On October 4, 2025, we wrote about how candidates competing for Minneapolis offices self-sort into two distinct groups, on the basis of endorsing and campaign behavior. The networks, and candidate list derived from them, were constructed using endorsements and shared campaign events known as of September 26. Below are the networks, and the candidate list, using endorsements and shared campaign events known through October 26.
The two groups, which we called Far-Left DSA and Traditional Liberal, still are distinct. There is no overlap. No candidate in one group endorses or campaigns with a candidate in the other. The main difference in one month is that the Traditional Liberal party has become much more internally cohesive, evident by the announcement, on October 26, of the “Team Minneapolis Slate,” led by incumbent Mayor Jacob Frey.
We encourage readers to use this guide, whatever their preferences. We believe that every voter should know which party best reflects their views and vote for candidates in that party accordingly. When most candidates in both parties call themselves “DFL” that label says nothing, it’s not a guide. We offer this updated list as a guide.
Why do you consider Steve Brandt to be DSA?
Probably because he broke with the DFL and endorsed Bernstein for B.E.T. and not Bob Fine.
This is red baiting. I personally don’t like DSA as an organization, and a candidate carrying that endorsement is less likely to get my support. That being said, there are many non-DSA progressives in Minneapolis – and all too often they get trashed as “DSA aligned” by Frey supporters. This has become so overused, that it has lost power as a means to persuade. Drawing spirograph like images to try to reinforce your categorization of people doesn’t clarify things. Democrats need to be a party of addition – not subtraction – and your tarring progressives with the DSA aligned label, subtracts, rather than adds, and hurts DFL changes in the general next year.
See my reply below. I did not draw the images. I defined “relationship” between two candidates if one endorsed the other and/or if they had one or more shared campaign events. The original data are in a Google sheet linked in the original article. I fed the data into Gephi, which is an open-source network visualization tool, and Gephi produced the images you see. The clustering you see is candidate behavior, nothing more, nothing less.
Important observation “The clustering you see is candidate behavior, nothing more, nothing less.”
Please read the methods. Candidate ID by party is based on their own endorsing and campaign behavior, not on what I think or what I consider. By definition, two candidates are linked if one endorsed the other and/or if they shared a campaign activity. The strength of the link is reflected in the line thickness in the diagram. Formal endorsement is stronger than a single campaign event; multiple campaign events are stronger than one. All this was explained in the original article, and in the original article there is a link to a Google sheet containing the data, with more explication of the methods.
The diagram was produced with Gephi, which is an open source network visualization tool. You feed Gephi the data, and it returns the network, using a principle in which “nodes” tightly connected are close to each other, while those connected but less tightly are farther apart. (The word ‘node’ comes from network theory, which actually is based on graph theory, the mathematics of modeling paired relationships).
If you examine the network, you’ll see that Brandt is way on top of the lower left cluster. Also in the original article, I made clear that we need names to succinctly refer to these two “parties”, because the word ‘DFL’ has lost all meaning. When almost everyone calls themselves “DFL”, but they endorse and/or campaign with with completely different people, they vote differently, and have completely different policy proposals, the label ‘DFL’ says nothing at all.
I decided to call the lower left cluster ‘Far-Left DSA’, because DSA-endorsed candidates are at the center of that cluster, and incumbent CMs in that cluster, who are not DSA-endorsed, vote with the DSA-endorsed CMs more than 88% of the time. A network showing these pairwise voting relationships also is in the original article.
You are welcome to invent and apply your own name if you feel that “Far-Left DSA” is emotionally triggering. For me it’s descriptive, and since I put in all the work to collect the data, figure out how to display it this way, and offer it to you, the readers, I get to pick the labels I want. Call it author privilege, or payment for the effort involved.
As for Brandt, he landed in that cluster because he endorsed DeWayne Davis, and Davis has many campaign events with Omar Fateh, and that cluster’s Council candidates, plus some endorsements from that cluster’s Council candidates. So through his endorsement of Davis, Brandt got associated with that cluster, although he is on the periphery.
Brandt did not endorse anyone in the other cluster, in the upper right, nor was he endorsed by anyone in the upper right cluster, nor did he shared any campaign events with any of them. So he is completely disconnected from the upper right cluster, which I called “Traditional Liberal”. Again, author privilege. You can name the two clusters however you want, but they are what they are, because of how they chose to associate themselves with other candidates.
This is the link to the original article, ICYMI. That came out on October 4, and it used endorsing and shared campaign events known through September 26. This one is an update that uses additional endorsements and/or shared campaign events through Oct 27.
https://minneapolistimes.com/mpls25-an-election-guide-for-the-perplexed/
1. I greatly appreciate the effort & work you put into analyzing these relationships & the visual representation provided to better understand the 2 factions that seem to represent ‘DFL’ politics these days. I think we can all agree keeping up with this new, extremely ‘progressive’ brand of ideology & knowing which candidate rep’s said ideology, is harder than staying up-2-date on the kids new slang terms (6-7?).
2. No issues on the Author Privilege – you frame it, you name it. If the labels provided in invoke a certain feeling/emotion, probably for a good reason stemming from personal experiences with said political ideology labels. A lot about politics is based on subjectivity.
Appreciate your efforts!