The Massie/Gallrein race and the youth vote
Yesterday I heard a discussion of the Massie defeat in Kentucky’s 4th, and I was alarmed to hear that the only age group Massie didn’t win was those over 65, who turned out in great numbers. I also heard that the young voted overwhelmingly for Massie to the tune of +30.
Usually numbers about age groups rest on exit polling. But today, when I tried to document the age-group results, I couldn’t find any exit polls at all. I did find this Daily Caller post on Facebook entitled, “Ed Gallrein Has Boomers To Thank For Defeating Thomas Massie.”
It began like this:
The Baby Boomer generation overwhelmingly voted for Trump-backed candidate Ed Gallrein, who ousted seven-term incumbent Republican Kentucky Rep. Thomas Massie in a GOP primary Tuesday night.
Seventy-two percent of voters between the ages of 26 and 35 supported Massie, while 16.2% backed Gallrein, according to a Quantus Insight poll from May 13. A majority of voters ages 66 to 75, 61.8%, supported Gallrein and 27.1% of that age group supported Massie.
What’s wrong with this picture? The poll’s date was May 13, not an exit poll at all and perhaps not reflecting the actual vote. The piece went on to state:
KY-04 GOP Primary: Results by Age Group
Age 17-25: Massie +25
Age 26-35: Massie +56
Age 36-45: Massie +38
Age 46-55: Massie +17
——
Age 56-65: Gallrein +18
Age 66-75: Gallrein +35
Age 76+: Gallrein +33@QuantusInsights | 5/11-12 | 908 LV https://t.co/H7juBRgxXr pic.twitter.com/pjcRguiF4a
— InteractivePolls (@IAPolls2022) May 13, 2026
Those younger age groups look disturbing, considering that it could reflect an anti-Israel antisemitic turn among young voters on the right. But when you look more carefully at the samples, the youngest groups are small – in some cases extremely small, so small that it makes the results meaningless or nearly so.
For example, there were 4 people in the 17-25 age group., In the 26-35 age group, the sample was bigger but still quite small: 67 people. In the 36-45 age group there were 125 people and in the 46-55 group there were 136. Those numbers over 100 increase the validity but still have a fairly large margin of error (couldn’t find that figure, though). And in the 46-55 group there were also 12.5% undecided, which could have nearly evened the vote if all of the undecideds went for Gallrein. In addition, in general, “undecideds” in the poll leaned toward Gallrein.
In the older age groups of 56-65 and 66-75 the samples were decent-sized: in the 240s. And those groups went strongly for Gallrein.
So it’s difficult to tell what it all means. Not only are the numbers flawed but they don’t say what actually happened in the voting booths (I assume there are still voting booths in Kentucky?). You might ask why I pay attention to polls, but my answer is that they’re pretty much all we have to go on in trying to understand election results, despite their many flaws.
Massie himself is claiming that despite his win, he’s the wave of the future:
The Kentucky congressman claimed in his speech after his defeat that young voters were still on his side.
“People that want somebody that will go along to get along, I’ve never heard of that strategy but that seems to be what the voters want,” Massie said. “But not the young voters.”
The crowd was still energetic despite Massie’s loss, and started a chant of “No more wars!” that the congressman joined in on. Massie’s speech meandered through different topics and touched on other politicians before another chant started of “America First!”
I think Massie’s future will involve a lot of talk show and podcast appearances, bragging about his success with youth. Maybe he’ll even score an appearance with Tucker Carlson, so they can rant about Jewish influence and the new wink/wink term for Jews, the “Epstein class.”
The good news is that young people become older people, and overall the population is getting older (the latter is not a good thing generally, but regarding voting patterns it helps).
Any one who prefers Hamas to Israel has been brainwashed or is unable to process reality. I listened to Megyn Kelly spout some preposterous nonsense about Israel’s influence to the Real Clear Politics team yesterday, and I am glad to say they pushed back and called her out for antisemitism.
The young (20-54) have been making noises about the political fighting, if they vote it will be democratic.
Democrats Propaganda Ministry seems to be losing personal so not looking like a good prospect there
You might ask why I pay attention to polls, but my answer is that they’re pretty much all we have to go on in trying to understand election results, despite their many flaws.
If polls were really just flawed, incomplete information, I would agree with you. But the free-to-us polls published in the media are cherry-picked to support a narrative, and I think paying attention to them gets us further away from understanding what is going on.
It’s a bit like expecting the bent-corner card at three-card monte to give you information about where the queen is, maybe “flawed” information, but it’s the best you have to go on. But the reality is that the game is rigged and they are trying to entice you to play by giving you the illusion of information.
Not to pick on the Daily Caller, but they were the ones who chose to use old polls with skewed demographics and try to pass them off as explaining the election. No one made them do that.
Niketas:
And yet many polls turn out, ex post facto, to have been correct. For example, this one correctly predicted the outcome within the margin of error. It has Gallrein winning by 5%, with 7.5% undecided but the majority of the undecided leaning to Gallrein. The margin of error was 3.3%.
The actual results had Gallrein winning by a little less than 10 points. So if 4 or 5 percent of the undecideds went to Gallrein, you have the actual result figure.
“People that want somebody that will go along to get along, I’ve never heard of that strategy but that seems to be what the voters want,” Massie said.”
That ‘strategy’ is the foundational principle of politics in a republic, where representatives are chosen through the democratic process.
Which leads to Massie’s assertion of “never (having) heard of that strategy”.
Massie has yet to learn that to ‘get along’ with other people, it’s best to be honest but only as honest as common sense and decency allow.
Clearly, Massie is a smart man. Sadly not a wise man.
Massie got to believing his own maverick pr. He forgot politics is the art of the possible . Politics is horse trading , it’s a dirty business. He should have resigned before his defeat. So much for being noble.
Never ever underestimate the electorate.
According to RealClear politics, three polls days before the election showed Gallrein winning by 7-8 points. The Big Data poll done by Rich Barris had the race as a dead heat. But Barris was clearly supporting Massie and I think his bias changed his turnout model.
Robert Barnes embarrassed himself by confidently predicting an easy Massie victory. He even advised people to bet heavily on Massie winning just a week before the election. Barnes regularly puts out odds on political elections and his advise to bet on Massie was won of his strongest recommendations. Like Baris he let his emotions influence his analysis.
(Neo, is the word “win” an error here: “Massie himself is claiming that despite his win…”)
The law of large numbers doesn’t apply to such small numbers.
But Barris was clearly supporting Massie
Rich Barris has become the Thomas Massie of pollsters.
@neo:And yet many polls turn out, ex post facto, to have been correct.
That’s not itself a basis for believing in them, however–to do so would be more cherry-picking, the “Texas Sharpshooter”. After all, half of people who guess at a coin flip get it right. That doesn’t mean they had a viable method of predicting the coin flips they happened to get right.
In addition, people expect a lot more from the polls than just do they predict vote percentages within the margin of error, as we saw with the Daily Caller article you criticized. And that’s the problem. The Caller had the narrative they wanted, and they even mixed and matched data from different polls to get what they wanted to support it, and you properly noted that the data was not anything like solid enough for them to draw those conclusions.
Do some polls get things right? Yes, we can see that. Do we know that they have a valid method that is expected to be more right than wrong, as opposed to sometimes getting lucky? Much harder to know. But with all the polls to choose from, the ones that are going to be brought to our attention are the ones that support the pre-chosen narrative, and that’s a totally different issue from whether some polls should be relied on.
If the poll says what the journalist wants it to say is the determining factor as to whether it’s brought to our attention, and that’s why I’m comparing it to three-card monte. (After all some marks do guess where the queen is. Three-card monte operators have a procedure for handling that.) If a poll is also particularly reliable that will be made part of the story, if it already says what they want.
A spectacular example was the late 2024 Des Moines Register poll by Ann Selzer predicting a Harris ahead of Trump. It told the media what they wanted to hear, so they promoted it as well as her record for reliability, which was based on cherry-picking her past successes (of 68 forecasts, 41% of them were outside the margin of error; not ridiculously bad but not indicative of a gold-standard either). Plenty of people at the time said the result was not believable. Nate Silver said it was very likely to be wrong. None of that mattered. This was not an unusual case except for the consequences for Selzer; in most circumstances Gell-Mann amnesia sets, pollsters continue to make predictions and journalists continue to cherry-pick them.
The key issue is separating polls of all asked, into polls of actual voters. And since we get actual only Afterwards, the closest is “likely voters”.
I guess lots of young folk, both pro- & anti- Trump, are more willing to answer a poll than to actually vote.
There is some number of voters more likely to vote, for their side, if they think their side will win. Less likely if uncertain, much less likely if they think they’ll be voting for a loser. Sadly, more of these are less politically obsessed conservatives, with a real life, who can be called Winning Voters.
The Dem media wants the Reps to be the losers and looks for polls that show the result they want, so as induce more Rep Winning Voters to not vote.
There is some very good news is that more young men are becoming Reps, and it’s likely this is attracting more of the hot young women who are more interested in Mrs, degrees, and even dates with straight guys. The Party that has more hot young women is more likely to grow among the young likely voters.
The question really is: Will.young voters show up?