THE EFFECTIVE PROBLEMSOLVER #111
St. Paul is on track for a 70 percent drop in homicides this year.
That’s an extraordinary result—one of the sharpest declines in the country.
But what’s most interesting isn’t the number.
It’s how it happened.
Ramsey County Attorney John Choi recently explained the turnaround in an interview with MPR News.
The conversation was about crime, but it revealed something bigger: a model for how complex problems actually get solved.
The Power of Focusing on What Matters
When St. Paul’s leaders looked at violent crime data a few years ago, they didn’t start with a new slogan or initiative.
They focused on a deceptively simple metric: solve rates—the percentage of crimes that are actually solved.
In 2021, St. Paul solved about 37 percent of non-fatal shootings. Today, that number is 70 percent.
Why does that matter?
Because what you measure determines how you behave.
Focusing on solve rates shifted attention from reacting to headlines to learning from patterns.
It created a feedback loop: the more cases solved, the more trust earned from the community, and the more people willing to cooperate in future investigations.
Choi put it plainly:
“If we don’t solve crimes, we lose the trust of communities.”
This is the kind of feedback loop I often write about in complex systems work.
Improving outcomes isn’t about one-time breakthroughs—it’s about strengthening the connections that make progress self-reinforcing.
If that idea sounds familiar, you might remember the bathtub analogy from TEP #040: the level of a problem depends less on effort and more on flow dynamics—the rates of what’s coming in and going out.
St. Paul didn’t just “try harder”; they changed the flows.
Tracking, Learning, and Adapting
St. Paul didn’t stumble onto this result.
They tracked data over time, set clear goals, and learned from others.
A group of local leaders visited Denver to study how that city had improved its clearance rates by treating non-fatal shootings like homicides—intense, immediate investigations that prioritize evidence and witness engagement.
They didn’t just copy the “Denver model.”
They adapted it.
They realized it wouldn’t work as an add-on inside their existing homicide unit, so they created a standalone non-fatal shooting unit.
That’s a structural decision—a recognition that you can’t expect new results from the same organizational design.
This move reflects a key systems thinking principle: when change stalls, don’t push harder on the same lever.
Redesign the system to fit the situation.
This kind of adaptive learning is at the heart of the problem-solving process I teach—what I described in TEP #029: How to Embrace a Strategy That Never Loses—learning to evolve your approach as the system itself changes.
Collaboration Based on Humility
None of this happened in isolation.
St. Paul formed a Violence Reduction Leadership Group in 2021, bringing together people from the courts, the prosecutor’s office, the police, city council, county commission, and mayor’s office.
At first, they spent their time blaming each other.
Crime was up, morale was down, and everyone was under public pressure.
But instead of walking away, they stayed in the conversation.
Over time, they shifted from defensiveness to curiosity—from “who’s at fault” to “what’s going on here?”
Choi said they had to “show up with some sense of humility and grace and start working better together.”
That line stuck with me.
It captures something essential about problem solving in complex systems: collaboration isn’t about aligning interests or signing an MOU.
It’s about building the trust and humility to learn together.
In TEP #085, I wrote about how 1:1 relationships—built over time, not through initiatives—are what really move systems.
This story is another example: progress came not from the perfect plan, but from people staying at the table long enough to learn together.
The Trust Factor
For all the structural and procedural improvements, St. Paul’s deep foundation of community trust might be the biggest reason for its success.
For decades, the city’s homicide clearance rate has been above 90 percent—unusually high nationally.
That history of responsiveness built credibility.
When residents believe police will take their reports seriously, they come forward with information.
When witnesses feel safe cooperating, cases get solved.
Trust isn’t just a moral value—it’s a performance driver.
You can’t buy it, mandate it, or post it on a website.
You earn it through consistent follow-through and results that matter to people’s daily lives.
Pragmatism Over Ideology
Choi also made a pragmatic point that’s easy to overlook: this kind of work isn’t possible without enough officers.
The city’s progress depended on having the capacity to assign dedicated investigators to a specialized unit.
That may sound obvious, but it’s worth saying in an era where public debate about policing tends to collapse into extremes—either “defund the police” or “back the blue.”
Neither slogan helps leaders understand how systems actually function.
Complex problems demand enough capacity to do the work, coupled with enough accountability to do it well.
You can’t have one without the other.
The Broader Lessons
For me, this story reinforces several lessons that apply far beyond public safety:
- Pick the right metric. Solve rates matter because they measure effectiveness, not just activity.
- Learn from the system. Track data over time, test changes, and watch for feedback loops.
- Collaborate with humility. True collaboration begins when the blame stops.
- Redesign the structure. Sometimes progress requires a new unit, not a new slogan.
- Build trust as you go. It’s the most important form of infrastructure in any system.
- Stay pragmatic. Ideological purity is no substitute for functional capacity.
A Model for What Progress Looks Like
When I first started working in systems change, I thought the key was to design the right policy or fund the right initiative.
Over time, I’ve learned that real progress looks more like what’s happening in St. Paul right now—practical, data-informed, collaborative, and rooted in trust.
It’s not flashy.
It doesn’t fit neatly into a campaign message.
But it works.
And that’s the quiet power of complex problem solving done well: it’s not about controlling the system from the top.
It’s about helping the people inside it learn faster, coordinate better, and keep faith with the communities they serve.
See you in two weeks.
-Bryan
