Cleaning up my email this week I found this from November 2024, right after Donald Trump recaptured the White House, that I forgot to post. Life happens sometimes. There’s a certain timeless relevance to the message, so I am going with it today, with updates in italics.
One of the regulars here asked about finding an impartial news source, given how wrong the mainstream news media was in predicting the outcome of the US presidential election. Except, they didn’t get it wrong – they just chose not to share the whole story. (I’m not sure there is such a thing as a completely impartial news source. There are some more trustworthy than others that work to overcome natural bias.)
The pollsters, who said the race was a dead heat, believed their numbers were accurate but the interpretation was misleading. As the cliche goes, there are “lies, damn lies and statistics.” To me the intangibles in the data indicated a Trump win
There isn’t much the average consumer of news can do about such things, except to become more media literate than most if us want to take the time for. It isn’t “fake news” that gets us, it is a system that has automatic biases built in.
A century ago, maybe a bit longer, news media used to make no pretence of impartiality. People subscribed to newspapers based on their political slant. That began to change in the early twentieth century, and by mid century news coverage was supposed to be fair and impartial. Reporters were told to put their personal leanings aside and report accurately and honestly.
When I was in Journalism school in the late 1980s, some of my classmates rejected that doctrine. They wanted to be influencers, not reporters. From what I have seen, the media has shifted in that direction ever since.
Which means you need to find news outlets that are accurate in their reporting, then strip out the ideological slant. Numbers are numbers – how you spin them makes a difference. And that’s why Harris and Trump were seen as virtually tied before the polls opened. The media, for the most part, were appalled at the thought of a Trump victory, so they subconsciously interpreted the polling data to make Harris look better.
Even as the numbers rolled in, they clung to their preconceptions. I could see the whole thing was over by nine p.m. on election night, but the networks were still saying it was too close to call.
Statistically they were correct. Theoretically Harris could still have won, if she got all the not yet counted ballots. But that wasn’t going to happen.
Polling is science, but interpreting the numbers is an art. Canadian Prime Minister Justin Trudeau knows that.
According to the polls, he should be packing his bags and riding off into the sunset rather than suffer the most embarrassing loss in Canadian electoral history. (And eventually the pressure from within his party was such that he did leave.)
However, he says he isn’t going anywhere. He knows the published polling numbers don’t tell the whole story. (He was right. I never expected to say that about Justin Trudeau.)
Yes Trudeau is behind among decided voters, so far behind that it is unlikely he can do much to improve by the scheduled October 2025 election date.
The real question though is how many people haven’t decided yet which way they are going to vote. And how many are open to changing their mind.
If that is ten per cent of the electorate, then Trudeau is about to become a footnote in history. If it is forty per cent, he may have a shot at defying the odds and winning. A lot can change in a year. (Mark Carney, Trudeau’s successor as Liberal leader, won an early election. His advisors read the mood of the country correctly. His popularity continues to rise, which is why I think anther election soon is quite possible, depending on the US midterm results.)
Just ask Donald Trump. According to the polls, he should never have become president. (Trump is possibly the most unpopular president in American history. That may not make much difference in the 2026 midterm elections. I’ll talk more about that before the vote.)
