Tag Archives: polling

Polling Accuracy

By Jim Ellis — Tuesday, Nov. 19, 2024

Polling

President-Elect Donald Trump / Photo by Gage Skidmore

There has been a great deal of controversial discussion about the 2024 political polling. Many continue to raise questions about the major polling firms’ cumulative accuracy rating, but is the criticism fair?

The perception that the polling industry missed President-Elect Donald Trump’s “landslide” win is not particularly accurate. Though Trump swept the battleground states, the margins were close, and virtually all of the 2024 presidential pollsters forecast close races in the critical states, which is the end result.

A typical polling margin of error is three percentage points. Using this as the accuracy scale, the Real Clear Politics polling average suggests that across the battleground board, the average survey fell within such a range.

In national polling, the ending Real Clear Politics average found Kamala Harris cumulatively leading by the smallest of margins, at 0.1 percent. These numbers include 24 polls from 14 different polling organizations.

While they did miss the projection of which candidate was leading, the cumulative margin was well within the margin of error. Since Trump is winning the national popular vote by 1.8 percentage points, the polling community average of 0.1 point separating the two candidate was a miss of 1.9 percentage points, but well within the margin of error.

In the battleground states, the least accurate polling came in Arizona. There, the Real Clear Politics average suggested a 2.8 percent lead for Trump, but he carried the state by a 5.5 percent margin. This was a 2.7 point miss.

The other states where the polling mark was two points off or more came in North Carolina and Nevada. The NC polling average suggested a Trump win margin of 1.2 percentage points. In actuality, the president-elect carried the state by 3.2 percent. In Nevada, the miss was a bit worse.

Pollsters pegged Trump with a lead of 0.6 percent, but he carried the Silver State with a 3.1 percent margin, thus the cumulative miss was 2.5 points.

The most accurate of the state presidential polling came in Georgia. The average projection suggested a Trump lead of 1.3 percent, but he won the state with a 2.2-point margin. Thus, the average poll figure missed the actual result by 0.9 points, again well within the polling margin of error.

In the seven battleground states, the polling community correctly projected five of the seven Trump wins. The two misses were Michigan and Wisconsin, but both fell only a half-point or less off the pace.

The one consistent error point in all of the battleground states, and nationally, was underestimating the Republican vote strength. This has been a consistent pattern during the Trump era, and it happened again in 2024 but to a lesser degree.

The numbers in the 11 competitive Senate races were not as accurate as the presidential projections.

The least accurately polled state was Florida where Sen. Rick Scott (R) recorded almost a 13-point victory, yet the polling average for his race against former Congresswoman Debbie Mucarsel-Powell (D) suggested a much closer outcome. The two were separated by an average of 4.6 points in Sen. Scott’s favor.

A similar pattern, but to a lesser degree, occurred in Texas. For most of the race, the polling average showed a close race between Sen. Ted Cruz (R) and Rep. Colin Allred (D-Dallas), but the end result proved different. While polling found Cruz extending his small lead within the race’s final stage to an average of 4.4 points, the end result yielded the two-term senator an 8.6 percent win, meaning a polling miss of 4.2 points, well outside the polling margin of error and far beyond the early race pattern.

The most accurately polled Senate race was Arizona, where the cumulative average found Rep. Ruben Gallego (D-Phoenix) leading former news anchor Kari Lake (R) by an average of 2.6 points in the later stages of the race, after he posted larger early advantages. Gallego would win the statewide race by 2.4 percentage points meaning a polling miss of only 0.2 percent.

As in the presidential race, the consistent polling flaw was under-counting the Republican support. In 10 of the 11 monitored races, the Republican candidate exceeded the suggested polling support figure. The only contest where the Republican candidate failed to top the polling average came in Montana where challenger Tim Sheehy (R) defeated Sen. Jon Tester (D) by 7.1 percentage points. The polling average in the race’s latter stage was 7.6 percent.

The polling community correctly projected the Senate winner in 10 of the 11 monitored races. The only contender to defy the polling projections was Pennsylvania Republican David McCormick (R) who many media outlets have projected as a winner opposite Sen. Bob Casey Jr. (D). The race is in a recount where McCormick maintains a lead of better than 25,000 votes.

For the most part, the repeated polling criticism is exaggerated. Overall, the pollsters had a relatively good year, and did correctly see the Trump polling trend in that he won the battleground states and the national popular vote by largely predicted close margins. The consistent underestimating of Republican support, however, does indicate future methodology correction is warranted.

Challenging the US House
Real Clear Politics Ratings

By Jim Ellis — Tuesday, Oct. 1, 2024

Polling

In a review of the national US House picture, we looked at various published race ratings. Those from Real Clear Politics are, to a degree, eyebrow raising.

The current RCP House rankings place 77 campaigns in the typical Likely Democrat, Lean Democrat, Toss-Up, Lean Republican, Likely Republican categories. An issue can be taken with 20 of the ratings, which seem to be mis-categorized based upon available recent polling data, resource imbalance, and historical voter performance.

Looking back at the 2022 RCP ratings just before the election finds that the Toss-Up and Lean Republican categories were particularly inaccurate. Just before people voted in ’22, RCP rated 34 campaigns as Toss-Ups. The end results found that Democrats won 28 of those contests versus only six for Republicans. This suggests that such a lopsided result means the category was somewhat miscast. Otherwise, the end result would have featured a more balanced partisan divide.

The Lean Republican category was a bad miss. According to the RCP ratings, 29 campaigns in the previous election cycle were housed in the column. Yet, Democrats won a majority of those contests, 15-14. This suggests that many of these races should have been labeled Toss-Ups, or even Lean Democrat.

In the current RCP ratings, 32 campaigns are rated as Toss-Ups. There is perhaps a better categorization for eight of the campaigns.

• Rep. Frank Mrvan (D-IN) defeated a stronger 2022 challenger, retired Air Force helicopter pilot Jennifer-Ruth Green (R), by a 53-47 percent margin in one of two Indiana districts drawn to favor a Democrat. Such a history suggests the 2024 contest should be considered Lean Democrat.

• Rep. Angie Craig (D-MN) has won two consecutive highly competitive campaigns with margins well-short of going into political overtime. This, and against a new opponent, suggests that her race should also be moved from Toss-Up to Lean Democrat.

• Though Rep. Don Davis’ (D-NC) CD is more competitive than the district to which he was initially elected in 2022, the Tar Heel State’s 1st Congressional seat is still held by a Democrat. Additionally, with a 40 percent black population, a total minority percentage that exceeds 49 percent, and reliable regional vote history, means this race should reside in the Lean Democrat category as opposed to an outright Toss-Up.

• Though New Hampshire’s 1st District has routinely defeated many past incumbents, Rep. Chris Pappas (D-Manchester) has won three consecutive elections. With such a short general election cycle from the Sept. 10 primary and being clear that Rep. Pappas has repeatedly defied the district’s partisan lean, it is reasonable to believe that he will do so again. This suggests the seat is no longer a Toss-Up seeing Rep. Pappas as his party’s nominee.

• Ohio Rep. Emilia Sykes (D-Akron) has a huge financial advantage over her opponent, GOP former state legislator Kevin Coughlin. The district statistically slightly leans Democrat, and considering this, Rep. Sykes’ incumbency, and her resource advantage provides the necessary evidence for a Lean Democrat rating.

• The FiveThirtyEight data organization rates Texas’ 34th District as D+17. This, plus the fact that Rep. Vicente Gonzalez (D-McAllen) defeated former Rep. Mayra Flores (R) with a substantial 53-44 percent margin in 2022, suggests that this campaign could even be considered Likely Democrat.

• Virginia’s open 10th District is not close to a Toss-Up. Everything here favors the Democrat nominee, state Del. Suhas Subramanyam (D-Fairfax), so consider this race Likely Democrat.

• Washington Rep. Kim Schrier (D-Sammamish) has won three close elections in a district that used to be Republican. The state’s jungle primary has always been a good indicator as to what happens in the general election, and this year’s primary performance was Rep. Schrier’s strongest. While the possibility of an upset exists, the race should now be rated as another Lean Democrat contest.

Several other ratings also seem in need of adjustment. Based upon the post-primary polling, the at-large Alaska race featuring Rep. Mary Peltola (D-Bethel) and Republican Nick Begich III, should move from the Lean Democrat to the Toss-Up category.

Florida Reps. Jared Moskowitz (D-Parkland) and Darren Soto (D-Kissimmee) should move from Lean Democrat to Likely Democrat. The same for the three Las Vegas Democrats, Reps. Dina Titus, Susie Lee, and Steven Hosford. Except for the Hosford race, the Republican candidates in these contests have badly under-performed.

Conversely, based upon polling, the AL-2 open campaign featuring Democrat Shomari Figures and Republican Caroleen Dobson should move from Likely to Lean Democrat. The Rep. Pat Ryan (D-Gardiner) tight Upstate New York seat should go to Lean Democrat from the Likely Dem category.

On the Republican side, Rep. David Valadao’s (R-Hanford) re-match battle in California should move into the Toss-Up category as his D+10 district is always very tight, and former Assemblyman Rudy Salas (D) is a strong opponent.

The two Iowa seats of Reps. Mariannette Miller-Meeks (R-Le Claire) and Zach Nunn (R-Bondurant) should also move into the Toss-Up category based upon the most recent dead heat polls and the former member’s tepid 2024 primary performance.

Finally, Rep. Ryan Zinke (R-MT), who scored an unimpressive 2022 win in the new western Montana seat, should be moved from Likely Republican to Lean Republican also based upon the latest published polling and regional voter history.

Gallup: A Changing Electorate

Gallup Poll Results: To see complete story/data/polling results, go to: Gallup News

By Jim Ellis — Friday, Feb. 23, 2024

Polling

Nationwide Polling: Multiple Demographic Swings — The Gallup research organization released a new nationwide study earlier this month, which is a part of the entity’s Gallup Poll Social Series. The surveys are conducted throughout the year, of at least 1,000 US adults, and they cover 12 different topics annually, meaning a different subject matter each month.

The current release covers where certain segments of the American electorate now stand with reference to their political party preference. One of the key findings is that both parties are gaining strength among some constituencies, while losing it among others.

Gallup finds the Republicans are gaining strength with minority voters, while Democrats are clearly becoming the party of the higher educated.

Among black voters, Republicans have gained almost 20 percentage points just since 2019 when compared to Gallup’s historical surveys. While still having a strong allegiance toward Democrats, blacks now only favor the party by a 47-point span. In 2019, the Democratic margin over the Republicans was 66 points, and even that figure is down from the Democrats’ apex point of 79, which was reached in 2008.

Among Hispanics, Democrats reached their apex in 2016 when they enjoyed a 36-point preference margin over Republicans within this demographic. The current Gallup national survey yields a stark result, finding the Hispanic Democratic advantage today has slipped to only 12 percentage points.

The non-Hispanic white category has also moved considerably toward Republicans, though it wasn’t long ago that the Democratic share of the national electoral vote was at parity with Republicans. In 2007, Democrats had a one-point edge over the GOP within the non-Hispanic white segment. Currently, the pollsters find the spread at 17 points between the two parties, favoring Republicans.

The news isn’t all bad for Democrats, however. Gallup continues to see major shifts among the higher educated voters who are significantly breaking away from the Republicans. Today, the trend shows a 29-point Democratic advantage over Republicans among postgraduate individuals, which is a considerable shift from 2010 when the Democratic edge was 11 percentage points. During that same time frame, the college graduate sector has gone from a nine-point preference for Republicans to what is now a five-point edge for Democrats, which is a swing of 14 points toward the latter party.

In the education category, however, the group that has demonstrated the most radical swing are those not having gone to college. In 2006 through ’08, the Democrats had a consistent 16-point edge. The latest Gallup survey finds almost the opposite result, with Republicans now holding a 14-point advantage within this same category.

Though these particular demographic and personal trait segments are reporting some different partisan allegiance predispositions this does not necessarily mean we will see a commensurate change in voting behavior. Obviously, the candidates matter as does the persuasion communication method developed for the individuals comprising these groups who are apparently more receptive to a different political message today than those from a previous time frame.

This at least partially explains why the vote patterns have not been running parallel to the sentiments detected in this Gallup poll and from other survey entities who are finding similar results.

This type of research does tell us, however, how campaign targeting strategy and communication approach might change for 2024 in comparison to previous elections from the past decade. The party that learns to best manage this changing electorate will be the one enjoying the most success in November.

Can DeSantis Overtake Trump?

By Jim Ellis — Tuesday, July 18, 2023

President

Former President Donald Trump; Florida Gov. Ron
Desantis (R)

Polling: DeSantis vs. Trump — A series of flash polls taken in June and early July within 13 states provides us a glimpse into the path Florida Gov. Ron DeSantis has ahead of him if he is to wrest the Republican presidential nomination away from former President Donald Trump. In summary, the deficit may be too large to overcome.

National polls have for weeks shown Trump holding huge leads over his GOP opponents, but these national surveys are virtually meaningless since the race will be determined in the state-by-state count. The winner must secure 1,234 pledged delegate votes from the 56 voting entities to clinch the party nomination.

The recent polling from the 13 states yields a combined 944 delegates, or 38.3 percent of the total Republican universe. The tallies come from 11 different pollsters during the period beginning June 5 and ending July 6.

Cumulatively, Trump would command a combined 47.4 percent support from these places, slightly less than he is showing nationally. Gov. DeSantis is so far recording 21.2 percent, and no other candidate even reaches five percent support. Therefore, the Florida governor must more than double his support base if he is to surpass Trump’s advantage.

Included in these 13 states are eight places where voters will cast their ballots on Super Tuesday, March 5, or before. The eight states are, in voting order: Iowa, New Hampshire, Nevada, South Carolina, California, North Carolina, Tennessee, and Texas. Of the pre-Super Tuesday states, only Michigan is not included.

In the early states, Trump’s margin, according to the latest available polling data, is actually an even larger 55-22 percent over Gov. DeSantis, with 23 percent choosing one of the seven minor candidates. Therefore, the Trump lead in the critical momentum building early states is larger than in the at-large universe, thus making DeSantis’s task even more difficult.

In addition to the eight early states mentioned above, the remaining tested domains, again in voting order, are Florida, Ohio, Wisconsin, Pennsylvania, and Montana. Of the 13, DeSantis’ strongest state is Wisconsin, where the Marquette University Law School survey, a frequent Badger State pollster, finds the Florida governor trailing Trump by just one percentage point.

The state will the largest chasm between the two men is Tennessee, where the Targoz Market Research survey finds a Trump margin of 61-12 percent. Perhaps most troubling to Gov. DeSantis is his most recent showing in his own home state of Florida. The latest Florida Atlantic University study that concluded on July 1 reveals a 50-30 percent margin in Trump’s favor.

This poll, however, may be an anomaly because no other data has shown such a large disparity. Since the beginning of May, one other research study finds Trump ahead by eight percentage points, while two more see a virtually even split between Trump and DeSantis. Still, for Trump to be doing this well in Gov. DeSantis’ home state is telling as to the strength of the former President’s national advantage.

While serious campaigning is just now getting underway, much can change between the present circumstances and the eventual outcome. With Iowa now moving their precinct caucuses from an original date of Feb. 5 to Jan. 15, the nomination campaign season grows even shorter.

For Gov. DeSantis, or any of the minor candidates to make a serious run at former President Trump, the activity pace will have to significantly quicken, and their momentum will need an abrupt about face.

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 →