Landed with Bad Ratings... Failed at polls...

Trump's still in trouble in first post-convention polling

Analysis by Harry Enten

President Donald Trump wanted a large bounce coming out of the Republican National Convention, and early evidence is that he did not receive one. Trump may have gotten a small bump out of his convention, but he still clearly trails former Vice President Joe Biden.

If later polling data confirms this early evidence, a Trump victory hinges on him becoming the first incumbent in over 70 years to come from behind after trailing following the major party conventions.

The new ABC News/Ipsos poll is especially harsh for Trump. Although it did not poll the horse race, it did poll the candidates' favorability ratings.

Trump's favorable rating stood at 32% in an Ipsos' poll last week. Today, after the Republican National Convention, it stands at 31%.

Biden's favorable rating, meanwhile, was at 45% last week. It's now at 46%.

These movements, of course, are well within any margin of error, though they line-up with the idea that Trump did not receive a large bump out of his convention.

There are no post-convention polls that measure the horse race of Biden vs. Trump that meet CNN standards. The evidence we do have, however, is largely consistent with what Ipsos is hinting at.

The USC Dornsife Presidential Election Poll is a tracking poll taken over a week period. It has both a unique probabilistic voting question as well as the more traditional "who are you voting for" question. With the days added after the Republican convention, the poll shows that Biden's advantage this week is about the same as it was last week.

Indeed, if you were to average all of the national polls taken at least partially after the conventions, there may have been a bump of maybe a point or 2 at most for Trump.

Interestingly, that might make for a slightly larger bounce than Biden received, which was basically nothing. That fits the historical pattern of elected incumbents usually getting the larger bounce from the conventions.

Still, it's not anywhere close to where Trump needed his bounce to be. Remember, Trump trailed by around 9 points heading into the conventions.

The lack of bounces does make sense given that fewer people watched the conventions this year, which is especially the case for the Republican convention.

Now obviously that could change as we get more polling data, which Trump better hope it does.

An examination of polling data since the end of World War II reveals the obvious problem for Trump. Elected incumbents tend not to gain after the conventions.

Looking at elected incumbents, 6 of 8 times the incumbent lost ground when comparing polls immediately following the convention and the final result. One was Dwight Eisenhower in 1956, who was leading by double-digits and went on to win by a slightly wider double-digit margin. The other was George H.W. Bush in 1992 who trailed by single digits and went on to lose by a slightly narrower single-digit margin.

No elected incumbent since World War II has ever picked up more than 4 points when comparing the post-convention polls and the final result. Most incumbents lose ground.

The biggest gains for incumbents come when they weren't elected, such as Harry Truman in 1948 and Gerald Ford in 1976.

The only instance of any incumbent coming back to win at this point in the campaign since World War II was Truman in 1948. And even he needed an additional polling error at the end of the campaign to complete the comeback.

The bottom line is this for Trump: something major needed to change in the post-convention polling. It doesn't look like it did in the first round of evidence. That means unless something shifts, Trump will have to do something quite unusual to earn another term in the White House.