Most Americans view election surveys with both hope and skepticism.
The large number of votes and the potential for incomplete or biased votes make poll monitoring a difficult and time-consuming task.
I wish there was a place where I could see all the polls and the unbiased averages that show trends.
I found out something.
I visit RealClearPolitics every morning looking for articles on both sides of the day's relevant issues.
When this website started publishing and updating a detailed list of election polls and their unweighted averages, I thought this was the answer to a political geek's prayers.
Apparently I wasn't the only one. RCP's polling averages became widely accepted throughout the media world.
But there was a problem. The ruling polling elite objected to the unfiltered results of the RCP.
Consider this: New York Times piece “Why the Right Thinks Trump is Running Away from the Race,” published five days before the presidential election. A righteous person naturally thought so because he was. But despite the political winds, Times staff could not see that possibility.
The newspaper said it was a misconception that Donald Trump was on track to victory and that “distorted polls and anonymous betting markets are boosting Republicans' hopes.”
The Times characterized it as something of a conspiracy aimed at creating a “narrative of unstoppable momentum” against President Trump and said it “could undermine confidence in the entire nation.” [electoral] system. “
In an effort to save our democracy, and perhaps encourage its progressive readership, the Times said that biased polling “does not treat all polls equal, which is why news organizations, including the New York Times, It has not had a significant impact on the public opinion poll average calculated by the government.” ”
Rather, traditional news organizations, in their greater wisdom, either “adjust their models to down-weight” or simply ignore investigations they deem biased.
The Times seriously thought people would see this as a plus.
The paper also criticized RCP for treating “all polls equally” and “unlike our competitors” and including polls “that other aggregators have rejected.”
Another objection is that RCP only reports results and “doesn't evaluate averages.”
The Times sees this as a negative. Really.
Of course, the paper's point is to exclude or downplay polls that it believes are flawed to give a more accurate reading of voter sentiment and election results.
Now, let's compare how well the Times and RCP's averages were maintained after the votes were counted.
President-elect Trump won the popular vote by about 1.7 percentage points (49.9% to 48.2%).
The final episode of the Times voting average was 2.7 percentage points behind that result, with Vice President Kamala Harris leading by 1 percentage point (49% to 48%). final match RCP average decreased by 1.8 points and Harris increased by 0.1 points (48.6% vs. 48.7%).
Therefore, the weighted average of all RCP polls was about 33% more accurate than the weighted average of some Times polls. Now, who would have thought of that?
Unsurprisingly, the Times' average was less accurate, in part because it underestimated or excluded polls it didn't like, where Trump was leading.
Trump fell one point short of the TIPP poll in which he won the popular vote. The Rasmussen poll, which had Trump ahead by 3 points, was excluded, leaving it 1.3 points off Trump's plus, but still a little more than 50% closer to the actual result than the Times average. RCP included both polls without weighting.
The Times also criticized RCP for its “no-toss-up” Electoral College map, which showed potential election outcomes assuming RCP averages for each state.
Horrifyingly, the Times found that the map showed “Trump wins in every battleground state” except for two. Of course, we all won in the end.
Finally, the Times reprimanded Republicans, noting that the betting odds were increasingly in Trump's favor. These odds are considered an indicator of voter sentiment because they reflect people funding candidates rather than simply answering questions over the phone or online. RCP provides readers with a list of odds for each betting site and averages those odds.
The paper was concerned that an unidentified Frenchman had skewed the odds with huge bets, and that “Mr. Trump's approval ratings had skyrocketed over the past month,” and that “reliable polling companies “It is not consistent with the overall situation of the current election campaign.” At least that was an accurate statement. However, the final results were tracked.
Fortunately for the Frenchman, he didn't rely on the Times or its so-called “reputable pollsters.” He “successfully bet that Donald Trump will win the presidency,” the Wall Street Journal reported, “earning about $85 million in profits.” report.
Bottom line: RCP's polling average is less biased and more accurate than the Times' average, RCP's “no toss-up” map showing Trump winning nearly every battleground state is mostly accurate, and betting odds was favored as the winning candidate. I believe the Times' criticism is inaccurate and self-serving, and is due to a failure to see outside the left's progressive bubble.
The Times poll's average results remind me of a misquote from New Yorker film critic Pauline Kael more than 50 years ago: I don't know anyone who voted for him. ” One might wonder if anyone on the Times staff knows anyone who voted for Trump.
In any case, we know where they'll be looking for polling data next cycle, and it certainly won't be the Times.
Andy Puzder, a distinguished fellow at the Heritage Foundation, served as CEO of CKE Restaurants for 16 years.
Twitter: @AndyPuzder
