Vice President Kamala Harris’s chance of winning the U.S. presidential election is lower than it was two weeks ago, according to a model that uses the stock market’s year-to-date performance to predict the incumbent political party’s likelihood of victory.

Nevertheless, that model still predicts that Harris, the Democratic presidential candidate, is likely to win the presidency on Tuesday — giving her a 70% probability of victory. The reason Harris’s likelihood of winning was lower on Nov. 1 than the 72% where it stood on Oct. 17, when I last wrote about this model, is that the Dow Jones Industrial Average DJIA has declined in the interim.

Most Read from MarketWatch

• ‘I don’t want to stop her from moving on with her life’: My second wife is younger than me. If I die first, how do I make sure she doesn’t cut off my children?

• ‘My mom still has his original will’: A few months before he died, my father went online and made a secret will, cutting off my mother. Can he do this?

• ‘No one writes about “rich guy” early retirement’: My wife and I earn $300,000 a year and have $3.75 million. We have 3 kids. When can I quit working?

• Here are the stakes for stocks, bonds, gold and other markets as investors await too-close-to-call election

• ‘I retired earlier than planned’: I’m 69 and will have $6,000 a month in retirement income. The bulk of my $3.6 million is equities. Is that OK?

Need to Know: The U.S. election need not be scary. History shows this is the way to trade it, says Citi.

This stock-market prediction model is not complicated. It exploits the historical tendency for the incumbent political party’s chance of electoral victory to reflect the Dow’s year-to-date performance. The model’s track record is statistically highly significant — at the 99% level.

Take a look at the chart above. It plots the trendline that best fits the historical data back to 1900. Is the model foolproof? Of course not. And bear in mind that a 70% probability is not 100%. Furthermore, even if the model had a perfect track record, there’s no guarantee that the future will be like the past.

That said, my simple model does have a better track record than the majority of models Wall Street uses, many if not most of which have no statistical validity. The model makes theoretical sense: The stock market is forward-looking, so a rising market means that most investors are upbeat about the economy’s prospects in coming months. Numerous studies have found that people tend to vote their pocketbooks.

I got a lot of angry emails in response to my mid-October column, with many of you arguing that a Harris presidency would be disastrous for the economy. I myself have no idea. But I do know that, if those dire forecasts were correct, we would expect the stock market to plunge whenever Harris’s chances of winning go up. That hasn’t been the case, as I pointed out in a column earlier this week. The stock market on average has risen in the weeks since July in which the Harris contract at PredictIt.org rose.