No ballistic information.
>> Are you sure?
Take your time, Major.
Any day now, Marvin.
>> It was at 38 and quarter on Erdman Avenue, sir.
>> So what you are in effect saying, Major is rather than looking at
this cluster of robberies you had along Monument Street,
you chose to investigate each using separate personnel.
>> No, sir.
Three of those cases were commercial robberies, so
we sent those downtown to CID.
My people had responsibility.
>> Goddammit, Marvin.
Did you just invoke the word responsibility to
explain why you didn't get on top of this shit?
Did you just do that?
I don't think you know what the word means.
When I look at the 24s and plainly see that some 38-caliber asshole is
raping your entire district all the way from Greenmount to fucking Edison Highway.
It's policing like that that makes it possible for
us to get hit with five fucking homicides in a single fucking night, people.
>> Poor fucking Marvin.
>> Believe it or not, he's right.
>> You're still not connecting the dots, Marvin.
>> In all the CompStat meetings I've ever been to,
I've never seen anything remotely like the kind of vituperative attack on leaders,
subordinate leaders in front of an audience like happens in those shows.
It happened to Bunny in that one.
It happened to, and again I said I, I believe in some of those, at least one
of those segments that I looked at, the, the commander Marvin I think his name was,
who was, did not appear to be on top of his crime patterns and
did not have thorough, detailed knowledge of the material in his own
briefing documents about forensics and the fact that there was, that there had been
a number of shootings involving a similar weapon, the same 38-caliber weapon.
The fact that he said there was no forensic data, ballistic data, rather and
then there was in his folder and he had to correct himself.
Those are things that probably would get a commander into some trouble at
a New York City CompStat meeting.
I think that I've written and I believe based on everything I've talked to and
seen in connection with CompStat in New York,
that people would never get in trouble because crime was not coming down.
They did get in trouble if they seemed to not be in touch with the crime that was in
their district that they were responsible for.
If they didn't sort of have an operating plan and
weren't familiar with how it was working or if there had been a lesson learned
in another previous edition of CompStat because after every CompStat meeting,
if there was something that was a success story, there would be a memo go out,
a CompStat memo that said, a commander who had faced this problem,
these weren't specifics, this is what he did and this was the result.
And if for people it seemed to be a successful result,
other commanders facing similar problems were supposed to learn that lesson.
And, you know, for me, the, the big advantage of performance-based management,
evidence-based management is if you have real-time feedback, it enables
you to accelerate the process of learning what's working, what's not working.
Prior to CompStat, prior to 1994, all of the crime data
analysis was done centrally by the Management Information Systems Division.
They would get the data from all the then 75 precincts.
It would be processed and three months later, a quarter of a year later,
a report would come out with detailed analysis of crime patterns in all
the subareas of the city and the boroughs of the city and the city as a whole.
Well, in policing, three months later is history, and
Bratton came in and said, every precinct needs its own computer.
Every precinct commander, precinct commander, he needs to
get the data about crime in his or her district at the same time we do and
then we both have to analyze it, and then we have a have a conversation about it.
That was, that was where data and statistics and comparative statistics
CompStat kind of really came, where the rubber met the road, and
it had to do with real-time information about crime patterns and
focus on those prime patterns, crime patterns at the local level, but
think learning from that in a conversation with leaders of the department.
That was revolutionary.
It used to be that the precincts were very far removed from headquarters in
its decision making, whether it was what the operators were telling 911 calls,
call, you know, callers about sending officers to them all over the city.
Basically there wasn't that kind of local area focus,
decentralized command responsibility for bringing crime down.
And shifting to that, along with the shift to pro,
either proactive, preventative crime-fighting rather than
reactive crime-fighting are a huge part of the story of success in New York City.
>> The fact is that homicides are off 14% of the iii this year, but I'm not
going to stand here and take credit for that when all other UCR felonies are up.
>> Why not, Major?
>> A wise man once told me years ago when I wasn't even a sergeant that you
should never take credit when the crime rate drops,
unless you want to take the blame when it rises.
>> Ha ha ha.
>> Ten-hut!
Mayor elective on deck.
>> Please, back to what you're doing.
I'm just a councilman here to observe.
>> At ease, gentlemen.
Major Daniels, continue your presentation.
>> In any event, rather than allow my post cars to patrol at
random, >> Deputy.
>> I put two-man cars in the posts with the Wosterdrum Courts, Edmonton Avenue,
Poplar Grove, Fit and Baltimore Streets, North Avenue.
>> Now if nothing else, a two-man car is more aggressive and that
helps push it away from the most volatile corners, which is maybe having an effect.
>> Just pushing it.
>> I'm not devoting my resources to cheap street arrest, commissioner.
Instead, I'm trying to bring every officer I
can to bear on making quality felony cases.
I could do more, but well, it's my opinion that not enough of the troops I
have are sufficiently trained to properly investigate such cases.
>> Major, you just insulted your command.
>> Respectfully, sir, too many of my people have
spent years chasing street-level arrests or they grab bodies and make stats, but
that doesn't teach you how to write a proper warrant or
testify properly as the probable cause or use and not get used by an informant.
I wish it were otherwise, but
if I'm chasing felony cases, I can only bring a minority of my officers to bear.
>> He was saying, I don't have the resources to address every priority,
so I'm having to, you know, I'm having,
having to focus on my highest priorities and do something about those.
And, you know, part of the message was, because the newly-elected mayor,
a former council member who'd been elected mayor comes in and
sits in on the meeting where that exchange is happening.
And the mayor saw somebody in the police
department who he hadn't previously even known existed and
had never met, never heard him speak, and never knew what he was thinking.
And that happened a lot in New York City CompStat meetings.
With a police department with more than 30,000 officers, the people at the top
were very far removed from even relatively important people a few layers down,
but every week, twice a week at the beginning, precinct commanders and
their lieutenants would come to the podium and answer questions and
do presentations and, and be part of the whole show.
The top leadership was seeing people that they wouldn't have ordinarily seen and
it had the effect of freeing up talent spotting within NYPD.
And that's something that, you know, leaders of complex organizations have to
be successful at if they want to improve their performance.
It used to be if you read some of the earlier literature,
a sociologist writing about NYPD and books like Behind, Behind The Shield, for
example, by a guy named Niederhoffer.
He talked about how to get promoted in New York City, you need a rabbi and
it referred a little bit to the fact that New York has a certain Jewish culture, but
it wasn't really pertaining to a Jewish person.
It just meant that you needed somebody who was watching out for
you, promoting your career, an uncle, a friend of your father's, somebody in
the department at a higher level who was kind of helping you advance your career.
And understandably, that's how people at a higher level would find about how,
you know, talented people at a lower level.
They would rely upon the people they know.
So I've got this job to do, is there, can you suggest somebody?
And they would suggest the people that they were connected to.
Maybe ethnically, maybe in some terms a church connection or,
a variety of different ways they could be connected, but
you needed a hook, you need a connection.
CompStat opened up a whole new opportunity for
leaders to review and spot talent.
And we knew from watching The Wire that Daniels was
a very serious talented person.
And the fact that the mayor saw him in this light, the mayor-elect,
which becomes again a political little thing that's going on the side of saying,
you know, you better watch out.
This guy, that guy is coming up fast to the very nasty deputy commissioner who's
been sort of leading the charge and sort of showing what a tough guy he is.
>> Year, we had MSAs in the 22 percentile, and
don't think they haven't noticed it down at the puzzle palace.
The word is they're looking for
at least a ten point increase from all city middle schools this time around.
>> We're still six weeks away from the MSAs and
you want us to start teaching the test now?
>> This year, the preferred term is curriculum alignment.
>> There's nothing wrong with emphasizing the skills necessary for the MSA.
If we can get them to write a paragraph without a four-letter word in it,
it'll have to have a better command of English.
>> Marsha, skill sets are one thing, but
this has us teaching test questions directly.
>> Test questions that involve skills.
I don't see your point.
>> I don't want to see the point.
>> Were you really expecting something different than last year?
>> I don't see the math section.
These are all Language Arts questions.
>> Our greatest failing on standardized tests last year.
So for the time being,
all teachers will devote class time to teaching Language Arts sample questions.
Now if you turn to page 11, please, I have some things I want to go over with you.
>> I don't get it.
All this so we score higher on the state tests?
If we're teaching the kids the test questions, what is it assessing in them?
>> Nothing.
It assesses us.
If the test scores go up, they can say the schools are improving.
The scores stay down, they can't.
>> This evidence-based measurement in management, performance measurement in
management is, and very much dependent upon whether or
not you have conceptualized performance in
a way that's consistent with what the public really wants, and whether or
not you figured out how to measure it in a way that's valid and reliable.
And I don't think we're there yet in terms of education.
You know, there are people, there's lots and
lots of controversy about what it is about education
we're trying to achieve, and it's because we're trying to achieve so much.
And in, in, in, in my lifetime, I think a lot of additional things have been
added to what it is we expect our schools to do for our kids.
And, you know, trying to come up with measures that capture that,
it's not where we, I don't think we've done that yet.
So that may be a difference between CompStatting schools and
CompStatting police departments.
That is a real distinction.
I have no doubt in my own mind having watched this
evidence-based real-time accelerated learning about what works and
doesn't work approach to management, like Call CompStat.
I have no doubt that if you have good measures, it's a powerful and
probably successful way to manage.
And my question is what's the alternative?
Michael Bloomberg regularly said, if you can't measure it, you can't manage it, and
I kind of believe that.
You know, if you don't have meaningful feedback in a control system,
a cybernetic system or any other system, how do you control it?
If, if we have difficulty agreeing on what it is that you're trying to measure,
then of course a, a, a measure, a measurement that guides management
decision making is going to satisfy some constituencies and
very much dissatisfy other constituencies.
Joan SubiratsCatedrático de Ciencia Política
Jaume BlascoMáster en Administración Pública
Ismael BlancoProfesor Agregado e Investigador Ramón y Cajal
Quim BruguéCatedrático de Ciencia Política
Raquel GallegoProfesora Agregada
Margarita LeónInvestigadora Ramón y Cajal
Marc ParésInvestigador y coordinador del grupo "Urban Governance, Commons, Internet and Social Innovation"
