Why Is So Much Happening, When The Polls Are So Stable?

An uncharacteristically insightful piece at Hot Air this morning leads in passing to an intriguing question:

First, Biden didn't really crater all that much from the debate. On June 27, the day of the debate, Biden's RCP average was 45.1, and it bottomed out on July 3 at 43.6. By the time Biden withdrew, though, he'd recovered most of that lost ground to have an RCP average of 44.8. Trump went from 46.6 on June 27 to 47.9 on July 21, but much of that gain came after the assassination attempt and then the GOP convention.

Second: Harris only negligibly improves on Biden, even in the honeymoon week. On June 27, before the debate, Biden was 1.3 points behind Trump in the RCP average. Trump had a 3.1-point lead on Sunday, when Biden withdrew. Now, at the end of what should be the friendliest news cycle Harris will ever get, Trump still leads by 1.7 points -- slightly better than his lead when the debate aired.

The writer thinks this is about Kamala, but I think the bigger question is the stability of the polls, not just since the debate, but since last fall. The fact is that statistical chatter has put Trump ahead in the meaningless Real Clear Politics national "average" between 1 and 3 points for the better part of a year. The June 27 debate didn't change the polls, and Joe dropping out of the race in favor of Kamala didn't change them, either.

Nevertheless, the story, at least for now, seems to be that the Lizard People decided that the June 27 debate was the factor that said it was time for Joe to go -- when, as I pointed out yestesrday, it was time for Joe to go at least a year ago, and the Lizard People missed it. Nevetheless, if the debate was what drove their decision over the past month, why did they discount the polls?

After all, right up to last Saturday night in Rehoboth when Joe is said to have met with his handlers who finally told told him it was over, Joe was insisting that the polls were stable in a margin-of-error race, and he was in fact correct. One answer might be that the RCP national "average" is in fact meaningless, but Trump was ahead in enough of the battleground states to put the race out of reach -- but even there, even if Joe consistently trailed Trump in each of them, it was often by margin-of-error amounts.

And in fact, respectable opinion had been stable since late last year that the race was a tossup. What put the Lizard People into such a panic after the debate? I think there are several explanations. One is almost certainly that the Lizard People don't take the polls seriously, especially the RCP "average". The polls underestimated Trump in the 2016 and 2020 elections by around 3 points, so a 1-point margin for Trump in the polls is actually 4 points on election day.

The second issue isn't the movement in the national "average", but the increase in the actual number of battleground states. This is a trend that had begun before the debate. The week before, RCP expanded its list of of battlegrounds to include Virginia, Minnesota, and New Hampshire, and by the end of last week, Trump was leading in Virginia and behind by roughly 3 points in the other two, which allowing for poll bias against Trump was worrisome.

But now we have a puzzle: even though the polls have been stable since Kamala replaced Joe, and we night expect them to continue his way in the new battlegrounds as well, RCP has removed the new battlegrounds from its list and reverted to the old standbys of Wisconsin, Michigan, Pennsylvania, Arizona, and Georgia. I noted on June 16,

It seems to me that the "close election" paradigm hinging on the "battleground" states of 2016 and 2020 is beginning to fail. As of right now, it seems likely that Joe can at best carry only two or three of those "battlegrounds", which makes his chances of a narrow Electoral College victory slimmer and slimmer. But with polling data coming in that Minnesota and Virginia are already in play, this puts a potential wider Trump Electoral College victory within the realm of possibility, which renders the whole "battleground" model inoperative.

But this was never reflected in the RCP "average", which was still unchanged even when the Lizard People realized it was time to dump Joe. So there must be another factor, which I think is the fallacy behind the Hot Air piece at the top link. A bigger factor driving the Lizard People's decision, I think, was the failed assassination attempt in Butler on July 13. Back in June, before Butler, I mused on the subliminatl effect of two other unsuccessful assassinations:

Many years ago, I read someone suggesting that Gerald Ford lost the 1976 election simply because he was the subject of failed assassination attempts by two women ( Lynette "Squeaky" Fromme in Sacramento September 5, 1975; and Sara Jane Moore in San Francisco on September 22, 1975).

Of all the other possible reasons for Ford's loss -- especially his pardon of Nixon -- this one immediately struck me as most persuasive. On one hand, that the attempts would be unsuccessful goes to the stereotype of the incompetent woman. On the other, that bumbling women would target Ford canceled his standing as an alpha male.

So the attempts against Ford somehow reinforced a subliminal context that Ford himself was bumbling and feckless. The attempt against Trump, on the other hand, was deadly, well-planned, and certain, enabled by the weaponized incompetence of a deep state agency. That this wasn't just a bumbling attempt like those against Ford was demonstrated by the death and wounding of bystanders. A credible explanation for Trump's near-brush with a Lee Harvey Oswald-style assassination was the intervention of the Almighty, not the incompetence of the assassin.

The powerful subliminal effect of this image was something Trump immediately and instinctively grasped after the shooting when he stood, pumped his fist, and uttered exclamations that might have been "Fight!" or maybe another F-word. This, I think, was the image the Lizard People grasped when, a week later, they finally impressed on Joe that it was time to go. It had nothing to do with the polls, or even the debate. It had everything to do with Trump's astonishing political instincts, which he'd been demonstrating repeatedly since last fall.

I'll have more to say about the subliminal aspect of the campaign.