When we published our independent poll of the Democratic primaries for Governor and U.S. Senate, we tried to be careful about one thing above all: the difference between a lead and a lock. In the two weeks since, we've watched the number travel — into speeches, into fundraising pitches, onto social media — almost always stripped down to a single story: the frontrunner is converting, the race is effectively decided.

We understand how that happens. It's the cleanest line in the dataset. It's also, on its own, the most misleading, because it answers a question the poll didn't ask ("who's ahead in June?") and quietly implies an answer to the one that matters ("who wins in August?"). Those are not the same question, and the gap between them is the whole story.

So this is a second pass at our own data: not new numbers, just the same crosstabs we published (which you can still inspect line by line). We're doing it because we think the consensus that's formed around this poll is more confident than the poll itself supports. If a race is broadly believed to be over, there's no reason for voters to keep paying attention. We don't think this one is over. Here’s why.

Start with the headline. In the collapsed ballot¹ — after we pushed undecided voters on which way they lean — Cindy Holscher sits at 37%, with Ethan Corson at 10, Curt Skoog at 7, and 44% undecided. That lead is real and it is far outside the poll's margin of error. We said that clearly the first time and we'll say it again: this is not a tossup at the top.

But "leads by a lot" and "has a lot of the vote" are different claims, and the second one isn't true. Break the 37% apart by how firm it is and it thins out fast. Only about 11% of the entire likely-primary electorate are committed Holscher voters — people who told us they've made up their minds. Another roughly 19% support her but say they could still change, and about 6 points are leaners we had to prompt out of the undecided column.

Put plainly: the frontrunner's rock-solid base is roughly one in nine of the people expected to vote. Everything above that line is genuine support, but it's provisional. A 37-point standing built on an 11-point foundation is a strong position. It is not a settled one.

Here's the finding we think is most underappreciated, and it comes straight out of the same "conversion" chart that's been making the rounds.

That chart's point (a fair one) is that among voters who've heard of each candidate, Holscher wins their support at a much higher rate (about 56%) than Corson (about 20%) or Skoog (about 13%). Those gaps are large and statistically solid. Voters who know Holscher tend to pick her, and that's a real asset.

But run the arithmetic one step further than the chart does. About 2/3 of likely primary voters have heard of Holscher. If 56% of those voters already support her, then she's already banked roughly 38 of a possible 67 points from the people who know her. That leaves only about 29% of the electorate who know Holscher and haven't chosen her — a pool she's spent a year working and has largely already harvested.

The uncomfortable implication for a frontrunner: the easy growth is mostly gone. The voters who were going to break her way on the basis of knowing her have, by and large, already broken. To grow from here, she has to reach the roughly one-third of the electorate who still don't know who she is — which is the exact same problem her challengers have, just approached from a higher starting point. The conversion chart is being used to argue momentum. Read completely, it argues something closer to the opposite: a candidate near the top of what her current level of exposure can produce.

We want to be precise, because this is the kind of claim that gets stretched. This is not "Holscher has peaked" — we can't measure a trend from one poll. It's a statement about structure, not trajectory: given who currently knows her and how they've already sorted, her remaining runway runs through the same low-information voters everyone else is chasing.

The obvious retort is that if Holscher's support is soft, the challengers' must be softer, and that's true. Among the small share of voters who've committed to anyone, only about a third say they're "definitely" locked in; the other two-thirds could still move. That softness runs across the whole field, frontrunner included.

But look at what that symmetry actually means. If nearly everyone's support is soft and 44% of voters haven't chosen at all, then the thing that decides this race isn't the June standings; it's who turns out and who breaks late. A shallow, broad lead in a low-information electorate is precisely the configuration in which polls and outcomes diverge. Not usually into an upset; more often into a result that's closer, or shaped differently, than the topline implied. The honest reading of symmetric softness is not "the challengers can win." It's "nobody in this field has this locked, and the frontrunner least of all can afford to act like it's over."

There is one place where Holscher's coalition is measurably, not speculatively, weaker. It survives statistical significance testing that a lot of crosstab "findings" don't.

Among women, Holscher runs at about 41%. Among men, she drops to about 30% — a gap of roughly 11 points that is well outside the noise, and her standing among men sits significantly below her own overall number. Half of male likely voters (about 50%) are undecided in the Governor's race. That's the single largest identifiable bloc of unclaimed, in-play voters in the primary, and the frontrunner has the weakest grip on it.

We're not going to tell you what any campaign should do about that (that's their job, not ours). But as a matter of description: if there's an opening in this race, this is where the data actually locate it. Not in a policy contrast, and not in consolidating the field. In the roughly one in two men who, a few weeks out, still haven't picked anyone.

We ran this poll so that every voter, donor, and candidate would be looking at the same independent information instead of dueling internal numbers. Two weeks in, our worry is the opposite of the one we started with: not that the poll would be ignored, but that it would be flattened into a verdict it doesn't deliver.

So, to be as clear as we know how:

The frontrunner has the best starting position in this race. She does not have it finished.

You can inspect every figure in this piece yourself: the full crosstabs, the topline results, and our primary models, which are projections, not predictions. Every number above traces to a specific crosstab cell; we'd rather you check our work than take it on faith.

Capitol Bee is a nonprofit civic journalism project of Civic Clarity, Inc., a 501(c)(3) organization. We don't have a paywall because we believe transparent government reporting should be accessible to everyone. State legislative bills are analyzed at BillBee.ai. Election forecasts live at forecast.capitolbee.com. Glossary of terms here.