Tag Archives: Florida

The Wild Polling Spectrum

By Jim Ellis

July 27, 2020 — We’ve seen a plethora of presidential state polls released in the past couple of days, and the results seem to swing wildly among the various pollsters. While most studies have shown former vice president Joe Biden at present winning most of the key states, his margins don’t seem particularly secure even while his national numbers appear stronger.

Let’s explore the polling divergence, and remember that sample selection, weighting, size of the respondent universe, and length of time to complete the survey process are all important variables in determining a ballot test result. Also, the huge number of people refusing to answer their phones or not participating in political surveys means that pollsters must further rely on weighting and mathematical supposition to compensate for the lack of live responses.

With that short background, let’s look at some of the key states producing the widest swings:

FLORIDA

• Quinnipiac University – July 16-20 (924 registered voters; live interview)

Biden	51
Trump	38
Biden	plus 13

• St. Pete Polls – July 13-14 (3,018 likely voters via automated system)

Biden	50
Trump	44
Biden	plus 6

Though Biden is up in both polls, the margin difference is stark. The Q-Poll, brandishing such a wide spread in what is typically a close political state raises reliability questions.

GEORGIA

• Garin Hart Yang Research – July 9-15 (800 likely voters)

Biden	47
Trump	43
Biden	plus 4

• Spry Strategies – July 11-16 (701 likely voters; IVR)

Trump	49
Biden	46
Trump	plus 3

Here we see two different leaders within two different polling samples within the same virtual period. Republicans tend to under-poll in the southern states, which could certainly be present in at least the GHY poll.

Continue reading →

Turnout 2020: Up, then Down

By Jim Ellis

June 8, 2020 — In most political campaigns, the final electoral result is determined not necessarily from transforming undecided individuals into positive votes, but rather ensuring that the candidate’s committed supporters actually cast their ballot. Therefore, accurately projecting and influencing voter turnout becomes critical for every campaign.

Before the COVID-19 virus struck, many analysts and political prognosticators were predicting a record turnout in the 2020 general election, thus exceeding 2016’s all-time high 136.8 million presidential election ballots. Many stated that breaking 150 million voters was possible, with some even believing that was likely. The post-COVID primary vote participation figures now suggest otherwise, however.

There is a big difference in voter turnout before and after the COVID-19 virus attack. Prior to the March 18 societal shutdown, 25 states had held presidential primary or major caucus elections, meaning up to and including the March 17 election date in Arizona, Florida, and Illinois. North Dakota and Wyoming, because of the small attendance figures in their caucuses and reporting system, are not included in this matrix.

By mid-March, former vice president Joe Biden had broken away from the pack of Democratic candidates, and all of his major opponents had either dropped out of the race or were headed down that path. When voters cast their ballots on March 17, only Sen. Bernie Sanders (I-VT) and Rep. Tulsi Gabbard (D-HI) were advancing to the April 7 primary in Wisconsin.

Through March 17, Democratic primary turnout was up substantially from 2016, and on projected pace to meet the high turnout general election predictions if such a trend continued throughout the remainder of the election year. Republican turnout was down substantially in comparison to 2016, but that is obviously because President Trump had no serious opposition for re-nomination. Therefore, only the Democratic turnout numbers are viable for making statistically relevant calculations and projections.

Through the 25 tested presidential primaries ending March 17, turnout was up 14.8 percent when compared to the open race four years ago in the 17 states that held primary or major caucus elections in both 2016 and 2020. Since the COVID shutdown, however, Democratic voter participation has fallen. In the 11 post-COVID states that held Democratic primary elections in both 2016 and 2020, turnout dropped 21.2 percent when comparing the participation figures from the aforementioned election years.

Continue reading →

Nevada: A Target?

By Jim Ellis

May 8, 2020 — The Silver State of Nevada, with six Electoral College votes, has been regarded as a swing state in most 21st Century presidential elections, but in projecting the 2020 vote, it is routinely considered as a place destined to land in the Democratic column. But, could Nevada ultimately be in play for President Trump?

With a general election electoral vote map looking ever closer as we move toward November, introducing a new target could drastically change the eventual outcome or at least the campaign focus and strategy.

A new ALG Research poll (April 27-30; 76 likely Nevada general election voters) reports findings that are consistent with virtually all of the 2016 Nevada general election surveys. Last week, the firm found former vice president Joe Biden leading President Trump with a 49-45 percent spread. Such a margin and preference percentages fall into the same realm as all 31 polls conducted in Nevada from May through the November 2016 election.

When ‘16 ended, Hillary Clinton carried the state, but her margin was only 2.4 percentage points, meaning a 27,000-vote spread of more than 1.12 million ballots cast. And, consistent with the large number of polls that concluded a close race within the 40s would be the actual result, Clinton defeated Trump, 47.9 – 45.4 percent, validating the plethora of research conducted over the final five-plus months of that election cycle.

Four years ago, 13 different pollsters conducted the almost three dozen surveys, and in no instance did either Trump or Clinton ever reach the 50 percent plateau. On the other end of the spectrum, in just three instances did one of the candidates dip below 40 percent. Therefore, in 28 of the 31 studies logged within the 2016 Real Clear Politics polling archives, both candidates fell within the 40s, meaning this result occurred over 90 percent of the time. In terms of range, the span stretched from Clinton plus-7 to Trump plus-6, but the average between the two was only 2.7 percent, which is almost exactly the actual final total. This obviously suggests a competitive political battlefield.

Continue reading →

COVID & Redistricting

By Jim Ellis

April 29, 2020 — The deadline for the Census Bureau to release the new population data is March 31, 2021, but with the entire process being delayed due to COVID-19 precautions, the ability to meet the requirement is becoming more difficult by the day. Already, the Bureau has been delayed in dispatching their door-to-door teams necessary in obtaining the responses from people who did not return their mail tabulation form.

The Trump Administration is reportedly suggesting that the March 31 deadline be postponed to sometime in the summer of 2021. If this happens, we will see a series of redistricting problems ignited in the states. First, the political leaders in New Jersey and Virginia, places that have 2021 elections and need their new state legislative lines in place well before that date, would find themselves in a difficult position.

Initially, the two states would certainly have to postpone their primary elections because both nominate their general election candidates in June. Beyond that, it is possible they would have to even postpone their general elections into 2020 or run in the obsolete boundaries that were drawn back in 2011. In either case, we could expect lawsuits being launched from whichever party loses a particular electoral contest.

Other states would be affected, too. Many have legal deadlines in place mandating that the new redistricting maps for state legislature and the US House delegation be adopted before the legislative sessions ends. Most states recess before mid-summer, which would mean special sessions being called if the legislature is to act.

The problem intensifies in the states that are either gaining or losing congressional districts in reapportionment. Currently, it appears that seven states will add seats to their delegations (the best projections suggest that Texas will gain three, Florida two, and Arizona, Colorado, Montana, North Carolina, and Oregon one apiece), while 10 will lose single districts (Alabama, California [for the first time in history], Illinois, Michigan, Minnesota, New York, Ohio, Pennsylvania, Rhode Island, and West Virginia).

The aforementioned apportionment is based upon calculations released publicly and not, of course, using the actual numbers. Therefore, we could see some differences between these projections and what the formulas actually produce when the Census Bureau finally can produce the updated real figures.

Continue reading →

Polling At Odds

By Jim Ellis

April 24, 2020 — A series of surveys were conducted by several independent pollsters in the most critical states that will likely determine the outcome of the 2020 presidential campaign, and the results are somewhat conflicting.

Largely consistent data comes in polls from three pollsters: Fox News, Ipsos, and Quinnipiac University. They surveyed the electorates in Florida (Quinnipiac), Michigan (Fox News and Ipsos), Pennsylvania (Fox News and Ipsos), and Wisconsin (Ipsos). All of the studies were conducted during the April 15-21 period. The pollsters did not collaborate, each directing their own surveys individually.

The results in all of these states found former vice president Joe Biden leading President Trump, obviously a strong positive sign for Biden since the candidate carrying the preponderance of these particular states will win the national election.

The Biden advantage in each poll ranges from three percentage points (Ipsos in Wisconsin) to eight (Fox News in both Pennsylvania and Michigan; Ipsos in Michigan).

Florida is one of the core states for President Trump, meaning that it’s one of the five most important places that he must carry to win the election. Here, Quinnipiac finds Biden leading the incumbent, 46-42 percent, which is more exaggerated than most current Sunshine State polls. We must also acknowledge that Republicans have been under-polling in Florida during recent elections by approximately two percentage points. If typical Florida political trends continue through this election, we will see a very close final tally.

The Great Lakes states covered in this Update are all in the swing category, and President Trump will have to at least carry one of them to claim a national victory. Conversely, if the President holds his five core states (Arizona, Florida, Georgia, North Carolina, and Texas), then Biden would likely be forced to carry all of the Great Lakes’ swings: Michigan, Pennsylvania, and Wisconsin.

Another research entity is also in the picture, however. Change Research, polling for CNBC (April 17-18; 5,787 likely voters in Arizona, Florida, Michigan, North Carolina, Pennsylvania, and Wisconsin) also tested the critical battleground states just after the other pollsters completed their questioning phase. The CNBC poll, as well, was an independent undertaking and not executed in conjunction with any other research firm.

Continue reading →

This Time in 2016

By Jim Ellis

Does Arizona hold the key in a Trump-Biden candidacy?

April 23, 2020 — It is interesting to revert back to April of 2016 to see just how Donald Trump was faring against Hillary Clinton and compare those results with today’s survey research. Thanks to the Real Clear Politics website and their polling archives, the day-by-day polling data from four years ago is still available so we can track the Trump-Clinton campaign historical progress with the new Trump-Joe Biden impending national contest.

The key point to remember about national presidential polling is that the aggregate ballot test means very little yet is the subject of most political research studies. Knowing what voters think about the campaign in places like Arizona, Florida, Georgia, Michigan, North Carolina, Pennsylvania, Texas, and Wisconsin is much more important when attempting to project a final outcome, but we see far fewer numbers coming from these places than we do nationally.

In 2016, we will remember that almost all analysts and political prognosticators were predicting a Clinton win, and virtually all national polling was revealing an advantage for the former secretary of state, US senator, and First Lady, yet Trump emerged the winner. After the election, most surface analysis reported that the polling was in error, but such was not generally the case. The preponderance of polling, which predicted a narrow Clinton popular vote victory was actually correct; as we will remember, Clinton finished ahead of Trump in a close plurality.

With this background in mind, let’s look at what the various polling firms are projecting this month for the Trump-Biden race and compare it to the available data from 2016.

In April 2016, through the 15th of the month, three national polls had been released from individual or collaborating media entities: CBS News (April 8-12), NBC News/Wall Street Journal (April 10-14), and Fox News (April 11-13). This year, we see a more active April polling month that yields nine studies from eight different pollsters.

In 2016, the three testing entities all predicted Hillary Clinton to be holding the advantage over Donald Trump, by margins of 10 (CBS), 11 (NBC/WSJ), and 7 (Fox) percentage points. A little over six months later, Clinton would carry the national popular vote by 2.1 percentage points but lose the presidency because the key states broke toward Trump, which added up to an Electoral College win.

Continue reading →

Sanders Out;
Focus Now on Trump-Biden

By Jim Ellis

Sen. Bernie Sanders

April 9, 2020 — Sen. Bernie Sanders (I-VT) suspended his presidential campaign yesterday, therefore making former vice president Joe Biden the Democratic Party’s unofficial nominee. Biden, still 766-bound delegate votes away from clinching a first-ballot victory is now unencumbered in his bid to become the party standard bearer. It is likely that he will secure the 1,991 bound first-ballot delegate votes once the June 2 primary — now featuring 10 states — is held.

Sen. Sanders conceded that he could not overcome Biden’s strong lead but stopped short of endorsing him, though it is clear that he eventually will, and called for the Democratic Party to pull together in order to defeat President Trump.

How will a Trump-Biden general election campaign unfold? Very likely, the race will come down to what happens in about 10 states. In 2016, President Trump defeated Hillary Clinton with an Electoral College margin of 306-232, giving him a 36-vote cushion against Biden. This is a relatively substantial margin, but when remembering that three critical states containing 46 electoral votes came down to an aggregate vote spread of just over 77,000 votes, such a gap could quickly dissipate.

To win again, President Trump must keep intact five states that he carried as part of his 2016 coalition, three of which are giving signs of moving closer to the political center since the last election, and two that are always in the swing category. Arizona, Texas, and Georgia are must-wins for the Trump campaign, but these states are no longer locks for the Republican nominee. Though they should still remain part of the 2020 Trump coalition, they cannot be taken for granted.

Florida and North Carolina are always swing states, and any Republican presidential nominee must carry them in order to win the national election. The Democrats, because they win most of the other big states, can claim a national victory without Florida and North Carolina but a Republican cannot.

Continue reading →