The Colossus
Francisco de Goya
As the giants strolled by, Francisco de Goya frantically painted them, but he painted so quickly the friction of his brushes caused almost all the paintings to catch fire and burn up.
This is why oil painting is dangerous.
The Polling Frightens Me, Help!
Yes, depending on who you are and, inversely, how little you know about poll accuracy and trustability, polls can be frightening. A writer using the dubious handle (I’ve been in social media since the early ’80s, so’s I gets to say “dubious handle”) gingytheelephantboy has an article out on Daily Kos that, again, is either reassuring or even more terrifying. Important points:
A special election was held to fill George Santos’ seat. The polls showed a toss-up. The Democrat won by 8 points. The last special election of the season was held in New Jersey last month. The seat was safely Blue and turnout was very small. The Democrat took 80 percent as compared to 75 the last time around. Yesterday in Fairbanks, a +14 Trump city, the Democrat won the mayoral election by 15 percent [sic]. Are we seeing a pattern here?
So why do I continue this series? I’m trying, as a non-specialist, to convey how to evaluate a campaign in terms of anticipating winners and losers during a campaign. I dare to say that my qualifications are the qualifications of most of the electorate, except I’m now (sigh) 60+ years old, and having 40 years in social media suggests I’ve had more experience watching and participating in arguments about politics and how the world works than most. I’ve seen a lot of weirdness, from the current absurdly blasphemous surrealism of Christian Nationalism to the assertion that America was about to experience a revolution because, at the time, the homeless in Denver were upset.
But to get back to the point, I try to share my thinking and evaluation methods so readers can get a leg up on not being 60+ years. I report polls, but I do not necessarily buy into them. Pollsters will tell you data collection has become more and more difficult. Progressives claim the youth vote, leaning Democratic, is undercounted. I’ve nearly been incessant in insisting the Dobbs decision is the most important factor in this election, and I do not expect to see that fade until the existential edge of reproductive health is once again dealt with in an adult manner – and not religious zealots screaming and running in circles.
Here’s the money quote from gingytheelephantboy:
Here is what I see. Democrats and pro-choice have been outperforming the polls by 5% or more in election after election for two years. And that is where we are today. By the way this doesn’t bother me a bit. If it motivates our voters to turn out like their lives depend upon it, then it is all good.
So the big question is why is this happening. My answer is that the polls are missing something important and it doesn’t take a rocket scientist to see what it is. They are just not polling enough women, particularly young women, a group notorious for not voting. And they are the very people who have been registering in droves. Of course they are not being polled, they don’t fit the old models.
Keep the above points in mind when evaluating campaigns, stay a bit skeptical of even top rank pollsters – and when choosing for whom to vote. You weren’t going to skip for despair, were you?
What’s this The Washington Post poll, And Why Is It Unrated?
Looking at one of the polls, WaPo doesn’t call out someone else doing, or sharing, the work, and all of FiveThirtyEight’s ratings of WaPo involved its partnering with someone else. Therefore, in the absence of a partner, they are unrated. Fortunately, they only appear once.
Biff! Pop! Pow! Aw, Adam West Escaped — Again!
- CNN/Politics has an article on Pennsylvania Senate challenger David McCormick’s (R-PA) former life as a hedge-fund manager. The article’s headline?
Senate candidate Dave McCormick led hedge fund that bet against some of Pennsylvania’s most iconic companies
The first couple of paragraphs are all that is necessary.
In the years that Pennsylvania Senate candidate Dave McCormick led one of the world’s largest hedge funds, the firm bet millions of dollars against some of the state’s biggest and most iconic companies, financial filings show.
Under McCormick’s leadership, Bridgewater Associates shorted the stocks of nearly 50 companies headquartered in Pennsylvania, including The Hershey Company and US Steel, a CNN review of records from the US Department of Labor found.
I don’t think most voters will care. Shorting is not an option generally known to the non-investing public, but it’s not an exotic or illicit investment strategy, no matter how much some amateur investors hop up and down in outrage. I have never used it, as gains are limited to roughly 100%, while potential for loss is unlimited; it’s a tool of the confident professionals. Who sometimes go broke using it, just like us long investors.
This strikes me as a scare article.
On the other hand, this article on a ham-handed visit by Mr McCormick to Philadelphia is not in the least surprising.
In other, more numerical news, highly respected Quinnipiac University (2.8) gives Senator Casey (R-PA) a 51%-43% lead over McCormick, a thoroughly reasonable lead, with a margin of error of ±2.6 points. Emerson College (2.9) gives the Senator a 48%-46% lead, which is within Emerson College’s ±3 point credibility interval. As with many Emerson College polls, it seems sometimes more rightward-leaning than many pollsters are measuring. TIPP Insights (1.8), working for Republican-aligned American Greatness, gives the Senator a 47%-43% lead with likely voters and a 48%-40% lead with registered voters. The big gap between likely and registered voters seems unlikely, but I’m not sure what it implies.
- In Maryland the University of Maryland Baltimore County Institute of Politics, an unknown pollster, gives Democrat Angela Alsobrooks (D-MD) a 48%-39% lead over former Governor Hogan (R-MD). This is in the neighborhood of a poll from a reputable pollster in the last report. Additionally, I know Maryland is a bastion of the Democratic Party, and the weight of an unknown pollster is difficult to measure, but this observation reinforces a point I made above:
Maryland voters are poised to enshrine the right to abortion access in the state constitution, with 69% of voters saying they will support it and just 21% opposed, according to a poll released Wednesday.
It’s certainly possible that abortion amendments on the ballot of conservative States may sway many voters to favor not only the amendment, but the liberal minority who put it on the ballot.
- Highly respected Quinnipiac University (2.8) has somehow found the Michigan Senate race to be even at 48%. This is certainly mysterious, as other respected pollsters have had Rep Slotkin (D-MI) up by ten or more points. Every pollster can have a blunder, I suppose. For comparison, Emerson College (2.9) has Rep Slotkin leading former Rep Mike Rogers (R-MI) 49%-44%. Emerson College has something called a ... credibility interval, similar to a poll’s margin of error, of ±3.1 points, which I take to mean Slotkin’s lead could be two points, or eight points. Then again, InsiderAdvantage (2.0) has more of a QU result of Rep Slotkin leading only 46%-45% – a statistical dead heat. However, InsiderAdvantage seems to lean to the conservatives – see that link for some right-wing speculation.
- Quinnipiac University (2.8) has Wisconsin’s Senator Baldwin (D-WI) up by four points, 50%-46%, over challenger Eric Hovde (R-WI), which feels a bit small. Still, the margin of error is ±3.0 points. Emerson College (2.9) has the Senator’s lead also at 50%-46%, with a credibility interval of ±3.0 points. Coincidence? Two converging pollsters? But InsiderAdvantage (2.0) has Baldwin’s advantage at only 48%-47%, or a statistical dead heat. But if InsiderAdvantage is right-leaning?
- In Florida we see discord between two prominent pollsters. The New York Times/Siena College (3.0) sees the Florida race as solidly Senator Scott’s (R-FL) at 49%-40%. Marist College (2.9) also sees the lead as the Senator’s, but only two points at 50%-48%., which is probably within the margin of error, but that is not listed. The former has had some other questionable results, while the latter’s result is more congruent with other recent results.The Marist College poll also notes:
37% have a positive impression of Mucarsel-Powell. 23% have a negative view of her. A notable 40% have either never heard of her or are unsure how to rate Mucarsel-Powell.
Much like Democratic challenger Rep Allred (D-TX) in Texas, Mucarsel-Powell making herself known to Florida voters may reap big benefits.
- Texas, like Florida, has been polled by The New York Times/Siena College (3.0) and Marist College (2.9), but in this case their results appear to be close, as the former gives Senator Cruz (R-TX) a 48%-44% lead over Rep Allred (D-TX), while the latter gives the Senator a 51%-46% lead. Allred’s chronic problem/opportunity continues, according to Marist:
39% have a favorable opinion of Allred. 36% have an unfavorable impression of him, and 25% have either never heard of Allred or are unsure how to rate him.
Allred must reach those 25% who don’t know he exists.
- Ohio gets its own paired of polls, but this time Marist College (2.9) is paired with WaPo Poll (unknown), and they both believe the race is within the margin of error, whatever that might be, with the former having Senator Brown (D-OH) leading challenger Bernie Moreno (R-OH), 50%-48%, while the latter has the Senator leading by 48%-47%.
- While I didn’t plan to mention Arizona again, Emerson College (2.9) is giving Rep Gallego (D-AZ) a small seven point lead over election denier Kari Lake (R-AZ), 50%-43%. The pollster comments,
Since last month, Gallego’s support increased two points while Lake’s support held at 43%.
So perhaps momentum is with Gallego. David Weigel has an interesting differential analysis of Mr. Gallego and Vice President Harris in Semafor here.
- In the shock of the day, The New York Times/Siena College (3.0) sees Montana Republican challenger Tim Sheehy (R-MT) leading Senator Tester (D-MT), 52%-44%. Public Opinion Strategies (1.6), sponsored by the Montana Republican Party, gives Sheehy the lead as well, 51%-45%, but the combination of a weak pollster and a partisan sponsor casts a pall over the credibility of that result.
- Envision two Vikings, whacking away at each other with foam light sabers: Nebraska Senator Fischer (R-NE) and challenger Dan Osborn (I-NE) each hired a pollster, ran a poll, and are advertising their results, thus the whacking reference. Senator Fischer’s pick is hardly that of the litter, as for reasons unknown and mysterious she selected unknown pollster Torchlight Strategies, a repeated action on her campaign’s part, and they’ve obligingly found her having the lead over Mr Osborn, 48%-42%.
Mr Osborn’s pollster is Change Research, which at least has a rating, but it’s a deeply unimpressive 1.4. Nor do they do much published work. They are giving Mr Osborn the lead, 46%-43%. Of course.
So how to evaluate this? It’s almost impossible to say, except maybe Don’t evaluate it. I’m going to recognize this as dueling propaganda. Both sides are trying to convince their partisans that the race is still winnable, so come on out and vote. The implied message is entirely honorable.
But if I were Nebraskan, I might be irritable. I dislike manipulation, and that’s the essence of this, to my mind.
- In case you wonder about Virginia, Republican challenger Hung Cao (R-VA), who seems to be a Democratic plant, is profiled here by Marc Fisher. If Cao is not a plant, then his approach to winning the seat is puzzling. Does he think a revolution will happen and put him in the seat? That a Trump victory is imminent and he’ll ride the coattails? Or is this a more devious strategy of gaining an objective by losing the race? I am puzzled.
Roll The Credits!
Nyah, too tired. Crank broke. Wife needs attention. Cat needs food.
Have a good weekend, folks.