Yesterday, veteran Louisiana pollster Bernie Pinsonat released a new survey of the Louisiana governor’s race which adhered to a similar narrative that fellow long-time pollster Verne Kennedy had offered a month ago – namely, that John Bel Edwards is on the cusp of potentially winning re-election in next month’s jungle primary.

Gov. John Bel Edwards leads his two Republican challengers by 7 points—47% to their combined 40%—in an independent statewide poll released today by Baton Rouge pollster Bernie Pinsonat, who conducted the survey for a group of undisclosed private clients.

But with 14% of voters undecided and the election is still a month away, it’s too soon to predict whether Edwards will win outright in the primary.

“He’s in good shape,” says Pinsonat, who conducted the poll from among 500 chronic voters from Sept. 3-6. “If the election were held when this poll was taken he would probably be elected. But we still have one month to go.”

Trailing Edwards with 24% of the vote is U.S. Rep. Ralph Abraham, followed by Baton Rouge businessman Eddie Rispone with 16%.

Among Republican voters, 42% said they would vote for Abraham, while 27% said they would support Rispone. Edwards picked up 15% of that Republican support, while another 15% were undecided or would not say.

Among Democrats, 79% said they would vote for Edwards, compared to 10% for Abraham and 5% for Rispone.

There are a couple of items which pop out to us from Pinsonat’s numbers, though we wouldn’t claim too forcefully that any of these are more than just casual impressions.

First, in the last round of polling we saw Eddie Rispone was knocking on the door of 20 percent. Kennedy’s poll had him at 19. Pinsonat now has him at 16. It’s a mistake to read too much into poll comparisons from differing sources, because the sample and the questions will vary from pollster to pollster. But given that Kennedy’s poll was completed in mid-August and Pinsonat was in the field with this survey last week, with the volume of TV ads Rispone is running and the fact the last couple of spots he’s put up on TV have generally been regarded as better than the initial batch (the biographical ad he’s running now is definitely his best one yet), you would expect him to be growing in support over that three-week period, not giving back three points.

Again, we’re not talking about the same methodology or sample, so that’s just an impression and not a judgement. But it did strike us as interesting, and perhaps indicative that Rispone may be hitting something of a wall with respect to momentum. We’ve heard rumblings from a couple of places that private and internal polling had told a similar story. So that’s something to watch as more data comes in.

Another item we noticed was that the Republican vote has 15 percent undecided, while there is only 6 percent undecided on the Democrat side. Presumably that 15 percent undecided in the GOP camp will choose between Abraham and Rispone, Given that Republican voters are 35 percent of the sample that’s five or six percent you can split between them, and so Abraham and Rispone really combine for 45-46 percent.

Does Edwards automatically pick up that six percent of Democrat undecideds? With Dems being 44 percent of the sample, that six percent translates into about three percent of the total vote, which would be enough to get him very close to winning the election in October.

But it’s by no means likely he’s going to get that vote. First of all, he’s the incumbent. Incumbents generally don’t get the undecided vote unless something dramatic happens (and nothing dramatic has happened in this race, nor does it look like anything dramatic in Edwards’ favor is coming). And second, there are other candidates in the race who can get that vote. After all, if you’re a Democrat and you still haven’t decided you want to vote for Edwards it means you just don’t like him.

That leads us to our major critique of this poll, which is that Pinsonat didn’t include Gary Landrieu and Omar Dantzler. Yes, those are minor candidates and they probably figure into the undecideds. We get that. But if you’re trying to do a good ballot test it’s helpful to make your poll look somewhat like the ballot – and Landrieu and Dantzler both have some semblance of a constituency. Dantzler is the black Democrat in the race, and we know from recent experience that black voters in Louisiana will, in some relatively significant numbers, vote for black candidates over more established white ones. And Landrieu has, as he brags, “$50 million in name identification,” which will translate into some number of votes likely north of 2 or 3 percent. Our Hayride-MultiQuest poll in late July showed Landrieu with four percent.

We would have liked to see how much of the undecided vote in this survey was true undecided and how much was leaning toward the two minor candidates as a protest vote.

The other thing we’re not sure about is the treatment of the black vote, particularly with Dantzler not listed as an option. The black community is going to know Dantzler is a black Democrat by Election Day on Oct. 12, and even if they don’t, when they see a guy named Omar with a “D” where party identification is listed, and that guy sits above John Bel Edwards on the ballot, they’re going to know they can vote for somebody who identifies with them demographically. This has to be understood as a factor, even if it only moves a few percentage points.

Pinsonat has his sample set at 60 percent white, 29 percent black and the rest as “other” within his sample of 500 respondents. But in the crosstabs under race, he has 298 whites, 145 blacks and 30 “others,” which adds up to 473 respondents. That would make blacks 31 percent of the sample rather than 29 percent, and that’s way too high a number without Barack Obama on the ballot.

All of which tells us that Edwards, at 47 percent, probably overpolls in Pinsonat’s survey.

But nobody should look at this poll from a Republican perspective and be satisfied. Abraham ought to be in the 30’s right now and Rispone ought to be in the 20’s, and neither one has really hit his stride. The poll has Edwards at 54-39 on approval, while Abraham is 41-41 and Rispone is underwater at 37-43. Yes, it’s a great bet that when President Trump comes down for the runoff and the GOP unites behind one candidate this race will start to look more like you would expect in a red state, but it still looks like there’s a lot of work to do.

So nobody ought to be happy.

That includes Edwards. He might look at this poll and think he’s poised to win in the primary, but if he can’t translate 54 percent approval to 50 percent electoral support that’s an indication that approval number is awfully lukewarm and Louisiana can take him or leave him, so one good political hit could finish him off.