Big gaps in 911 response in Minneapolis

A Star Tribune analysis of more than 70,000 calls shows disparities in response times across the Twin Cities.

Rick Dunbar watched from his car as a man tried to force open the door to his daughter’s Minneapolis rental property. He called 911.

And then he waited — for 31 minutes. By the time police arrived at the Fulton neighborhood house, the man had left.

“If this guy would have just kicked the door in ... he could have took TVs and whatever else and they still would have never caught him,” Dunbar said of the 2017 incident.

The response might have been faster if Dunbar called from another neighborhood. But police take longer on average to respond to high-priority 911 calls near Minneapolis’ southern border, according to a Star Tribune analysis of more than 70,000 calls in Minneapolis and St. Paul. In some neighborhoods, median response times for these calls were 5 minutes longer than Minneapolis’ citywide median of just under 8 minutes.

Response times on high-priority calls
< 6m
6m+
7m 30s+
9m+
10m 30s+
12m+
Fewer than 10 incidents
No incidents
Select feature
Median response times for high-priorty 911 calls in 2017 for the Minneapolis and St. Paul police departments in the Twin Cities, divided by quarter-mile hexagons.
Map showing:

Neal Hagberg frantically called 911 last fall near Minnehaha Parkway after two men chased him for several blocks during an attempted carjacking. It took about 12 minutes for officers to arrive. “I felt really vulnerable at the time that it took for them to get to me,” Hagberg said.

Fewer people call 911 from the areas where the analysis shows higher response times, police department officials said, leaving more officers clustered closer to parts of the city with more calls. Officers traveled more than 4 miles, from near Lake Street, to respond to Dunbar’s call.

“We have to walk that line between, ‘Do I add more officers over here so I can get response times down? Or do I add officers over here to try and reduce violent crime?’ ” said assistant chief Mike Kjos. “And it’s not an easy answer. There’s no formula.”

The department has already changed some procedures in response to Star Tribune inquiries. The department reviewed 50 high-priority calls across the city — chosen at random by the Star Tribune — that took more than 30 minutes before officers arrived. In a number of those cases, the department discovered, callers probably waited longer because officers were changing shifts. The data show that citywide median response times rose dramatically during the 6 a.m. hour, when the day shift begins and calls are also at their lowest volume.

Response time spikes at 6 a.m. shift change
Median response time of high-priority 911 calls in 2017 by hour for the Minneapolis Police Department. During the 6 a.m. hour the median response time jumps up significantly; this is also the hour with the lowest number incidents.

Now, Kjos said, precincts will keep a squad car available during shift change.

“It brought to light a clear issue that needed to be addressed,” Kjos said.

The department also found that response times sometimes exceeded 30 minutes because the priority of the call had been escalated or downgraded as more information became available. Sometimes the situation was less urgent than first thought, or officers were busy during high-demand times like bar close. In other cases, call records did not offer a clear explanation for the delay.

A half-hour response to high-priority calls is not the norm, even in neighborhoods with higher response times. When it happens, 911 callers take notice.

Just east of Lake Nokomis, in the parking lot of Oxendale’s Market, a man who had just agreed to give another man a ride pulled a sawed-off shotgun out of his trunk and said, “You know what this is. Give it up.” The would-be passenger handed over $305, but then tackled the gunman and a melee ensued.

The driver fired the gun as the victim ran away. No one was hit.

Police arrived 40 minutes after the 911 call. The precinct inspector later wrote in an e-mail to the local council member that all precinct squads were tied up on other calls.

“I was dumbfounded by it, to say the least,” said Alan Zocher, Oxendale’s store manager. “I thought that would have been a big enough issue where they would have gotten here faster.”

‘Always been a challenge’

The department regularly reports response times for the city’s five precincts, which are broad areas that include both crime hot spots and sleepier environs near the border. The Star Tribune used 2017 data from computer-aided dispatch to calculate the median times for much smaller areas, illustrating dramatic differences between border neighborhoods and the urban core.

Dispatchers determine whether a call is a high priority, and the severity of those calls ranged from a suspicious person to an assault in progress. One woman called to report hearing screaming while walking her dog. In another instance, people called about an emotionally disturbed man carrying a knife into neighborhood businesses. Others involved home and business alarm systems.

There was less evidence of disparate response times in St. Paul, according to an analysis of the dispatch data. St. Paul had about a fifth as many high-priority calls during the period analyzed, but priority classifications aren’t identical between the cities.

“I’ve got neighborhoods next to each other that have vastly different times,” said Minneapolis Council Member Jeremy Schroeder, who represents a ward bordering Richfield and the airport. “That’s disturbing.”

Median response times stretched from 12 to nearly 14 minutes in the Page, Morris Park, Wenonah, Diamond Lake, Armatage, Fulton and Shingle Creek neighborhoods. Most of that time is spent driving to the scene. By contrast, response times were closer to 7 minutes in neighborhoods like Jordan, Hawthorne and Whittier, which have some of the highest numbers of calls in the city.

“The people that live around those [Chain of Lakes] areas and get higher property taxes aren’t getting nearly the patrol that’s going on at Lake and Nicollet and Franklin and Lyndale,” said Lt. Bob Kroll, who leads the city’s police union. “They’re not getting the same amount of coverage. That’s always been a challenge. And it’s always been a shame.”

Some, including Kroll, say the data illustrate a need for more police officers. Police Chief Medaria Arradondo has said he would like to see the department grow from 888 officers to 1,000. But police department funding has been a controversial topic at the City Council.

“From a budget perspective, why don’t we staff our precincts appropriately? Because 911 callers need help now,” said Council Member Linea Palmisano, who represents the city’s southwestern corner. “Part of what we need is the political will to give police departments the resources they need to be effective first responders.”

Some of the response time differences are a result of the dispatch process. Kjos said officers are responsible for handling calls in their respective response zones, which are generally larger in areas with lower call volumes.

“I think that it’s unfortunate that when you live in an area that doesn’t have very many calls, that that area tends to get to be a larger space for an individual squad,” Kjos said. “Which makes it more difficult to get from point A to point B in a timely manner.”

‘Where are they?’

In Dunbar’s case, he later learned that the man trying to force his way into the house was probably an acquaintance of a tenant. Dunbar had to replace the door because it was so damaged from being kicked.

“It just seems like it’s taking longer and longer for them to respond,” Dunbar said.

In October 2017, Liz Friedman was walking her dog in Fulton when she heard a woman yell, a man’s voice and a loud bang — like a large piece of furniture crashing to the ground. It sounded to her like a domestic disturbance.

“It was just very out of character for this neighborhood to have such noises in the middle of the day,” Friedman said. She called 911 to report what she heard, while second-guessing whether the call was warranted.

Her call took more than 20 minutes to be assigned to an available squad, which spent another 13 minutes driving to the scene. She hopes the response would be faster for other high-priority calls like an intruder.

“That’s too long for an emergency response,” Friedman said. “An emergency means fast.”

Credits

  • Edited by James Schiffer
  • Data analysis by Eric Roper and Alan Palazzolo
  • Design and development by Alan Palazzolo

Methodology

The Star Tribune analyzed more than 700,000 dispatch records involving Minneapolis and St. Paul police for 2017. The data, provided by Ramsey County and the city of Minneapolis, included both 911 calls and incidents reported by officers.

The analysis focused on high-priority calls that were not initiated by an officer (such as a traffic stop or a crime witnessed by an officer). "High-priority" comprised priority codes 1 and 0 in Minneapolis, and 2 and 1 in St. Paul. The two dispatch centers have different methods of prioritizing calls, and the priority level of a call is sometimes adjusted after additional information becomes available.

Incidents that were likely reported by officers were removed from the analysis using timestamps in the data.

Response times were calculated as the time between "first keystroke" (or "call pickup", if keystroke data was not provided) to "first unit arrived". Incidents where this could not be calculated were not included in the main analysis.

Records from incidents handled by the University of Minnesota police were not included in the analysis.

Code and more details for the analysis can be found here on Github.