2026 Maine Democratic Senate Delegate Results: The Popular Vote

In my last post I mentioned that, based on the voting method, we should expect an overwhelming delegate victory in the Democratic Senate Convention vote. An overwhelming victory is exactly what we saw with Troy Jackson securing over 92% of the delegates. This means that more voters voted for Troy Jackson than any other candidate, but it doesn't tell us much else. A popular vote primary between these candidates probably wouldn't have resulted in such wide margins.

We can't get a direct popular vote because a popular vote didn't happen. People voted for delegates, and many of those delegates weren't pledged to a candidate at all. At least some people surely split their vote across delegates who supported different candidates and many voters did not vote for as many delegates as they could have. It is not obvious how these would be translated into a popular vote and voter level data isn't public (for good reason) so I couldn't try to reconstruct this if I wanted to.

What we can get at is popularity. By classifying every delegate, win or lose, and comparing the percentage of votes delegates for each candidate received we can get at an idea of how popular each candidate was. This is not exactly what the popular vote would have been, but it should be dramatically closer to what the popular vote would have looked like.

The biggest source of difference is that voting for fewer delegates than you were allowed to reduces how much your vote counts. While this takes us further away from the popular vote, it does still capture how enthusiastic voters were for a candidate on some level.

Popularity

In popularity, Troy Jackson received 45.8% of the vote. Does this make Jackson's win less impressive? I would say no. Full disclosure, I do favor Troy Jackson so I may be a bit biased, but I personally find this reassuring. There were four leading candidates going into this election and a lot of unaffiliated delegates. Jackson could have won that 92% with only 25% popular support, possibly less. But instead, in a crowded field, he won an outright majority of votes that could be linked to a candidate. Although he did not receive a majority when counting unaffiliated delegates, he received more votes than the 2nd and 3rd place candidates combined.

County By County

The county by county results of the popularity vote line up identically with the delegate results: Jackson won every county except Washington County.

I felt like this map was a bit boring because it doesn't really give you a sense of the geographic support for each candidate so I also plotted four mini-maps: one for each of the top three candidates and one for unaffiliated.

This was more interesting and I noticed that we could actually see candidate support around their home counties so I outlined them in yellow.

Here is a detailed breakdown of the popular vote for each county.

Slate Voting

In my previous post I found that ~97% of the winning delegates were from slates which suggested a high degree of slate voting. However, this was the winners only so it didn't tell us much about the slate voting for other candidates. Now we can look at the vote breakdown for every slate vs non-slate delegate win or lose.

Slate delegates still received more votes but only ~85% of votes compared to ~97% of the winning delegates. This suggests that slate voting was fairly popular even among other candidates and voters in general but it was not universal.

One wrinkle is that I would assume slate voters would have filled up their full ballots while non-slate voters would be much more likely to vote for fewer delegates. This would decrease the representation of non-slate voters. Unfortunately, it is tough to quantify this without vote-level data. To get a rough idea of how many ballots were left blank I did a breakdown of votes vs potential votes for each county.

There was no county with less than ~58% of votes filled so this suggests that for the most part people were filling out most of their ballots. This combined with the vote breakdown between slate and non-slate delegate vote totals suggests that slate voting was fairly popular amongst voters even though an exact ratio is impossible to calculate.

An interesting thing I noticed while plotting this is that by far the biggest determiner of how much of their ballot people filled was how many total seats there were in their county. This suggests that voters were overwhelmed by picking too many delegates.

Slate Discipline

At least to me, there seems to be a perception that Jackson organized the hardest around slate voting and I think there was some concern that his win came more from a disciplined slate voting strategy than popular support. This is actually something we can investigate in the data.

If every voter was doing perfect slate voting then we would expect every candidate in a county slate to get the exact same number of votes. The more the slate candidates vote totals differ from one another the less people are voting for them as a slate. We can plot the distribution of votes received by slate delegates for each candidate we can get a sense of how much their voters were slate voting. There is also a metric called QCD (quartile coefficient of dispersion) that is a single number that captures how spread out the vote totals are. The closer to 0 the less distance there is between the 1st and 3rd quartiles of votes received for each slate delegate.

The data somewhat backs up the notion that Jackson voters tended to vote the slate more consistently but compared to Shah it was not that much more consistently. The distributions between Jackson and Shah don't look that different and their QCDs are much closer to each other than either of them are to Bellows. The more disciplined slate voting must have been helpful to Jackson on some level but it doesn't appear to be enough on its own to explain the popular vote difference.

Takeaways

The takeaways here are pretty boring and I personally think that is a good thing. The candidate that won was the most popular but within margins you would expect given the recent governor results. To me, this suggests these are results we can be confident in and we should be proud we put a process together like this in such a short time.