Commentary: Measuring Metrics

• Author: Rocco DeBenedetto
• DOI: http://doi.org/10.1111/puar.12743
• Published date: 01 March 2017
• Date: 01 March 2017

Measuring Metrics 193

Public Administration Review,

Vol. 77, Iss. 2, pp. 193–194. © 2017 by

The American Society for Public Administration.

DOI: 10.1111/puar.12743.

Rocco DeBenedetto , a 20-year veteran

of the New York City Police Department,

rose from officer to deputy inspector and

commanded Brooklyn ’ s 84th Precinct. He

now serves as senior consultant with a

global security solutions firm located in

New York City.

E-mail: roccodeben@yahoo.com

Commentary

T he New York City Police Department ’ s

(NYPD) introduction of CompStat radically

changed policing, most notably at the

executive level. It was a novel concept in 1994: the

parading of precinct commanders to a miked-up

podium where they stood naked (figuratively),

exposed to relentless grilling by the department

hierarchy, with no apparent boundaries or etiquette

and with rules made on the fly—by the upper

echelon, of course. A good number of qualified and

competent commanders ran for the hills, begging

for assignments within the property clerk division

or in some carpeted, air-conditioned headquarters’

office. Others filed for early retirement. Many stayed

in command, including several future chiefs of

department, who seized the opportunity to be on

stage and exhibit their knowledge and readiness to

be held accountable and demonstrate some much

needed leadership. Twenty-three years later, no one

can dispute or diminish the overall success of the

CompStat strategy and the critical role it has played

during this historic reduction of crime within New

York City.

The new age of CompStat was no place for timid

souls. This had some value. Commanders had to

now know who had been shot, where, when, and,

perhaps most telling, why. They had to coordinate

with detective squad supervisors to ensure that

investigations remained current and thorough. They

needed to know the hotspots where drugs were being

sold and coordinate with narcotic supervisors to

ensure that efforts and resources were directed toward

and targeted those locales.

But what soon evolved was a giant, all-encompassing

numbers game. Commanders eagerly awaited weekly

CompStat reports from borough headquarters every

Tuesday morning; all else took a backseat to the

consumption and digestion of this sea of numbers.

The first priority being “how ’ d I do?” If another

commander posted fewer robberies, the inference

was that he or she was a doing better job. Each

commander worried that if a neighboring precinct

issued more “C” summonses (summonses requiring a

warrant check, more serious than a traffic summons),

that commander might get the promotion that, of

course, should have been his or hers. The department

hierarchy successfully pitted the 95 or so alpha males

and females serving as precinct, housing, and transit

district commanders against each other, attempting to

stimulate passion, ownership, and ideas to foster crime

reduction as the endgame. CompStat clearly made

and destroyed careers.

My long experience as a police officer and police

commander in the NYPD allows me to recognize

several of the problems with CompStat cited by

John A. Eterno, Christine S. Barrow, and Eli B.

Silverman in their article “Forcible Stops: Police

and Citizens Speak Out.” The genesis of one of the

more significant problems became visible when the

individual metrics that were selected and accepted

as performance evaluators somehow became more

important than the overall mission and commitment

that each commander must demonstrate to his or her

community and constituents.

A commander ’ s worst fear might be standing at

the podium without a viable answer to the chief ’ s

observation and question, “You have a spike in

robberies in Sector Adam, Captain. What are

you doing about it?” A survival response could be

something like, “I regularly meet with and brief

my anticrime teams who have been directed to

concentrate on the affected areas. Within the past

three weeks, they have conducted 42 stop and frisks

and robberies in Sector Adam are down.” But should

such a response have offered a lifeline? Were the

stops conducted during the times the robberies had

occurred? Were they conducted in a legal manner,

and were the subjects stopped similar in description

to those described by victims and/or witnesses? Are

the officers able to articulate the level of reasonable

suspicion necessary? If officers had conducted 90

stop and frisks, would the number of robberies

Rocco DeBenedetto

Measuring Metrics