.7 14.4 10.6 7.9

United States Senator (“Senator”) from Vermont Bernie Sanders would appear to have a significant edge—and momentum—heading into the final day of caucusing in Nevada. Increasingly falling behind Sanders are five candidates bunched relatively close together: former Vice President Joe Biden; Massachusetts Senator Elizabeth Warren; former South Bend, Indiana Mayor Pete Buttigieg; billionaire activist Tom Steyer and Minnesota Senator Amy Klobuchar. However, while Buttigieg, Steyer and Klobuchar appear to have some momentum heading into Saturday, Biden and Warren are fading somewhat, though we have no idea how Warren’s dominant February 19 debate performance will impact her final results. After a similarly-strong performance, Klobuchar outperformed her final New Hampshire WAPA by an average of 11.5 percentage points (“points”); were Warren to jump even one-third that amount—around four points—she could easily finish in 2nd place.

The bottom line.

As with the earlier Iowa Caucuses, anybody who thinks they know what will happen in the 2020 Nevada Democratic Caucuses has absolutely no idea what will happen in the 2020 Nevada Democratic Caucuses. Polling is very difficult to do in a multi-candidate race, with caucuses adding the additional wrinkle of unknowable “backup” choices or even who will participate. Combine this with a noticeable lack of recent high-quality polling, and we have something more akin to a gut-level hunch than a scientific projection.

It may very well be Sanders is the heavy favorite to win the Nevada Caucuses with literally every other candidate struggling to reach the 15% viability threshold, but I would not bet anything remotely of value on it.

We shall see.

Until next time…

2020 New Hampshire Primary: How did my polling averages fare?

Given the extremely volatile polling for the 2020 Democratic presidential nomination following the start of primary and caucus voting, I will not provide global monthly updates for the next few months. Instead, I will focus on the first handful of primaries and caucuses: Iowa on February 3, New Hampshire on February 11, Nevada on February 22, South Carolina on February 20, the 14 Super Tuesday contests on March 3, and so forth.

Also: I now weight higher polls conducted partially (1.33 or 1.67) or entirely (2.00) after February 3, 2020, than polls conducted entirely before February 4, 2020. I similarly weight higher polls conducted partially (2.33 or 2.67) or entirely (3.00) after Feb 11, 2020.

Unlike the molasses-slow pace it took for results from the 2020 Iowa Caucuses to be released—ultimately leading to the resignation of Iowa Democratic Party Chair Troy Price—95+% of the results from the 2020 New Hampshire Primary were tabulated and released within five hours of the final poll closings at 8 pm EST on February 11, 2020.

This improvement is undoubtedly due to the two “sharing size” bags of M&M’s—one plain, one peanut—I purchased at our local CVS after dropping our eldest daughter at her swim team workout. These candies have long been a staple of our election night watch parties, yet I neglected to buy some for the Iowa Caucuses the previous week.

Scan0046

Earlier that day, meanwhile, I had published my final New Hampshire Primary WAPA (weighted-adjusted polling average) for the 11 then-declared Democratic presidential candidates, calculated five different ways (Table 1):

  • 71 since January 1, 2019
  • 57 since the 1st Democratic debate on June 26, 2019
  • 35 since the 5th Democratic debate on November 19, 2019
  • 27 since the 7th Democratic debate on January 14, 2020
  • 15 since the Iowa Caucuses on February 3, 2020
    • 14 starting February 4 or later
    • 1 (Monmouth University) on February 3

Table 1: Final New Hampshire Primary WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate Since

Iowa

Sanders 23.0 23.1 24.8 25.5 26.4
Buttigieg 16.7 17.0 18.8 19.4 21.4
Biden 15.8 15.3 14.0 13.0 12.2
Warren 14.3 14.7 12.9 12.6 13.1
Klobuchar 7.0 7.2 8.6 9.3 9.7
Gabbard 3.4 3.5 3.7 3.5 3.2
Yang 3.2 3.3 3.6 3.4 3.2
Steyer 2.5 2.6 2.9 2.8 2.6
Bennet 0.4 0.5 0.5 0.5 0.6
Bloomberg 0.3 0.3 0.4 0.3 0.1
Patrick 0.3 0.3 0.4 0.4 0.4
DK/Other 13.1 12.2 9.6 9.2 7.1

Based solely on these numbers, one could reasonably draw the following conclusions:

  • United States Senator (“Senator”) from Vermont Bernie Sanders; former South Bend, Indiana Mayor Pete Buttigieg; and Minnesota Senator Amy Klobuchar were rising in the polls heading into the Iowa Caucuses.
  • Former Vice President Joe Biden and Massachusetts Senator Elizabeth Warren were declining in the polls.
  • No other candidate was moving in the polls one way or the other.

Comparing WAPA to results. Table 2 lists the results of the 2020 New Hampshire Democratic Primary as of 2 am EST on February 13, 2020:

Table 2: Percentage of vote received in 2020 New Hampshire Democratic Primary

Candidate % Vote
Sanders 25.8
Buttigieg 24.5
Klobuchar 19.9
Warren 9.3
Biden 8.4
Steyer 3.6
Gabbard 3.3
Yang 2.8
Write-in votes (including Bloomberg) 1.4
Patrick 0.4
Bennet 0.3
All others 0.4

Table 3 lists the arithmetic differences between each candidate’s final Iowa Caucuses WAPA and each of the three reported measures; positive values indicate better performance in the Caucuses than in the polls.

Table 3: Arithmetic difference between Vote % and WAPA, 2020 New Hampshire Democratic Primary

Candidate All

Polls

Since 1st Debate Since 5th Debate Since 7th Debate Since

 Iowa

Mean

Differnce

Sanders 2.8 2.7 1.0 0.3 -0.6 1.2
Buttigieg 7.8 7.5 5.7 5.1 3.1 5.8
Biden -7.4 -6.9 -5.6 -4.6 -3.8 -5.7
Warren -5.1 -5.5 -3.7 -3.4 -3.9 -4.3
Klobuchar 12.9 12.7 11.3 10.6 10.2 11.5
Gabbard -0.1 -0.2 -0.4 -0.2 0.1 -0.2
Yang -0.4 -0.5 -0.8 -0.6 -0.4 -0.5
Steyer 1.1 1.0 0.7 0.8 1.0 0.9
Bennet -0.1 -0.2 -0.2 -0.2 -0.3 -0.2
Bloomberg 0.4 0.4 0.3 0.4 0.6 0.4
Patrick 0.1 0.1 0.0 0.0 0.0 0.0

Two candidates—Buttigieg and Klobuchar—strongly outperformed their final New Hampshire Primary WAPA, by an average of 5.8 and 11.5 percentage points (“points”), respectively. The polling momentum they appeared to have coming out of the Iowa Caucuses was apparent in the final results, especially given that the polls conducted in the following week were the least inaccurate.

Sanders narrowly edged Buttigieg, 25.8% to 24.5%, to win the 2020 New Hampshire Democratic Primary, finishing an average 1.2 points higher than his WAPA; again WAPA was more accurate the closer in time it was calculated. Businessman Tom Steyer also finished nearly one point better in the voting (3.6%) than his average WAPA, which differed little in accuracy by time of calculation.

By contrast, Biden (-5.7 points) and Warren (-4.3 points) strongly underperformed their WAPA, with the more recent estimates again the least inaccurate.

One immediate consequence of these results is that three candidates, Colorado Senator Michael Bennet, former Massachusetts Governor Deval Patrick and entrepreneur Andrew Yang, ended their presidential campaigns. As with the 15 former candidates who preceded them, all three men did so with grace and class. One other candidate, United States House of Representatives member from Hawaii Tulsi Gabbard, finished 7th in New Hampshire, barely registered in Iowa and is not close to making any future debates, so her status in this race is somewhat shaky.

Bottom line. To evaluate these comparisons globally, I calculated two difference measures for each of the five WAPA, excluding “DK/Other” (Table 4):

  1. Means of the absolute value of each candidate’s value in Table 3
  2. Sums of the squared differences (“SSE”) between each of the five WAPA value and the results value.

Table 4: Global differences between WAPA and results, 2020 New Hampshire Democratic Primary

Polling period Mean AV Difference SSE
All Polls 3.5 317.4
Since 1st Debate only 3.4 304.2
Since 5th Debate only 2.7 207.7
Since 7th Debate only 2.4 172.4
Since Iowa Caucuses only 2.3 145.3

All five versions of WAPA were quite accurate, despite sharp late movement by Biden, Buttigieg, Klobuchar and Warren, missing by between 2.3 and 3.5 points in either direction, on average. Not surprisingly, the older the collection of polls used to calculate WAPA, the higher the average “miss.”

As with the Iowa Caucuses, meanwhile, and despite newer polls greatly outweighing older polls: the older the set of polls used to calculate New Hampshire Democratic Primary WAPA values, the less predictive they were of the actual results. And, again, the older-polls WAPA may shift too slowly to capture significant late movement.

Given this consistency over the first two contests, I will continue to use this template to assess WAPA.

Now, on to the Nevada Caucuses on February 22, 2020!

Until next time…

2020 New Hampshire Primary: Final Polling Update

[This poll was updated at 4:15 EST on February 11, 2020 to account for a Change Research poll (C) conducted February 8-9, 2020.]

At just after midnight EST on February 11, 2020, a total of 27 Democratic voters gathered in the small New Hampshire hamlets of Dixville Notch, Hart’s Location and Millsfield to cast the first votes in the 2020 New Hampshire Presidential Primary. Here are the tallies from those early votes (which, while fun to watch, have historically shown little relationship to the way the statewide electorate votes); I jotted them down as they were being broadcast live on MSNBC:

Table 1: 2020 New Hampshire Presidential Primary votes from three towns voting just after midnight

Candidate Dixville Notch Hart’s Location Millsfield Total
Klobuchar 0 6 2 8
Sanders 1 2 1 4
Warren 0 4 0 4
Yang 0 3 0 3
Biden 0 1 1 2
Buttigieg 1 0 1 2
Bloomberg 2 0 0 2
Steyer 0 1 0 1
Gabbard 0 1 0 1
TOTAL 4 18 5 27

The perhaps-surprising winner in these early returns is United States Senator (“Senator”) from Minnesota Amy Klobuchar, who had a very well-received debate performance in Manchester, New Hampshire on February 7; as the data in Table 2 suggest, she may have some momentum from that performance. Also, the name of former New York City Mayor Michael Bloomberg does not appear on the either the Democratic or Republican New Hampshire Primary ballot; these votes, as well as the only Republican vote cast in Dixville Notch, are write-in votes.[1]

To learn how I calculate candidate WAPA (weighted-adjusted polling average), please see here. I weight polls conducted partially (1.33 or 1.67) or wholly (2.00) after the Iowa Caucuses higher than prior polls.

Here is a photograph of a now-defunct ice cream parlor in scenic Portsmouth, New Hampshire in early September 2005; I was never the owner of this establishment:

Scan0046

**********

As of 2 a.m. EST on February 11, 2020, here is a breakdown of publicly-available New Hampshire Primary polls:

  • 72 since January 1, 2019
  • 58 since the 1st Democratic debate on June 26, 2019
  • 36 since the 5th Democratic debate on November 19, 2019
  • 28 since the 7th Democratic debate on January 14, 2020
  • 16 since the Iowa Caucuses on February 3, 2020
    • 15 starting February 4 or later
    • 1 (Monmouth University) on February 3

The final two tracking polls by Emerson College and the University of New Hampshire overlapped in time with each organization’s prior tracking polls, so I only used the most recent ones.

Table 2: Final New Hampshire WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate Since

Iowa

Sanders 23.0 23.1 25.0 25.7 26.6
Buttigieg 16.7 17.0 18.8 19.4 21.4
Biden 15.7 15.2 13.9 12.9 12.1
Warren 14.0 14.2 12.8 12.5 12.7
Klobuchar 7.0 7.2 8.5 9.3 9.6
Gabbard 3.5 3.6 3.8 3.6 3.3
Yang 3.3 3.4 3.6 3.5 3.3
Steyer 2.5 2.6 2.9 2.8 2.6
Bennet 0.4 0.4 0.5 0.5 0.5
Bloomberg 0.3 0.3 0.4 0.2 0.1
Patrick 0.3 0.3 0.4 0.4 0.5
DK/Other 13.4 12.6 9.4 9.2 7.3

Besides Klouchar, the top two finishers in the Iowa Caucuses—Vermont Senator Bernie Sanders and former South Bend, Indiana Mayor Pete Buttigieg—also appear to have momentum going into New Hampshire. The likelihood is that Sanders will win the primary—though not by anywhere close to his 2016 winning margin of 60% to 38% over former Secretary of State Hillary Clinton—perhaps followed by Buttigieg.

By contrast, former Vice President Joe Biden and, to a lesser extent, Massachusetts Senator Elizabeth Warren appear to be fading somewhat. United States of House of Representatives member Tulsi Gabbard of Hawaii, entrepreneur Andrew Yang and businessman Tom Steyer remain clustered around 2.5-3.7 percentage points, while Colorado Senator Michael Bennet and former Massachusetts Governor Deval Patrick appear to be non-factors at this point.

We shall see.

Until next time…

[1] President Donald Trump received 15 votes in Hart’s Location and 16 votes in Millsfield, while former Massachusetts Governor William Weld received 4 votes and 1 vote, respectively. Mary Maxwell received 1 vote in Hart’s Location.

2020 Iowa Caucuses: How did my polling averages fare?

Given the extremely volatile polling for the 2020 Democratic presidential nomination following the conclusion of the Iowa Caucuses, I will not provide global monthly updates for next few months. Instead, I will focus on the first handful of primaries and caucuses: Iowa on February 3, New Hampshire on February 11, Nevada on February 22, South Carolina on February 20, the 14 Super Tuesday contests on March 3, and so forth.

Also: I now weight polls conducted partially after February 3, 2020 either 1.333 or 1.667 times higher, and polls conducted entirely after February two times higher, than polls conducted entirely before February 4, 2020.

On the night of February 3, 2020, I was sitting on my usual spot on our sofa, watching MSNBC and anticipating returns from that day’s Iowa Caucuses.

Iowa Visitor Center Sep 1990

Earlier that day, I had published my final WAPA (weighted-adjusted polling average) for the 11 declared Democratic presidential candidates, calculated four different ways (Table 1):

  • Using all 58 polls conducted since January 1, 2019
  • Using only the 45 polls released since the 1st Democratic debate on June 26, 2019
  • Using only the 21 polls released since the 5th Democratic debate on November 19, 2019
  • Using only the 15 polls released since the 7th Democratic debate on January 14, 2020

Table 1: Final Iowa Caucuses WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate
Biden 19.9 19.8 20.1 20.3
Sanders 18.4 18.8 21.0 22.7
Warren 17.1 18.1 15.6 15.6
Buttigieg 15.9 16.8 16.7 16.7
Klobuchar 6.9 7.3 9.1 9.7
Yang 3.0 3.2 3.6 3.9
Steyer 2.8 3.1 3.1 3.5
Gabbard 1.5 1.6 1.5 1.6
Bloomberg 0.4 0.4 0.6 0.5
Bennet 0.3 0.3 0.2 0.3
Patrick 0.0 0.0 0.0 0.1
DK/Other 13.8 10.6 8.5 5.2

Based solely on these numbers, one would reasonably draw the following conclusions:

  • United States Senator (“Senator”) from Vermont Bernie Sanders and Minnesota Senator Amy Klobuchar were rising in the polls heading into the Iowa Caucuses, as to a lesser extent were entrepreneur Andrew Yang and businessman Tom Steyer.
  • Massachusetts Senator Elizabeth Warren was declining in the polls.
  • No other candidate was moving in the polls one way or the other.

By 11:37 pm EST, however, I had grown tired of waiting for results other than successive waves of entrance polls, so I tweeted the following:

RIP, Iowa Caucuses (1972-2020)

I have defended their idiosyncrasies for decades, believing the retail aspects of campaigning there outweighed the low-turnout mischegoss of the process.

 No more.

 This is ridiculous.

 #IowaCaucuses #iowacaucus2020

I will not relitigate here the myriad problems the Iowa Democratic Party had with tabulating, validating and releasing three distinct measures:

  1. Initial headcount of support for each Democratic candidate (“Initial tally”)
  2. Post-realignment headcount of support for each Democratic candidate (“Final tally”)
  3. Allocation of “state delegate equivalents,” or SDE’s, the only measure ever previously reported

Moreover, my annoyance has abated since Monday night, primarily because I suspect these vote-reporting snafus revealed that the byzantine process of converting persons standing in rooms, then possibly standing in different parts of the room, into SDE’s has always been “riddled with errors and inconsistencies,” to quote a recent New York Times headline. And if this marks the beginning of the end of using caucuses to allocate delegates to each party’s nominating conventions, so be it; they are undemocratic, exclusionary and overly complex.

As for which states “should” come first in future presidential nominating processes, I am currently agnostic.

Three days later, we finally have near-final results from the Iowa Caucuses (Table 2):

Table 2: Near-final Iowa Democratic Caucuses results, February 3, 2020

Candidate Initial Tally Final Tally SDE’s
Biden 15.0 13.7 15.8
Sanders 24.8 26.6 26.1
Warren 18.4 20.2 18.0
Buttigieg 21.3 25.0 26.2
Klobuchar 12.7 12.3 12.3
Yang 5.0 1.0 1.0
Steyer 1.7 0.2 0.3
Gabbard 0.2 0.0 0.0
Bloomberg 0.1 0.0 0.0
Bennet 0.1 0.0 0.0
Patrick 0.0 0.0 0.0
Uncommitted 0.6 0.1 0.2

The following three tables list the arithmetic differences between each candidate’s final Iowa Caucuses WAPA and each of the three reported measures; positive values indicate better performance in the Caucuses than in the polls.

Table 3: Arithmetic difference between Initial Iowa Caucuses % of vote and Iowa Caucuses WAPA

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate Mean

Difference

Biden -4.9 -4.8 -5.1 -5.3 -5.0
Sanders 6.4 6.0 3.8 2.1 4.6
Warren 1.3 0.3 2.8 2.8 1.8
Buttigieg 5.4 4.5 4.6 4.6 4.8
Klobuchar 5.8 5.4 3.6 3.0 4.5
Yang 2.0 1.8 1.4 1.1 1.6
Steyer -1.1 -1.4 -1.4 -1.8 -1.4
Gabbard -1.3 -1.4 -1.3 -1.4 -1.4
Bloomberg -0.3 -0.3 -0.5 -0.4 -0.4
Bennet -0.2 -0.2 -0.1 -0.2 -0.2
Patrick 0.0 0.0 0.0 -0.1 0.0
DK/Other -13.2 -10.0 -7.9 -4.6 -8.9

Initial tally. If the Iowa Caucuses were instead the Iowa Primary, this would have been the only vote reported. On this measure Sanders, Klobuchar and former South Bend, IN Mayor Pete Buttigieg averaged 4.5-4.8 percentage points (“points”) higher in the initial tally than in their WAPA. And the closer in time the polls were to the Iowa Caucuses, the more “accurate” the WAPA.

Warren (+1.8 points) and Yang (+1.6) also overperformed their WAPA in the initial tally, albeit by smaller margins. And for Warren, older polls were more predictive than recent polls.

By contrast, former Vice President Joe Biden did an average of 5.0 points worse in the initial Iowa Caucuses tally than his WAPA. Steyer and United House of Representatives Member from Hawaii Tulsi Gabbard (-1.4 each) also performed somewhat worse than their WAPA.

Table 4: Arithmetic difference between Final Iowa Caucuses % of vote and Iowa Caucuses WAPA

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate Mean

Difference

Biden -6.2 -6.1 -6.4 -6.6 -6.3
Sanders 8.2 7.8 5.6 3.9 6.4
Warren 3.1 2.1 4.6 4.6 3.6
Buttigieg 9.1 8.2 8.3 8.3 8.5
Klobuchar 5.4 5.0 3.2 2.6 4.1
Yang -2.0 -2.2 -2.6 -2.9 -2.4
Steyer -2.6 -2.9 -2.9 -3.3 -2.9
Gabbard -1.5 -1.6 -1.5 -1.6 -1.6
Bloomberg -0.4 -0.4 -0.6 -0.5 -0.5
Bennet -0.3 -0.3 -0.2 -0.3 -0.3
Patrick 0.0 0.0 0.0 -0.1 0.0
DK/Other -13.7 -10.5 -8.4 -5.1 -9.4

Final tally. Only three candidates improved their vote totals after supporters of non-viable candidates shifted to a viable candidate (15% of attendees at a precinct caucus):

  • Buttigieg (+5,638 supporters; +3.7 points)
  • Warren (+2,238; +1.8)
  • Sanders (+2,155; +1.8)

These three candidates, as well as Klobuchar (-1,288; -0.4), performed better in the final tally than their WAPA, on average. As with the initial tally, WAPA using more recent polls was most predictive for Sanders, Buttigieg and Klobuchar, while WAPA using older polls was most predictive for Warren.

Biden, on the other hand, lost 2,693 supporters and dropped 1.3 points between the initial and final tallies; Yang and Steyer also lost considerable support between the initial and final tallies. For all three candidates, WAPA using earlier polls was most predictive.

Table 5: Arithmetic difference between Iowa Caucuses SDE % and Iowa Caucuses WAPA

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate Mean

Difference

Biden -4.1 -4.0 -4.3 -4.5 -4.2
Sanders 7.7 7.3 5.1 3.4 5.9
Warren 0.9 -0.1 2.4 2.4 1.4
Buttigieg 10.3 9.4 9.5 9.5 9.7
Klobuchar 5.4 5.0 3.2 2.6 4.1
Yang -2.0 -2.2 -2.6 -2.9 -2.4
Steyer -2.5 -2.8 -2.8 -3.2 -2.8
Gabbard -1.5 -1.6 -1.5 -1.6 -1.6
Bloomberg -0.4 -0.4 -0.6 -0.5 -0.5
Bennet -0.3 -0.3 -0.2 -0.3 -0.3
Patrick 0.0 0.0 0.0 -0.1 0.0
DK/Other -13.6 -10.4 -8.3 -5.0 -9.3

SDEs. The same pattern holds for SDEs as for final vote tally, with one minor modification.

  • Buttigieg, Sanders and Klobuchar outperformed their WAPA, with the difference decreasing with more recent polls
  • Warren outperformed her WAPA, with the difference increasing with more recent polls
  • Biden, Steyer and Yang underperformed their WAPA, with the difference increasing with more recent polls.

The bottom line. To evaluate these comparisons globally, I used the sum of the squared differences (“SSE”) between each WAPA value and the results value. Excluding “DK/Other,” Table 6 lists the SSE for each comparison; higher values indicate lower predictive power.

Polling period Initial Tally Final Tally SDEs
All Polls 136.5 240.5 224.9
Since 1st Debate 115.8 210.8 198.2
Since 5th Debate 88.3 190.4 168.0
Since 7th Debate 77.1 177.8 156.1

WAPA was most predictive of the initial tally, not surprising given that poll respondents are asked which candidate they planned to support upon arriving at the caucus site, and not about second or third choices. WAPA was also slightly more predictive of the distribution of SDEs than of the final raw tally of supporters, though neither was especially predictive.

For each reported measure, WAPA was more predictive the closer the polls were to the Caucuses; I will admit this rather surprised me, given the candidate-specific differences detailed above. One explanation is that including older polls, however low-weighted, masks late polling movement of the kind that occurred to Sanders, Buttigieg and Klobuchar.

For now, however, I will continue to report multiple versions of WAPA, if only to see if this pattern holds for later contests.

Now, on to New Hampshire!

Until next time…