Continuing in our weekly series, the latest Morning Consult poll of the presidential race is out and this time we’ve seen a slowing of the shift that Donald Trump has recorded in his favor over the past few weeks. It remains a two point race with plenty of voters still either undecided or leaning toward a third party option. Neither major party candidate is close to a majority at this stage.

Donald Trump remains within the margin of error of Hillary Clinton in Morning Consult’s new national survey of likely voters.

The new poll, conducted from Sept. 6 through Sept. 8, shows the Democratic nominee leading the Republican nominee, 41 percent to 39 percent, if the election were held today. Libertarian nominee Gary Johnson and Green Party nominee Jill Stein are the first choice for 10 percent and 3 percent of voters, respectively, well below the 15-percent threshold required to make the presidential debate stage on Sept. 26.

In a two-way match up of likely voters, Clinton’s lead is even smaller, 44 percent to Trump’s 43 percent.

This survey really isn’t any different in a statistically significant way than last week’s entry when Clinton also led among likely voters by two points, 42-40. This represents a stall in the post-convention trend. On August 16th she led by seven, dropping to six on August 22nd and three on August 28th. The effect of third party candidates in the four way contest is similarly unchanged, with Gary Johnson picking up one point and Jill Stein dropping one.

The real shocker for me comes in the break between registered and likely voters. For some time now I’d been anticipating the shift from the former to the latter, expecting Trump to continue to do better among likely voters base on his slightly better enthusiasm numbers. For this survey, Morning Consult actually breaks down the difference between the two models and it doesn’t push the margins at all.

mc911

Once again we may have to delve into the crosstabs for answers. The sample for their registered voter results this week comes up with a split of 38 Democrat, 32 Independent and 30 Republican. Last week it was 35/32/32. For the likely voter model things are even more strange. This week the D/I/R ratio jumped to 40/27/32. That’s a net five percent jump for the Democrats and a matching five percent drop for independents.

Still, even if we set aside the party breakdown shift, is there really no difference between the registered and likely voter models? Not every survey finds this and it stands in contrast to some other current samples. Take the latest from ABC and the Washington Post for example.

Clinton holds a 46 percent to 41 percent edge over Trump among likely voters, followed by Libertarian candidate Gary Johnson at 9 percent and the Green Party’s Jill Stein at 2 percent. Clinton’s lead swells to 10 percentage points among the wider swath of registered voters, 45 percent to 35 percent, similar to her 45 percent to 37 percent edge last month.

Trump is down by five in that poll, but their split is 34 Democrat, 24 Republican, 33 Independent (with 5 “other” and 3 “No Opinion”). Still, the difference between the registered and likely voters is telling. Trump does five points better with likely voters than the wider registered voter sample, so take from that what you will. For even more background, you can read about how Morning Consult builds their likely voter model here.

hillary-kaine