Consider this a Rorschach poll, especially in the context of the swing-state reports of late in the presidential race. Democratic leaning pollster PPP has the first survey of Pennsylvania in over a month, a state Donald Trump’s campaign says it will win but which has gone Democratic since 1988. Readers who like Trump’s chances will feel pretty good about the toplines, but they come with some caveats as well:

Clinton has 41% to 40% for Trump, with Gary Johnson at 6% and Jill Stein at 3%. In a head to head match up, Clinton and Trump tie at 44%. Similarly to what we found on the Florida poll we released yesterday, Republicans are more unified around Trump (79/8) than Democrats are around Clinton (75/15). That dynamic is what’s making the state competitive.

Democrats lead a generic ballot question for President 45/41, which is pretty similar to Barack Obama’s margin of victory in the state in 2012. Bernie Sanders has a substantial advantage over Trump, 45/36, with Johnson at 5% and Stein at 1%. In a head to head contest Sanders leads Trump 51/39.

To recall a point made in my earlier post, Obama won Pennsylvania by more than five points, 52/46.6 against Romney four years ago. The best any Republican has done since 1988, when George H. W. Bush became the most recent Republican to win the state, was his son’s 48.5% in 2004 — still 2.5 points behind John Kerry. And that was as an incumbent; George W. Bush only got to Romney’s level in 2000, with just 46.4% of the vote. Republicans don’t get blown out in Pennsylvania, except maybe in 2008, but they generally don’t compete well there either.

Being within a point of Clinton in the three-way race puts Trump closer than any other GOP candidate’s final outcome since 1988, and that’s a potentially encouraging sign. However, we’re not at the general election yet, and Trump’s share of the poll still comes up short of John McCain’s 44.3% in 2008. Remember, Trump’s been campaigning for over a year and has been the presumptive Republican nominee for almost a month. He should be well known, and his support solid. Solid at 44% is not a very good position in a two way race under these circumstances.

As PPP notes, the division within the Democratic bloc makes this an uneven comparison, too. If Hillary fixes that unity disparity, suddenly she polls at 47/40, which would be the worst gap for the GOP in PA since 2008. Perhaps a signal of what that might look like comes in the head-to-heads with Bernie Sanders and Trump, which Sanders leads by wide margins at 45/36/5 (with Johnson) and 51/39 in the two-way race. Trump doesn’t draw enough from the Sanders bloc to go ahead even with the sharp division still festering among Democrats, and when that bitterness winds down, Trump will draw even less from it.

There is one other caveat, albeit small. The sample does have a 49/41 split on the 2012 vote for Obama, slightly larger than the 5.4-point final gap, so this may be a bit friendlier to Hillary, but … that’s not much of a difference. Even with the results from Luzerne, there isn’t much evidence that Trump changes the situation in Pennsylvania for Republicans. At least, not so far.