Trump Vs. Clinton Presidential Polls: Trump Pulls Ahead In Race — Or Does He? November Election Picture Still Cloudy

Two new presidential election polls show Donald Trump actually leading Hillary Clinton nationwide among voters, who responded to the question of which candidate they would pick if the November vote was held today. One poll showed Trump ahead by a five-point margin which, if accurate, would present Clinton with a steep challenge to defeat the New York real estate mogul and become the first woman elected president of the United States.

So, should Clinton be worried about these new polls? Should Trump be excited? A closer look at the polls shows the answer is most probably “none of the above.”

A Fox News poll released earlier this week showed Trump with a three-point lead over Clinton, but the poll came after Clinton had led in all but two — including one poll that was tied — of the previous 25 major national polls pitting Trump vs. Clinton listed by the poll aggregating site Real Clear Politics.

Nonetheless, the Fox News poll received broad, almost saturation media coverage this week. In the video below, the Morning Joe panel on MSNBC attempts to analyze the single Fox News poll.

At the same time, independent election forecasters and political scientists who looked at the Fox News poll spotted some significant problems — including the most obvious one, namely that the poll appears to be a statistical “outlier.” In other words, unless the results of the Fox News poll are repeated in a significant number of upcoming polls, the poll is essentially meaningless.

In fact, on Thursday a CBS News/New York Times poll showed Clinton with a six-point national lead over Trump, and the latest Reuters/Ipsos poll found Clinton with a five-point lead, adding further evidence that the Fox News poll was an outlier.

In addition, as Emory University political scientist Alan Abramowitz noted, the Fox News poll contained apparent errors in sampling — that is, the pollsters surveyed too many members of one particular group. In this case, Republicans.

Also, as Abramowitz noted in his tweet, the Fox News poll showed 23 percent of Latino voters saying they would cast their ballots for Donald Trump. That result seems questionable at best, given that a recent poll of Latino Americans found a whopping 87 percent of them holding an “unfavorable” view of Trump.

However, the results of the Fox News poll were, in fact, repeated in a poll released the next day. In fact, the Rasmussen Reports poll taken on May 17 and May 18 showed Trump ahead by not just three but five percentage points.

Was there a problem with the Rasmussen poll? Polling experts have found that Rasmussen polls, in general, tend to show a bias toward Republican candidates. In the 2012 election, according to a study by FiveThirtyEight founder Nate Silver, Rasmussen had the fourth-highest average error rate among 23 polling firms in the study.

Additionally, Silver found, Rasmussen had also had the second-highest bias toward the Republican in that presidential election, Mitt Romney — overestimating Romney’s performance in the election by an average of 4.5 percentage points per poll.

This year, of the previous 25 Trump vs. Clinton polls prior to this week, Rasmussen was responsible for the only two polls that failed to show Clinton in the lead, as an April 26 poll showed the race tied, and an April 28 Rasmussen survey gave Trump a two-point advantage.

A third Trump vs. Clinton poll came out this week as well. McLaughlin and Associates, operated by self-described Republican strategists John and Jim McLaughlin, released a national poll on May 17 showing Hillary Clinton leading Donald Trump by four points, 46-42.

The Fox News poll had Trump ahead 45-42, while Rasmussen showed a 42-37 race in favor of Trump.

Polling experts caution against attaching too much significance to any individual poll — regardless of how much news coverage it gets. The recent Fox News poll showing Trump apparently overtaking Clinton received blanket news coverage because it showed a remarkable reversal from almost all previous polls.


But statisticians say that such polls do not merit the intense coverage they receive because, by showing such drastic reversals, they are likely to be inaccurate.

“It takes a lot to shift public opinion, especially when pitting two candidates who are as distinct as Clinton and Trump,” wrote Columbia University political scientist Andrew Gelman in The Washington Post on Thursday. “Preelection polls jump all over the place. Sometimes Hillary Clinton’s ahead by 20 points, other times she and Donald Trump are nearly tied. Jumping polls can make for frenzied reporting, but much of this is just noise.”

In another example of a poll that “jumps all over the place,” a Reuters/Ipsos Trump vs. Clinton presidential election tracking poll has Trump coming from behind to virtually tie Clinton on May 12. But by May 15, just three days later, Clinton had opened up a 6.2 point lead in the same poll. The Real Clear Politics polling average as of May 19 shows Hillary Clinton with a 3.3 point national lead over Donald Trump.

[Featured Photos By Mary Altaffer / Andrew Harnik / Associated Press]