WEBVTT
00:00:00.239 --> 00:00:04.240
Halfway through the year, I can tell you that our violent crime is down 19%.
00:00:04.879 --> 00:00:08.160
Our violent and property crime is down a whopping 22%.
00:00:09.119 --> 00:00:14.400
And our homicides are down over 63%, which is truly amazing.
00:00:14.640 --> 00:00:18.079
We could not accomplish this without your help and your support.
00:00:18.239 --> 00:00:26.160
Our detectives and officers are out there doing great, amazing work, and you all are sharing things with us and reporting crime and suspicious activity.
00:00:26.399 --> 00:00:29.920
Thank you for helping us make Tampa safer together.
00:00:30.320 --> 00:00:30.879
Hello all.
00:00:30.960 --> 00:00:33.039
Welcome to the Chicago Justice Podcast.
00:00:33.119 --> 00:00:34.640
I'm your host, Tracy Siska.
00:00:34.880 --> 00:00:44.560
I'm also Executive Director of the Chicago Justice Podcast and adject instructor in the Department of Justice, Law, Criminology, and Security at American University.
00:00:46.320 --> 00:00:52.399
The exploitation of crime and crime and violence data in America has reached epic proportions.
00:00:53.679 --> 00:01:02.079
Social media and the rapid spread of misinformation and disinformation have exponentially increased the impact of the misuse of crime data on communities.
00:01:03.200 --> 00:01:09.200
Pair this with the rapid rise of partisanship and extremism, and you have a recipe for disaster in our criminal justice system.
00:01:10.959 --> 00:01:23.040
One of the consequences of hyperpartisanship and the rise of extremism on the right is that we, as a side, have moved away from a climate of searching for and expanding best practices within our justice system.
00:01:24.079 --> 00:01:30.400
Instead, we have adopted an attitude of absolute denialism when confronted with data that challenges political views.
00:01:31.359 --> 00:01:42.319
This has led us to a point where, as crime goes down around the country, unethical and immoral politicians exploit individual violent incidents to challenge realities that cities across the country are experiencing.
00:01:42.959 --> 00:01:49.120
The media laps it up because they benefit as much as anyone from the political expectation of crime and violence.
00:01:49.920 --> 00:01:52.319
They're also a major exploiter of crime and violence.
00:01:53.040 --> 00:02:00.560
To dig into all these issues, we sit down with Jeff Asher, a nationally recognized crime analyst and co-founder of AH Datalytics.
00:02:01.280 --> 00:02:07.680
He writes a weekly newsletter and hosts the Jeffalytics Podcast, a show about measuring crime and communicating trends.
00:02:09.199 --> 00:02:10.319
Let's get to our discussion.
00:02:10.960 --> 00:02:12.800
Jeff Asher, thank you so much for joining us.
00:02:12.879 --> 00:02:13.680
We really appreciate it.
00:02:14.240 --> 00:02:14.960
Thanks for having me.
00:02:15.280 --> 00:02:31.599
So give my audience a brief explanation of the Uniform Crime Reports and how the national incident-based reporting system is a little more detailed than what people or police, I should say, report to the feds typically.
00:02:31.759 --> 00:02:33.599
And what are the differences between the two?
00:02:34.159 --> 00:02:44.719
So so the Uniform Crime Report was created in the 1930s, first by the International Association of the Chiefs of Police, and then FBI took it over a couple of years later.
00:02:44.960 --> 00:02:50.479
Um it is it NIBERS is a part of UCR.
00:02:50.560 --> 00:02:55.680
So UCR is sort of the big picture umbrella of crime data reporting.
00:02:55.919 --> 00:02:58.319
Um we get hate crime data through that.
00:02:58.479 --> 00:03:01.680
There's something law enforcement officers killed or assaulted, Leoka.
00:03:02.240 --> 00:03:06.800
Um there's all sorts of different cargo theft is one that they deal with.
00:03:06.879 --> 00:03:15.919
So all sorts of different data collection um things that the FBI has happening under this one UCR umbrella.
00:03:16.159 --> 00:03:21.680
One of those, which is what people are most familiar with, is the summary reporting system or SRS.
00:03:22.159 --> 00:03:38.080
That's the seven main categories of crime that the FBI created or the IECP, the standardized crime that we're all going to count that was created in 1930 and has basically been in effect for the last hundred years.
00:03:38.319 --> 00:03:43.360
In the late 80s, the FBI said, hey, we need modernized crime statistics.
00:03:43.439 --> 00:03:45.360
And so NIBERS was created.
00:03:45.599 --> 00:03:54.479
NIBERS is instead of seven crime types, it's like almost 60 crime types, 52 to 60-ish crime types.
00:03:54.719 --> 00:04:07.199
And it allows for rather than reporting aggregated, okay, we're going to tell you how many assaults there were, how many robberies there were, how many thefts there were, we're going to, in NIBERS, they're going to give you the incidents.
00:04:07.280 --> 00:04:10.800
So they're going to say, here is a theft incident.
00:04:11.280 --> 00:04:13.280
Here's, you know, this is what was stolen.
00:04:13.439 --> 00:04:14.960
This is who the perpetrators were.
00:04:15.120 --> 00:04:22.639
This is where it took place, all of this information that you don't necessarily get from just aggregating data.
00:04:22.879 --> 00:04:33.839
And so while SRS is something, you know, a couple of hundred thousand rows takes up the entirety of SRS.
00:04:33.920 --> 00:04:39.600
You could aggregate SRS and easily open it in an Excel spreadsheet going back to 1960.
00:04:39.920 --> 00:04:44.800
NIBERS, on the other hand, is a couple of million rows per year.
00:04:45.040 --> 00:04:49.360
And so it's a lot of data that you don't necessarily get from the old system.
00:04:49.680 --> 00:04:59.360
Okay, so how give us an idea of how regularly, I know SRS, they're supposed to report numbers in Ibers, I think it's every year.
00:04:59.680 --> 00:05:03.839
How regularly do departments actually report?
00:05:04.000 --> 00:05:10.000
So is this like 80% of them report every year, 90% of them, 50%?
00:05:10.720 --> 00:05:15.199
Like how what how regular is this data actually sent to the FBI?
00:05:15.519 --> 00:05:26.319
So SRS data is typically you're looking at 95-ish percent of the country covered by an agency that reports under the summary reporting system.
00:05:26.800 --> 00:05:38.560
The new system, when it it started in the late 80s, um, by 2015, it was about a third of the country that was reporting via NIBERS.
00:05:38.720 --> 00:05:50.560
That has since increased significantly to the point where the most recent report in 2025 was 90-ish percent of the country being covered by a NIBERS reporting agency.
00:05:50.720 --> 00:05:52.800
So that's one part of uh to answer your question.
00:05:52.879 --> 00:05:54.959
The other part is how frequently do they report?
00:05:55.199 --> 00:05:59.920
Agencies report annually, they have until April of next year to report this year's data.
00:06:00.160 --> 00:06:11.920
But under NIBERS, part of what NIBERS has brought is a reconsideration, a reorganization of how data is reported to the FBI, a modernization of how data is reported to the FBI.
00:06:12.240 --> 00:06:17.519
And the result is that many agencies are able to report monthly.
00:06:17.680 --> 00:06:23.279
Some states report basically daily crime data to the FBI, other states report monthly.
00:06:23.439 --> 00:06:32.480
It's a dramatic uptick in what is coming to the FBI regularly that enables for much stronger analyses of basically what's happening.
00:06:34.319 --> 00:06:46.800
And so some of the critics of both the SRS or NIBERS is that there's some missing data, and that missing data is missing to such a point that you can't compare year to year.
00:06:48.560 --> 00:06:50.639
And I'm one, I'm wondering if that's true.
00:06:51.040 --> 00:06:54.720
Because I think that would throw off all kinds of crim research if that were true.
00:06:54.959 --> 00:07:04.160
But also I think let's say there are errors to some degree, and and there are gonna be errors in any national data system like this.
00:07:04.480 --> 00:07:08.480
If there is consistent, can't we still report and compare?
00:07:08.959 --> 00:07:11.600
Like we know nothing is gonna be a hundred percent perfect.
00:07:11.759 --> 00:07:21.439
I think this idea of expecting any system we're you know, we're gonna talk 18,000 plus police agencies, there's no way the data's gonna be perfect from every single one of them.
00:07:21.680 --> 00:07:32.959
So what do you say to people who say, well, there's errors in this from year to year, and that makes you know, comparing year to year or decade to decade, uh, it's just you can't do it.
00:07:33.759 --> 00:07:36.639
Uh so first off, there's always been errors.
00:07:36.800 --> 00:07:47.360
Um, in our our crime data from 1930 to the late 60s, early 70s was just agencies kind of reporting what they wanted.
00:07:47.519 --> 00:07:55.680
And so there's lots of evidence that our our information about what was happening in the 60s and before that is is probably understating how much crime there was.
00:07:55.920 --> 00:08:10.800
We know that there's constantly times where agencies don't report, where agencies report wrong data, um, tiny places reporting thousands upon thousands of offenses at random in a given month or two.
00:08:11.040 --> 00:08:21.120
What we know is that the the data is never perfect, but it's sort of consistently imperfect, imperfect for the most part.
00:08:21.439 --> 00:08:45.279
And you can kind of assume that because we know that the raw numbers are wrong, they're they're consistently wrong, but the the gaps are usually large enough that if you use sort of estimated language and you say it's likely down, it's um, you know, if it's up half a percentage point or down half a percent, that you say it's about even.
00:08:45.440 --> 00:08:51.440
So using language that that appreciates the sort of uncertainty in crime data, I think, is the easy workaround.
00:08:51.600 --> 00:08:54.720
And then relying on data that you know is is more reliable.
00:08:54.879 --> 00:08:57.519
We know that our murder data is is pretty reliable.
00:08:57.600 --> 00:09:03.120
It matches what comes from the CDC, it matches what's coming from other sources, it matches what's coming from cities themselves.
00:09:03.279 --> 00:09:08.240
So we have a pretty good idea that the murder data we're getting is reliable.
00:09:08.480 --> 00:09:19.840
Focusing on the things that tend to be reported more often tend to be reported uh better by police departments is a real workaround to understanding what our trends are.
00:09:20.240 --> 00:09:23.120
Oftentimes people have that sort of question in good faith.
00:09:23.200 --> 00:09:24.559
And I think that that's the best answer.
00:09:24.720 --> 00:09:31.120
People that approach it in bad faith because the crime trend that is happening is not their preference.
00:09:31.279 --> 00:09:32.799
I don't know how to handle those people.
00:09:33.679 --> 00:09:47.679
If you don't want to believe it, then you know you're welcome to sort of live out in fairyland or fantasy land or wherever it is that you want to live that where the data is, it's either the data is incorrect or more usually crime is going up.
00:09:47.840 --> 00:09:53.360
People that are not able to uh uh accept that sometimes the data is correct, even if the data is not perfect.
00:09:53.600 --> 00:09:55.840
I think that that's an unfortunate worldview.
00:09:56.000 --> 00:10:02.320
And and kind of the the goal of of my work at least is to sort of correct those those incorrect worldviews.
00:10:02.559 --> 00:10:13.120
Okay, so here's another thing I want to talk about related to we used to call the, and you may know this term, in my crim uh master's program, we called it the dark figure.
00:10:13.279 --> 00:10:21.679
And that is crime that's occurring that's not being reported and not being captured by local police departments and thus not coming in up in the SRS or in NIBERS.
00:10:21.840 --> 00:10:32.960
Can you explain to my audience what the National Crime Victimization Survey is and how that gets us a peek into that and kind of puts a little more gives us a little more understanding in the crime incident data.
00:10:33.200 --> 00:10:35.840
And CVS is run by the Bureau of Justice Statistics.
00:10:36.000 --> 00:10:49.759
It's a survey that they've been doing reasonably consistently since they've been doing it before the 90s, but the the current version is is reasonably comparable to what they've been running since sort of the early 90s.
00:10:50.000 --> 00:11:05.600
And what it does is BJS goes out and they interview roughly 250 or so thousand people nationwide, and they interview them every every every month, I believe, or every every few months.
00:11:05.679 --> 00:11:13.440
It's the census that runs the survey, and they ask, have you been the victim of a crime in the six months prior?
00:11:13.919 --> 00:11:23.840
And so they add up that, and you know, if you've got a large enough survey, you can create a an answer of how many victimizations have occurred over that span.
00:11:24.000 --> 00:11:33.279
And in general, it matches what we get from the reported data, but it is a means of evaluating what our trends are that doesn't rely on reported data.
00:11:33.360 --> 00:11:37.759
And one of the things they do in the survey is they ask, did you report this offense to the police?
00:11:37.919 --> 00:11:49.759
And they can measure through that what uh scale, what, what, what share of aggravated assaults, what share of motor vehicle thefts are being reported to police each year.
00:11:50.000 --> 00:12:00.080
And that tells us that some crimes, you know, rape and sexual assault, uh, theft, um, burglary get dramatically under-reported to police every year.
00:12:00.320 --> 00:12:12.639
About half of all assaults get reported to police each year, and about 80% or so of auto thefts get reported each year because you need the police incident number to get your insurance claim.
00:12:12.879 --> 00:12:20.320
So it gives us an understanding of what are the contours of unreported crime every year.
00:12:20.639 --> 00:12:23.360
That said, there's a bunch of limitations to NCVS.
00:12:23.519 --> 00:12:27.679
First and foremost, they're not able to obviously interview murder victims.
00:12:27.919 --> 00:12:32.000
So murder is not a category that's captured in NCVS.
00:12:32.159 --> 00:12:37.919
It's the crime with the highest societal cost, but it's not necessarily a crime that we get data from NCVS from.
00:12:38.159 --> 00:12:46.320
They're only interviewing kids older than 12 or adults, so you're not getting really juvenile crime type stuff.
00:12:46.399 --> 00:12:50.000
Uh, not a huge deal, but a thing to mention all the same.
00:12:50.240 --> 00:12:55.120
It's a survey, and people often ignore completely the margin of error.
00:12:55.360 --> 00:13:01.519
And, you know, if one year to the next is within the margin of error, it's not statistically significant.
00:13:01.759 --> 00:13:08.720
You'll oftentimes see people that want to make a point about crime going up or going down ignore that that margin of error completely.
00:13:08.879 --> 00:13:29.039
And I think the biggest issue that when people misuse it and try to use NCVS to measure or to analyze crime from one year to the next is the way that it works is if I'm interviewed in January 2026, I'm asked in the six months prior, so July 2025 to December 2025, was I the victim of a crime?
00:13:29.279 --> 00:13:32.320
Let's say I was mugged in August 2025.
00:13:32.399 --> 00:13:33.679
I'll say, Yeah, I was robbed.
00:13:33.759 --> 00:13:35.440
Um, I'll give all of the details.
00:13:35.600 --> 00:13:37.519
I did, I reported it to police.
00:13:37.679 --> 00:13:47.120
They will mark that as a 2026 offense because they're using the year of uh of interview rather than year of offense.
00:13:47.360 --> 00:14:06.879
And so by doing that, it creates a wider range basically from any anything conduct any uh offense that occurred from July 2025 through November 2026 is what makes up the 2026 NCVS NCVS.
00:14:07.120 --> 00:14:21.279
So if you want to say, you know, maybe there was a change in crime in the second half of 2025 that got captured in the interviews or in 2024 or whatever, and so it creates a lot more uncertainty in terms of year-to-year change.
00:14:21.440 --> 00:14:35.440
It does a great job of capturing the contours of crime, it does a great job of capturing what doesn't get reported, but people want to use it like it's analogous to SRS and NIBERS as far as this is what was reported in a given year to have occurred in that year.
00:14:35.600 --> 00:14:37.200
And it's really not apples to apples.
00:14:37.600 --> 00:14:43.840
Another criticism of both the left and the right that about crime, that seems one thing they're united on.
00:14:43.919 --> 00:14:50.080
They don't believe the crime numbers, the violence numbers one way or the other, um, that it fits their political allegiance.
00:14:50.320 --> 00:14:59.440
So there's a lot of pressure on police departments, especially municipal police department, large city police departments, to when they release their annual numbers.
00:14:59.679 --> 00:15:06.879
Crime went up, it's they're horrible, it's time to get rid of the mayor and the police chief, crime went down, oh my god, it's the best police department ever.
00:15:07.039 --> 00:15:15.519
But the local media puts pressure on these police departments, then the national media does these comparisons across big cities or across states.
00:15:15.759 --> 00:15:32.480
So, what do you say to people who say, well, there's all kinds of pressure on these police departments to what we would call in the criminal kill crime or doctor the numbers, and that and they're doctoring the numbers is skewing UCR or the SRS or NIBERS.
00:15:32.559 --> 00:15:34.240
What do you what's your take on that?
00:15:34.480 --> 00:15:45.679
I'll often get when crime is going up, that those numbers are being doctored so that police departments can improve their budgets and you know get more money because crime's out of control.
00:15:46.159 --> 00:15:54.879
And when they're going down, they're being doctored because the police departments want, you know, they want to be able to say that they're fighting crime effectively.
00:15:55.120 --> 00:16:13.360
And it, you know, it fits into this, I think, in a lot of things in American society right now, and probably world society as well, that there are issues that require being careful, being, you know, looking at it, understanding the data, understanding the the uncertainty in the data, and understanding the potential misuses of the data.
00:16:13.519 --> 00:16:19.440
But there's a tendency to just not believe information that comes, even if the information is correct.
00:16:19.679 --> 00:16:21.600
I I see that very frequently.
00:16:21.679 --> 00:16:24.799
I I don't find it it necessarily compelling.
00:16:24.960 --> 00:16:29.039
And I think that there are ways to use other sources.
00:16:29.360 --> 00:16:33.840
So if you you know you have a problem with FBI data, you can use an agency's data.
00:16:33.919 --> 00:16:49.840
If you have a problem with an agency's reported data, you can use sometimes if they have it publicly available, you can use their calls for service data, which it's a lot harder to doctor a half million calls for service that come in through a dispatcher versus what the police department actually reports itself.
00:16:50.080 --> 00:16:56.960
You don't like the police department at all, you can use the gun violence archive and evaluate gun violence through a data source like that.
00:16:57.279 --> 00:17:01.600
So there are ways to do um different data sets.
00:17:01.840 --> 00:17:08.480
DC is a good example because DC has been sort of in the news for the unreliability of their crime data.
00:17:08.720 --> 00:17:11.119
And you can look at the gun violence archive.
00:17:11.200 --> 00:17:14.960
The DC Public Health Department has a gun violence dashboard.
00:17:15.119 --> 00:17:27.200
You can compare the data coming from the FBI with the data coming from the city itself, and you can see, okay, maybe robbery and aggravated assault aren't really accurate.
00:17:27.359 --> 00:17:30.400
There's there's questions about the reliability of that data.
00:17:30.720 --> 00:17:41.200
And you can then look at something like the FBI data, their carjacking data, and and their murder data, and say, okay, these look a lot more comparable and this is more reliable.
00:17:41.359 --> 00:17:48.000
So there's all sorts of ways that you can curiously investigate data without sort of out of hand saying, no, this is wrong.
00:17:48.079 --> 00:17:48.640
I don't believe it.
00:17:48.960 --> 00:17:51.839
So you and I share a strong dislike.
00:17:52.000 --> 00:18:01.119
I've looked at your Substack for hyperbolic or make-believe analyses uh published by these what I call real estate adjacent websites.
00:18:01.920 --> 00:18:05.200
For instance, I see you referring your substack to neighborhood scour.
00:18:06.880 --> 00:18:12.480
So what are your biggest problems with these types of analysis these websites do?
00:18:12.799 --> 00:18:21.920
Because I know I get asked all the time, and ever since I started grad school, really from friends, from family, from friends of friends of friends.
00:18:22.160 --> 00:18:30.480
They see some site and they're not sure if they should move into this neighborhood because some site like neighborhood scout says that's a really dangerous neighborhood.
00:18:31.039 --> 00:18:44.240
But you could never you'd have to really dig in to try to figure out what because I was always the maze in first blush looking at them, wow, how did they get access to all that data to come up with that analysis?
00:18:44.799 --> 00:18:46.000
That's kind of amazing.
00:18:46.400 --> 00:18:49.440
So I just wanted to get your uh get your take on this.
00:18:50.559 --> 00:18:54.640
Yeah, a lot of these sites do that type of analysis.
00:18:54.799 --> 00:19:09.759
It's unfortunate, it's oftentimes taking analysis uh or taking data from UCR and saying, okay, there were a thousand robberies in this city last year.
00:19:10.000 --> 00:19:20.559
Let's basically use our own proprietary algorithm and say that okay, the you know, the the poor areas are gonna have 70% of the robberies.
00:19:20.720 --> 00:19:29.200
So we'll say that this neighborhood had, you know, 700 robberies, and this area had you know 10%, probably.
00:19:29.359 --> 00:19:32.079
So then this and there's no way to really track it.
00:19:32.240 --> 00:19:37.519
And if a city doesn't report data to the FBI, they will make up a number.
00:19:37.839 --> 00:19:42.720
And they won't they'll they won't be transparent about when the data came from.
00:19:42.880 --> 00:19:50.720
And oftentimes there'll be splashy graphics that people want to say, hey, this is the worst place in America, or this is the best place in America.
00:19:50.960 --> 00:19:53.519
That's a story that people sometimes want to tell.
00:19:53.759 --> 00:20:02.559
And just because the data is very clearly should not be compared, um, we shouldn't be ranking places using the FBI data.
00:20:02.640 --> 00:20:10.640
Uh, that doesn't stop organizations that don't care about the the inefficiencies or the uncertainties of the data.
00:20:10.720 --> 00:20:12.640
Um, they're just gonna publish it anyway.
00:20:13.119 --> 00:20:22.960
So, in general, what are your feelings on how local media typically reports on crime and violence and crime data generally?
00:20:23.359 --> 00:20:34.079
And I'm not talking the one-offs because there's you know, we feature a lot of journalists on our podcast doing great reporting, but I'm talking about the daily news coverage around the country.
00:20:34.559 --> 00:20:45.920
Because you know, we know that for the last few decades at least, the fear of crime has always far outpaced the likelihood of people being victimized.
00:20:46.160 --> 00:20:50.640
So I just wanted to get your feelings being an expert in the data, how you felt it was being reported on.
00:20:50.960 --> 00:20:51.920
It's really hard.
00:20:52.079 --> 00:21:02.319
It the the challenge that I know media faces a lot is that it's easy to report things that happen, it's not easy to report things that don't happen.
00:21:02.640 --> 00:21:06.160
So there's that saying that the media doesn't cover the planes that land.
00:21:06.319 --> 00:21:17.920
And I think that that's apt for crime data because you'll never see a story that, hey, there was no carjacking last week, or um, Walgreens didn't report any shopliftings.
00:21:18.240 --> 00:21:30.799
And so when people un try to understand the their environment through anecdotes, it leads to a belief that there's always things are always getting worse.
00:21:30.880 --> 00:21:33.920
Um, and that's really hard, I think, for media to do.
00:21:34.160 --> 00:21:41.119
And this is coming from someone that thinks it's really important that when crime occurs, like it's important that we understand our crime trends.
00:21:41.200 --> 00:21:49.680
So when things are getting worse, that's an important story for local media to be telling, but they really struggle with things are getting better.
00:21:50.000 --> 00:21:54.400
And and I wish I had a better answer for how to solve that problem because I I don't know what it is.
00:21:54.559 --> 00:21:59.759
There certainly are a lot of journalists out there that are very data literate and try to address that issue.
00:22:00.160 --> 00:22:10.319
And then there's journalists out there that aren't data literate and don't try to address, you know, what what is the overall context of with which this crime is occurring.
00:22:10.720 --> 00:22:15.839
Back many years, God, 10 years ago, maybe even more, 12 years ago, we did an analysis.
00:22:16.000 --> 00:22:23.039
We studied one year of sexual assault and and domestic violence reporting in the Chicago Tribune and Chicago Sundi.
00:22:23.359 --> 00:22:26.160
And we basically figured out that.
00:22:33.359 --> 00:22:36.960
Domestic violence, the only thing that made the news were hyper-violent cases.
00:22:38.480 --> 00:22:42.559
And for sexual assault, it was almost it was like 90% stranger assaults.
00:22:42.960 --> 00:22:50.400
The media, when we released it, we got a I got a lot of criticism from journalists saying, well, how do we report on sexual assault?
00:22:51.119 --> 00:22:56.160
The the victims, we can't use the victims' names and they hardly talk to us.
00:22:57.200 --> 00:23:05.680
And I'm like, well, the and me thinking being a big data guy, it's like it's not the details of an individual case that matter.
00:23:06.240 --> 00:23:10.960
Like if you if there's an assault at DePaul University, it isn't about Jane Smith that gets assaulted.
00:23:11.119 --> 00:23:17.759
It's about whether there's a trend going up or down, and what's the trend in the university responding to those things.
00:23:18.160 --> 00:23:22.799
Yeah, it it and it's hard to do on a daily basis and do it coher, you know.
00:23:23.279 --> 00:23:37.839
We talk a lot on this podcast about I think unfortunately, and it's doing us a great disservice, that crime and violence data is exploited for political purposes, for agenda purposes.
00:23:38.160 --> 00:23:42.000
I think the media exploits it for clicks.
00:23:42.240 --> 00:23:45.920
I think polit politicians exploit it for their own agenda.
00:23:46.160 --> 00:23:56.079
I know even in like the and you've talked talked about this on your uh Substack, crime is going down, has gone down, violent crime has gone down over the last couple years.
00:23:56.319 --> 00:24:10.799
While there's a report of a really bad violent incident in a city, and the politicians who are not in office or not in the powerful office they want to be, will criticize and say, Well, crime can't be going down.
00:24:11.359 --> 00:24:12.160
It's impossible.
00:24:12.319 --> 00:24:13.440
Look what just happened.
00:24:13.680 --> 00:24:17.839
And I try to tell people like bad things still happen as crime goes down.
00:24:18.000 --> 00:24:20.480
It just means less of the bad things are happening.
00:24:20.640 --> 00:24:22.640
It doesn't mean they're completely eliminated.
00:24:22.799 --> 00:24:34.160
So I just wanted to get your take on whether or not you think those things, like crime data, violent crime and crime and violence data is being gets exploited to the detriment of the public.
00:24:34.640 --> 00:24:35.599
I mean, entirely.
00:24:35.759 --> 00:24:55.440
And you see this in especially in politics, where uh I did a uh congressional hearing in in North Carolina last year, and they had, you know, the parents of people that had been murdered and the siblings and a police officer that had been shot and these terrible incidents.
00:24:55.839 --> 00:25:03.119
And then, you know, they had all of these people interviewing, and then I came up as the witness that basically said, like, these are horrible tragedies, it's awful.
00:25:03.279 --> 00:25:06.319
I can't imagine what all of this is going down.
00:25:06.480 --> 00:25:10.240
And if you can't accept that it's getting better, then you can never learn why.
00:25:10.319 --> 00:25:14.079
And if you never learn why, then you can never make things actually better.
00:25:14.240 --> 00:25:28.720
And I mean, I got I got a ton of flack from most of the Congress people out there that were like, Well, you know, you're, you know, you're lying, or you're uh you're you're, you know, you didn't consider X moving the goalposts and things like that.
00:25:28.880 --> 00:25:46.880
And it's just, it's not an easy thing, I think, to talk about, because if we talk about how murders probably gonna have we're probably gonna have the fewest murders ever recorded in 2026, like by far, the lowest murder rate ever recorded, you're still talking about 12,000 murders.
00:25:47.119 --> 00:25:58.240
And so if you want to find 12,000 tragedies or 12,000 individual stories of people who lost their lives when they shouldn't have, like that's terrible.
00:25:58.400 --> 00:26:03.680
And absolutely we should always acknowledge that that that the issues remain unsolved.
00:26:04.160 --> 00:26:07.920
But that doesn't mean that the issue isn't getting better.
00:26:08.079 --> 00:26:16.640
And it doesn't mean that it that it would be a wasted opportunity to not know why it's getting better because not learning why is the worst thing we could possibly do.
00:26:16.960 --> 00:26:23.039
Especially on this issue, people don't care why, they just want it going in whatever direction they want it going.
00:26:23.759 --> 00:26:24.480
That's it.
00:26:24.720 --> 00:26:35.359
And for a lot of people, a lot of politicians, especially the media to some extent, it can't ever be going down when their person is not in power.
00:26:35.759 --> 00:26:38.559
Then everyone's lying if it's going down.
00:26:38.720 --> 00:26:44.400
And then once they get power or their people get power, then it's starting to go down and it's okay.
00:26:44.559 --> 00:26:48.960
Then it's directly related to their policies, which just annoys the hell out of me.
00:26:49.119 --> 00:26:55.599
It's like you look, if you're talking about the fewest murders in America, most likely in 2026, maybe for forever, right?
00:26:56.000 --> 00:27:01.200
And people are like, you have all these local politicians and local police chiefs and editorial pages.
00:27:01.279 --> 00:27:03.119
Look, how good are police departments doing?
00:27:03.279 --> 00:27:06.240
And there's no doubt the police department has some impact on that.
00:27:06.319 --> 00:27:08.960
There's unit's undoubtedly undoubtedly, right?
00:27:09.200 --> 00:27:12.640
But at the same point, we're a country of 350 million people.
00:27:12.799 --> 00:27:28.319
And if crime is going down generally all over the country, all at the same time, we might want to look at something other than just the policies and practices of 18,000 separate police agencies as for why the crime is crime numbers are going down.
00:27:28.400 --> 00:27:34.640
It doesn't mean the police don't have a role, but we may look try to find a solution that is more universal throughout our country.
00:27:34.799 --> 00:27:42.240
There probably are disparate policies that are probably opposite of each other in two different cities across the country that have crime drops that are very similar.
00:27:42.400 --> 00:27:45.759
What troubles you the most about how America deals with crime data?
00:27:46.000 --> 00:28:01.920
I mean, I think it it's very similar to your your previous statement that we have a tendency to that whatever was happening at the moment that the trend changed is what we're going to credit or blame for the change.
00:28:02.160 --> 00:28:10.000
And so, you know, sometimes it makes sense that we did X gun violence reduction policy, and hey, gun violence went down.
00:28:10.079 --> 00:28:10.559
That's great.
00:28:10.640 --> 00:28:12.960
That's that's pretty good evidence that this thing worked.
00:28:13.039 --> 00:28:15.920
And if we study it, we can find that you know there's a correlation there.
00:28:16.079 --> 00:28:21.119
Other times it's like I was hopping on one leg and then crime started falling.
00:28:21.200 --> 00:28:26.240
And so if we keep hopping on one leg, then that's gonna lead to fewer crimes.
00:28:26.319 --> 00:28:40.640
And and so sometimes I'm I'm looking, I have a piece coming out soon on looking just at at San Francisco vehicle burglaries and auto thefts and the degree to which technology influenced those those things just plunging.
00:28:40.960 --> 00:28:55.839
And it it makes sense, but then you you one, you look at it, the data, and you see that hey, they went from 30,000 vehicle burglaries in a year to 5,000 vehicle burglaries a year in the span of like four years.
00:28:56.000 --> 00:29:02.160
And all of that started before they, you know, before their flock cons before their flock um contract.
00:29:02.240 --> 00:29:04.079
It started before they had drones.
00:29:04.319 --> 00:29:08.079
Auto thefts again started falling in in the beginning of 2023.
00:29:08.160 --> 00:29:11.359
And it wasn't until mid-2024 that they had Flock.
00:29:11.599 --> 00:29:21.839
And so if you want to credit the technology, you have to explain why these huge declines started even before that and why they match national standards.
00:29:22.000 --> 00:29:32.799
And then, you know, how drones are gonna actually stop uh vehicle burglaries, or how drones are gonna stop, or how uh license plate readers are gonna stop vehicle burglaries when the car doesn't move.
00:29:32.960 --> 00:29:33.759
So good questions.
00:29:33.839 --> 00:29:41.200
Maybe there's a role there, but I think that it's a lot harder to say than just this is a technology we implemented in the last three years.
00:29:41.359 --> 00:29:43.599
It must be responsible because we're seeing a drop.
00:29:43.839 --> 00:29:59.440
And so I wish that we had much more academic research into why the trends we are seeing are happening, and that level research is happening, but maybe not to the degree that the the potential to learn from this historic time is warranting.
00:29:59.759 --> 00:30:03.200
Yeah, I used to have well, I have a friend who's now retired.
00:30:03.279 --> 00:30:08.319
He used to be a district commander in Chicago over the area where Wrigley Field is, where the Cubs play.
00:30:08.400 --> 00:30:10.960
It's a very popular lots of bars and restaurants.
00:30:11.279 --> 00:30:14.319
Gets crowded on the weekends, super crowded on game days.
00:30:14.480 --> 00:30:20.160
And he, when he got there, they had a lot of problems with bar, the bar fights, and violence around the bars and alcohol.
00:30:20.240 --> 00:30:30.319
So he created a special unit and worked with the bars and had a huge presence on those on the strip after game before games, after games on the weekend nights.
00:30:30.480 --> 00:30:36.960
And lo and you know, they had the Com Stamp meetings, and he got a big applause because his numbers plummeted during the Cub season.
00:30:37.200 --> 00:30:42.960
And then he goes to one and they're like, Well, why did your you know your uh car thefts are up and car burglars are up?
00:30:43.119 --> 00:30:45.519
What the hell are you why aren't you dealing with that?
00:30:45.759 --> 00:30:52.079
He's like, Because the other guy had a special unit working on that, but this this other thing went up, the bar issue went up.
00:30:52.160 --> 00:30:58.799
So I took that special unit, moved to the bar thing, which was more violent than the car break-ins, and that's those numbers have plummeted.
00:30:58.960 --> 00:31:01.200
Well, you got to do something about those car things.
00:31:01.279 --> 00:31:01.920
He goes, I am.
00:31:02.000 --> 00:31:05.599
I'm gonna take the unit and I'm gonna move it over to deal with the car burglaries.
00:31:05.759 --> 00:31:07.920
Okay, well, you know, we'll see you next month.
00:31:08.000 --> 00:31:09.519
I'm like, you will, and you know what's gonna happen?
00:31:09.680 --> 00:31:13.759
That's gonna go down, and my numbers at the bars and the violence is gonna go up.
00:31:13.839 --> 00:31:15.519
And you're gonna tell me to do something about it.
00:31:15.680 --> 00:31:18.640
I'm gonna take the guys that are on the cars and move them back to the bars.
00:31:18.799 --> 00:31:21.680
And then he goes, and he would tell me, he goes, No, this is whack-a-mole.
00:31:21.759 --> 00:31:23.200
It's my turn they go get yelled at.
00:31:23.279 --> 00:31:24.720
And he goes, It's whack-a-mole.
00:31:24.799 --> 00:31:28.240
Wherever we have the resources we put it, the thing gets addressed somehow.
00:31:28.319 --> 00:31:34.400
But the other things in the city that are in my area that aren't getting addressed because I have them concentrating on the other thing, it's happening.
00:31:34.480 --> 00:31:35.440
It's whack-a-mole.
00:31:35.519 --> 00:31:38.960
So I just thought that has just stayed with me from a police and perceptive.
00:31:39.200 --> 00:31:39.440
Okay.
00:31:39.599 --> 00:31:42.000
Jeff Asher, thank you so much for jumping in with us.
00:31:42.079 --> 00:31:43.119
We really appreciate it.
00:31:43.440 --> 00:31:44.079
My pleasure.
00:31:44.240 --> 00:31:47.440
You know, that was Jeff Asher from AH Datalytics.
00:31:49.119 --> 00:31:54.880
When the only facts that can exist are the ones that reinforce our political views, the world is in grave danger.
00:31:55.119 --> 00:31:58.319
We have been at this point for many years on our criminal justice system.
00:31:58.799 --> 00:32:06.240
The rabid belief that crime and violence go up or down solely because of the actions of local police and politicians is crazy.
00:32:07.359 --> 00:32:18.480
Especially given the fact that national patterns across dozens, if not hundreds, of cities cannot be based solely on the policies and practices of individual police agencies across all those jurisdictions.
00:32:18.960 --> 00:32:25.440
Arguing we need better crime and violence data and transparency from justice agencies is an argument CJP was built on.
00:32:25.680 --> 00:32:26.559
We're there with you.
00:32:30.640 --> 00:32:37.519
That was That was Jeff Asher from AH Datalytics.
00:32:39.200 --> 00:32:44.960
When the only facts that can exist are the ones that reinforce our political views, the world is in grave danger.
00:32:45.759 --> 00:32:48.960
We have been at this point for many years on our criminal justice system.
00:32:49.680 --> 00:32:57.279
The rabid belief that crime and violence go up or down solely because of the actions of local police and politicians is crazy.
00:32:58.240 --> 00:33:09.599
Especially given the fact that national patterns across dozens, if not hundreds, of cities cannot be based solely on the policies and practices of individual police agencies across all those jurisdictions.
00:33:11.039 --> 00:33:14.799
Arguing that we need better crime and violence and data because of that.
00:33:43.119 --> 00:33:50.400
But arguing we cannot trust any of the data because it doesn't reinforce our political agenda is dangerous and bad for communities.
00:33:51.039 --> 00:33:54.559
Thanks for listening, and as always, please share with your friends and contacts.
00:33:54.720 --> 00:33:56.240
I'll be back with you next week.
00:00:00.239 --> 00:00:04.240
Halfway through the year, I can tell you that our violent crime is down 19%.
00:00:04.879 --> 00:00:08.160
Our violent and property crime is down a whopping 22%.
00:00:09.119 --> 00:00:14.400
And our homicides are down over 63%, which is truly amazing.
00:00:14.640 --> 00:00:18.079
We could not accomplish this without your help and your support.
00:00:18.239 --> 00:00:26.160
Our detectives and officers are out there doing great, amazing work, and you all are sharing things with us and reporting crime and suspicious activity.
00:00:26.399 --> 00:00:29.920
Thank you for helping us make Tampa safer together.
00:00:30.320 --> 00:00:30.879
Hello all.
00:00:30.960 --> 00:00:33.039
Welcome to the Chicago Justice Podcast.
00:00:33.119 --> 00:00:34.640
I'm your host, Tracy Siska.
00:00:34.880 --> 00:00:44.560
I'm also Executive Director of the Chicago Justice Podcast and adject instructor in the Department of Justice, Law, Criminology, and Security at American University.
00:00:46.320 --> 00:00:52.399
The exploitation of crime and crime and violence data in America has reached epic proportions.
00:00:53.679 --> 00:01:02.079
Social media and the rapid spread of misinformation and disinformation have exponentially increased the impact of the misuse of crime data on communities.
00:01:03.200 --> 00:01:09.200
Pair this with the rapid rise of partisanship and extremism, and you have a recipe for disaster in our criminal justice system.
00:01:10.959 --> 00:01:23.040
One of the consequences of hyperpartisanship and the rise of extremism on the right is that we, as a side, have moved away from a climate of searching for and expanding best practices within our justice system.
00:01:24.079 --> 00:01:30.400
Instead, we have adopted an attitude of absolute denialism when confronted with data that challenges political views.
00:01:31.359 --> 00:01:42.319
This has led us to a point where, as crime goes down around the country, unethical and immoral politicians exploit individual violent incidents to challenge realities that cities across the country are experiencing.
00:01:42.959 --> 00:01:49.120
The media laps it up because they benefit as much as anyone from the political expectation of crime and violence.
00:01:49.920 --> 00:01:52.319
They're also a major exploiter of crime and violence.
00:01:53.040 --> 00:02:00.560
To dig into all these issues, we sit down with Jeff Asher, a nationally recognized crime analyst and co-founder of AH Datalytics.
00:02:01.280 --> 00:02:07.680
He writes a weekly newsletter and hosts the Jeffalytics Podcast, a show about measuring crime and communicating trends.
00:02:09.199 --> 00:02:10.319
Let's get to our discussion.
00:02:10.960 --> 00:02:12.800
Jeff Asher, thank you so much for joining us.
00:02:12.879 --> 00:02:13.680
We really appreciate it.
00:02:14.240 --> 00:02:14.960
Thanks for having me.
00:02:15.280 --> 00:02:31.599
So give my audience a brief explanation of the Uniform Crime Reports and how the national incident-based reporting system is a little more detailed than what people or police, I should say, report to the feds typically.
00:02:31.759 --> 00:02:33.599
And what are the differences between the two?
00:02:34.159 --> 00:02:44.719
So so the Uniform Crime Report was created in the 1930s, first by the International Association of the Chiefs of Police, and then FBI took it over a couple of years later.
00:02:44.960 --> 00:02:50.479
Um it is it NIBERS is a part of UCR.
00:02:50.560 --> 00:02:55.680
So UCR is sort of the big picture umbrella of crime data reporting.
00:02:55.919 --> 00:02:58.319
Um we get hate crime data through that.
00:02:58.479 --> 00:03:01.680
There's something law enforcement officers killed or assaulted, Leoka.
00:03:02.240 --> 00:03:06.800
Um there's all sorts of different cargo theft is one that they deal with.
00:03:06.879 --> 00:03:15.919
So all sorts of different data collection um things that the FBI has happening under this one UCR umbrella.
00:03:16.159 --> 00:03:21.680
One of those, which is what people are most familiar with, is the summary reporting system or SRS.
00:03:22.159 --> 00:03:38.080
That's the seven main categories of crime that the FBI created or the IECP, the standardized crime that we're all going to count that was created in 1930 and has basically been in effect for the last hundred years.
00:03:38.319 --> 00:03:43.360
In the late 80s, the FBI said, hey, we need modernized crime statistics.
00:03:43.439 --> 00:03:45.360
And so NIBERS was created.
00:03:45.599 --> 00:03:54.479
NIBERS is instead of seven crime types, it's like almost 60 crime types, 52 to 60-ish crime types.
00:03:54.719 --> 00:04:07.199
And it allows for rather than reporting aggregated, okay, we're going to tell you how many assaults there were, how many robberies there were, how many thefts there were, we're going to, in NIBERS, they're going to give you the incidents.
00:04:07.280 --> 00:04:10.800
So they're going to say, here is a theft incident.
00:04:11.280 --> 00:04:13.280
Here's, you know, this is what was stolen.
00:04:13.439 --> 00:04:14.960
This is who the perpetrators were.
00:04:15.120 --> 00:04:22.639
This is where it took place, all of this information that you don't necessarily get from just aggregating data.
00:04:22.879 --> 00:04:33.839
And so while SRS is something, you know, a couple of hundred thousand rows takes up the entirety of SRS.
00:04:33.920 --> 00:04:39.600
You could aggregate SRS and easily open it in an Excel spreadsheet going back to 1960.
00:04:39.920 --> 00:04:44.800
NIBERS, on the other hand, is a couple of million rows per year.
00:04:45.040 --> 00:04:49.360
And so it's a lot of data that you don't necessarily get from the old system.
00:04:49.680 --> 00:04:59.360
Okay, so how give us an idea of how regularly, I know SRS, they're supposed to report numbers in Ibers, I think it's every year.
00:04:59.680 --> 00:05:03.839
How regularly do departments actually report?
00:05:04.000 --> 00:05:10.000
So is this like 80% of them report every year, 90% of them, 50%?
00:05:10.720 --> 00:05:15.199
Like how what how regular is this data actually sent to the FBI?
00:05:15.519 --> 00:05:26.319
So SRS data is typically you're looking at 95-ish percent of the country covered by an agency that reports under the summary reporting system.
00:05:26.800 --> 00:05:38.560
The new system, when it it started in the late 80s, um, by 2015, it was about a third of the country that was reporting via NIBERS.
00:05:38.720 --> 00:05:50.560
That has since increased significantly to the point where the most recent report in 2025 was 90-ish percent of the country being covered by a NIBERS reporting agency.
00:05:50.720 --> 00:05:52.800
So that's one part of uh to answer your question.
00:05:52.879 --> 00:05:54.959
The other part is how frequently do they report?
00:05:55.199 --> 00:05:59.920
Agencies report annually, they have until April of next year to report this year's data.
00:06:00.160 --> 00:06:11.920
But under NIBERS, part of what NIBERS has brought is a reconsideration, a reorganization of how data is reported to the FBI, a modernization of how data is reported to the FBI.
00:06:12.240 --> 00:06:17.519
And the result is that many agencies are able to report monthly.
00:06:17.680 --> 00:06:23.279
Some states report basically daily crime data to the FBI, other states report monthly.
00:06:23.439 --> 00:06:32.480
It's a dramatic uptick in what is coming to the FBI regularly that enables for much stronger analyses of basically what's happening.
00:06:34.319 --> 00:06:46.800
And so some of the critics of both the SRS or NIBERS is that there's some missing data, and that missing data is missing to such a point that you can't compare year to year.
00:06:48.560 --> 00:06:50.639
And I'm one, I'm wondering if that's true.
00:06:51.040 --> 00:06:54.720
Because I think that would throw off all kinds of crim research if that were true.
00:06:54.959 --> 00:07:04.160
But also I think let's say there are errors to some degree, and and there are gonna be errors in any national data system like this.
00:07:04.480 --> 00:07:08.480
If there is consistent, can't we still report and compare?
00:07:08.959 --> 00:07:11.600
Like we know nothing is gonna be a hundred percent perfect.
00:07:11.759 --> 00:07:21.439
I think this idea of expecting any system we're you know, we're gonna talk 18,000 plus police agencies, there's no way the data's gonna be perfect from every single one of them.
00:07:21.680 --> 00:07:32.959
So what do you say to people who say, well, there's errors in this from year to year, and that makes you know, comparing year to year or decade to decade, uh, it's just you can't do it.
00:07:33.759 --> 00:07:36.639
Uh so first off, there's always been errors.
00:07:36.800 --> 00:07:47.360
Um, in our our crime data from 1930 to the late 60s, early 70s was just agencies kind of reporting what they wanted.
00:07:47.519 --> 00:07:55.680
And so there's lots of evidence that our our information about what was happening in the 60s and before that is is probably understating how much crime there was.
00:07:55.920 --> 00:08:10.800
We know that there's constantly times where agencies don't report, where agencies report wrong data, um, tiny places reporting thousands upon thousands of offenses at random in a given month or two.
00:08:11.040 --> 00:08:21.120
What we know is that the the data is never perfect, but it's sort of consistently imperfect, imperfect for the most part.
00:08:21.439 --> 00:08:45.279
And you can kind of assume that because we know that the raw numbers are wrong, they're they're consistently wrong, but the the gaps are usually large enough that if you use sort of estimated language and you say it's likely down, it's um, you know, if it's up half a percentage point or down half a percent, that you say it's about even.
00:08:45.440 --> 00:08:51.440
So using language that that appreciates the sort of uncertainty in crime data, I think, is the easy workaround.
00:08:51.600 --> 00:08:54.720
And then relying on data that you know is is more reliable.
00:08:54.879 --> 00:08:57.519
We know that our murder data is is pretty reliable.
00:08:57.600 --> 00:09:03.120
It matches what comes from the CDC, it matches what's coming from other sources, it matches what's coming from cities themselves.
00:09:03.279 --> 00:09:08.240
So we have a pretty good idea that the murder data we're getting is reliable.
00:09:08.480 --> 00:09:19.840
Focusing on the things that tend to be reported more often tend to be reported uh better by police departments is a real workaround to understanding what our trends are.
00:09:20.240 --> 00:09:23.120
Oftentimes people have that sort of question in good faith.
00:09:23.200 --> 00:09:24.559
And I think that that's the best answer.
00:09:24.720 --> 00:09:31.120
People that approach it in bad faith because the crime trend that is happening is not their preference.
00:09:31.279 --> 00:09:32.799
I don't know how to handle those people.
00:09:33.679 --> 00:09:47.679
If you don't want to believe it, then you know you're welcome to sort of live out in fairyland or fantasy land or wherever it is that you want to live that where the data is, it's either the data is incorrect or more usually crime is going up.
00:09:47.840 --> 00:09:53.360
People that are not able to uh uh accept that sometimes the data is correct, even if the data is not perfect.
00:09:53.600 --> 00:09:55.840
I think that that's an unfortunate worldview.
00:09:56.000 --> 00:10:02.320
And and kind of the the goal of of my work at least is to sort of correct those those incorrect worldviews.
00:10:02.559 --> 00:10:13.120
Okay, so here's another thing I want to talk about related to we used to call the, and you may know this term, in my crim uh master's program, we called it the dark figure.
00:10:13.279 --> 00:10:21.679
And that is crime that's occurring that's not being reported and not being captured by local police departments and thus not coming in up in the SRS or in NIBERS.
00:10:21.840 --> 00:10:32.960
Can you explain to my audience what the National Crime Victimization Survey is and how that gets us a peek into that and kind of puts a little more gives us a little more understanding in the crime incident data.
00:10:33.200 --> 00:10:35.840
And CVS is run by the Bureau of Justice Statistics.
00:10:36.000 --> 00:10:49.759
It's a survey that they've been doing reasonably consistently since they've been doing it before the 90s, but the the current version is is reasonably comparable to what they've been running since sort of the early 90s.
00:10:50.000 --> 00:11:05.600
And what it does is BJS goes out and they interview roughly 250 or so thousand people nationwide, and they interview them every every every month, I believe, or every every few months.
00:11:05.679 --> 00:11:13.440
It's the census that runs the survey, and they ask, have you been the victim of a crime in the six months prior?
00:11:13.919 --> 00:11:23.840
And so they add up that, and you know, if you've got a large enough survey, you can create a an answer of how many victimizations have occurred over that span.
00:11:24.000 --> 00:11:33.279
And in general, it matches what we get from the reported data, but it is a means of evaluating what our trends are that doesn't rely on reported data.
00:11:33.360 --> 00:11:37.759
And one of the things they do in the survey is they ask, did you report this offense to the police?
00:11:37.919 --> 00:11:49.759
And they can measure through that what uh scale, what, what, what share of aggravated assaults, what share of motor vehicle thefts are being reported to police each year.
00:11:50.000 --> 00:12:00.080
And that tells us that some crimes, you know, rape and sexual assault, uh, theft, um, burglary get dramatically under-reported to police every year.
00:12:00.320 --> 00:12:12.639
About half of all assaults get reported to police each year, and about 80% or so of auto thefts get reported each year because you need the police incident number to get your insurance claim.
00:12:12.879 --> 00:12:20.320
So it gives us an understanding of what are the contours of unreported crime every year.
00:12:20.639 --> 00:12:23.360
That said, there's a bunch of limitations to NCVS.
00:12:23.519 --> 00:12:27.679
First and foremost, they're not able to obviously interview murder victims.
00:12:27.919 --> 00:12:32.000
So murder is not a category that's captured in NCVS.
00:12:32.159 --> 00:12:37.919
It's the crime with the highest societal cost, but it's not necessarily a crime that we get data from NCVS from.
00:12:38.159 --> 00:12:46.320
They're only interviewing kids older than 12 or adults, so you're not getting really juvenile crime type stuff.
00:12:46.399 --> 00:12:50.000
Uh, not a huge deal, but a thing to mention all the same.
00:12:50.240 --> 00:12:55.120
It's a survey, and people often ignore completely the margin of error.
00:12:55.360 --> 00:13:01.519
And, you know, if one year to the next is within the margin of error, it's not statistically significant.
00:13:01.759 --> 00:13:08.720
You'll oftentimes see people that want to make a point about crime going up or going down ignore that that margin of error completely.
00:13:08.879 --> 00:13:29.039
And I think the biggest issue that when people misuse it and try to use NCVS to measure or to analyze crime from one year to the next is the way that it works is if I'm interviewed in January 2026, I'm asked in the six months prior, so July 2025 to December 2025, was I the victim of a crime?
00:13:29.279 --> 00:13:32.320
Let's say I was mugged in August 2025.
00:13:32.399 --> 00:13:33.679
I'll say, Yeah, I was robbed.
00:13:33.759 --> 00:13:35.440
Um, I'll give all of the details.
00:13:35.600 --> 00:13:37.519
I did, I reported it to police.
00:13:37.679 --> 00:13:47.120
They will mark that as a 2026 offense because they're using the year of uh of interview rather than year of offense.
00:13:47.360 --> 00:14:06.879
And so by doing that, it creates a wider range basically from any anything conduct any uh offense that occurred from July 2025 through November 2026 is what makes up the 2026 NCVS NCVS.
00:14:07.120 --> 00:14:21.279
So if you want to say, you know, maybe there was a change in crime in the second half of 2025 that got captured in the interviews or in 2024 or whatever, and so it creates a lot more uncertainty in terms of year-to-year change.
00:14:21.440 --> 00:14:35.440
It does a great job of capturing the contours of crime, it does a great job of capturing what doesn't get reported, but people want to use it like it's analogous to SRS and NIBERS as far as this is what was reported in a given year to have occurred in that year.
00:14:35.600 --> 00:14:37.200
And it's really not apples to apples.
00:14:37.600 --> 00:14:43.840
Another criticism of both the left and the right that about crime, that seems one thing they're united on.
00:14:43.919 --> 00:14:50.080
They don't believe the crime numbers, the violence numbers one way or the other, um, that it fits their political allegiance.
00:14:50.320 --> 00:14:59.440
So there's a lot of pressure on police departments, especially municipal police department, large city police departments, to when they release their annual numbers.
00:14:59.679 --> 00:15:06.879
Crime went up, it's they're horrible, it's time to get rid of the mayor and the police chief, crime went down, oh my god, it's the best police department ever.
00:15:07.039 --> 00:15:15.519
But the local media puts pressure on these police departments, then the national media does these comparisons across big cities or across states.
00:15:15.759 --> 00:15:32.480
So, what do you say to people who say, well, there's all kinds of pressure on these police departments to what we would call in the criminal kill crime or doctor the numbers, and that and they're doctoring the numbers is skewing UCR or the SRS or NIBERS.
00:15:32.559 --> 00:15:34.240
What do you what's your take on that?
00:15:34.480 --> 00:15:45.679
I'll often get when crime is going up, that those numbers are being doctored so that police departments can improve their budgets and you know get more money because crime's out of control.
00:15:46.159 --> 00:15:54.879
And when they're going down, they're being doctored because the police departments want, you know, they want to be able to say that they're fighting crime effectively.
00:15:55.120 --> 00:16:13.360
And it, you know, it fits into this, I think, in a lot of things in American society right now, and probably world society as well, that there are issues that require being careful, being, you know, looking at it, understanding the data, understanding the the uncertainty in the data, and understanding the potential misuses of the data.
00:16:13.519 --> 00:16:19.440
But there's a tendency to just not believe information that comes, even if the information is correct.
00:16:19.679 --> 00:16:21.600
I I see that very frequently.
00:16:21.679 --> 00:16:24.799
I I don't find it it necessarily compelling.
00:16:24.960 --> 00:16:29.039
And I think that there are ways to use other sources.
00:16:29.360 --> 00:16:33.840
So if you you know you have a problem with FBI data, you can use an agency's data.
00:16:33.919 --> 00:16:49.840
If you have a problem with an agency's reported data, you can use sometimes if they have it publicly available, you can use their calls for service data, which it's a lot harder to doctor a half million calls for service that come in through a dispatcher versus what the police department actually reports itself.
00:16:50.080 --> 00:16:56.960
You don't like the police department at all, you can use the gun violence archive and evaluate gun violence through a data source like that.
00:16:57.279 --> 00:17:01.600
So there are ways to do um different data sets.
00:17:01.840 --> 00:17:08.480
DC is a good example because DC has been sort of in the news for the unreliability of their crime data.
00:17:08.720 --> 00:17:11.119
And you can look at the gun violence archive.
00:17:11.200 --> 00:17:14.960
The DC Public Health Department has a gun violence dashboard.
00:17:15.119 --> 00:17:27.200
You can compare the data coming from the FBI with the data coming from the city itself, and you can see, okay, maybe robbery and aggravated assault aren't really accurate.
00:17:27.359 --> 00:17:30.400
There's there's questions about the reliability of that data.
00:17:30.720 --> 00:17:41.200
And you can then look at something like the FBI data, their carjacking data, and and their murder data, and say, okay, these look a lot more comparable and this is more reliable.
00:17:41.359 --> 00:17:48.000
So there's all sorts of ways that you can curiously investigate data without sort of out of hand saying, no, this is wrong.
00:17:48.079 --> 00:17:48.640
I don't believe it.
00:17:48.960 --> 00:17:51.839
So you and I share a strong dislike.
00:17:52.000 --> 00:18:01.119
I've looked at your Substack for hyperbolic or make-believe analyses uh published by these what I call real estate adjacent websites.
00:18:01.920 --> 00:18:05.200
For instance, I see you referring your substack to neighborhood scour.
00:18:06.880 --> 00:18:12.480
So what are your biggest problems with these types of analysis these websites do?
00:18:12.799 --> 00:18:21.920
Because I know I get asked all the time, and ever since I started grad school, really from friends, from family, from friends of friends of friends.
00:18:22.160 --> 00:18:30.480
They see some site and they're not sure if they should move into this neighborhood because some site like neighborhood scout says that's a really dangerous neighborhood.
00:18:31.039 --> 00:18:44.240
But you could never you'd have to really dig in to try to figure out what because I was always the maze in first blush looking at them, wow, how did they get access to all that data to come up with that analysis?
00:18:44.799 --> 00:18:46.000
That's kind of amazing.
00:18:46.400 --> 00:18:49.440
So I just wanted to get your uh get your take on this.
00:18:50.559 --> 00:18:54.640
Yeah, a lot of these sites do that type of analysis.
00:18:54.799 --> 00:19:09.759
It's unfortunate, it's oftentimes taking analysis uh or taking data from UCR and saying, okay, there were a thousand robberies in this city last year.
00:19:10.000 --> 00:19:20.559
Let's basically use our own proprietary algorithm and say that okay, the you know, the the poor areas are gonna have 70% of the robberies.
00:19:20.720 --> 00:19:29.200
So we'll say that this neighborhood had, you know, 700 robberies, and this area had you know 10%, probably.
00:19:29.359 --> 00:19:32.079
So then this and there's no way to really track it.
00:19:32.240 --> 00:19:37.519
And if a city doesn't report data to the FBI, they will make up a number.
00:19:37.839 --> 00:19:42.720
And they won't they'll they won't be transparent about when the data came from.
00:19:42.880 --> 00:19:50.720
And oftentimes there'll be splashy graphics that people want to say, hey, this is the worst place in America, or this is the best place in America.
00:19:50.960 --> 00:19:53.519
That's a story that people sometimes want to tell.
00:19:53.759 --> 00:20:02.559
And just because the data is very clearly should not be compared, um, we shouldn't be ranking places using the FBI data.
00:20:02.640 --> 00:20:10.640
Uh, that doesn't stop organizations that don't care about the the inefficiencies or the uncertainties of the data.
00:20:10.720 --> 00:20:12.640
Um, they're just gonna publish it anyway.
00:20:13.119 --> 00:20:22.960
So, in general, what are your feelings on how local media typically reports on crime and violence and crime data generally?
00:20:23.359 --> 00:20:34.079
And I'm not talking the one-offs because there's you know, we feature a lot of journalists on our podcast doing great reporting, but I'm talking about the daily news coverage around the country.
00:20:34.559 --> 00:20:45.920
Because you know, we know that for the last few decades at least, the fear of crime has always far outpaced the likelihood of people being victimized.
00:20:46.160 --> 00:20:50.640
So I just wanted to get your feelings being an expert in the data, how you felt it was being reported on.
00:20:50.960 --> 00:20:51.920
It's really hard.
00:20:52.079 --> 00:21:02.319
It the the challenge that I know media faces a lot is that it's easy to report things that happen, it's not easy to report things that don't happen.
00:21:02.640 --> 00:21:06.160
So there's that saying that the media doesn't cover the planes that land.
00:21:06.319 --> 00:21:17.920
And I think that that's apt for crime data because you'll never see a story that, hey, there was no carjacking last week, or um, Walgreens didn't report any shopliftings.
00:21:18.240 --> 00:21:30.799
And so when people un try to understand the their environment through anecdotes, it leads to a belief that there's always things are always getting worse.
00:21:30.880 --> 00:21:33.920
Um, and that's really hard, I think, for media to do.
00:21:34.160 --> 00:21:41.119
And this is coming from someone that thinks it's really important that when crime occurs, like it's important that we understand our crime trends.
00:21:41.200 --> 00:21:49.680
So when things are getting worse, that's an important story for local media to be telling, but they really struggle with things are getting better.
00:21:50.000 --> 00:21:54.400
And and I wish I had a better answer for how to solve that problem because I I don't know what it is.
00:21:54.559 --> 00:21:59.759
There certainly are a lot of journalists out there that are very data literate and try to address that issue.
00:22:00.160 --> 00:22:10.319
And then there's journalists out there that aren't data literate and don't try to address, you know, what what is the overall context of with which this crime is occurring.
00:22:10.720 --> 00:22:15.839
Back many years, God, 10 years ago, maybe even more, 12 years ago, we did an analysis.
00:22:16.000 --> 00:22:23.039
We studied one year of sexual assault and and domestic violence reporting in the Chicago Tribune and Chicago Sundi.
00:22:23.359 --> 00:22:26.160
And we basically figured out that.
00:22:33.359 --> 00:22:36.960
Domestic violence, the only thing that made the news were hyper-violent cases.
00:22:38.480 --> 00:22:42.559
And for sexual assault, it was almost it was like 90% stranger assaults.
00:22:42.960 --> 00:22:50.400
The media, when we released it, we got a I got a lot of criticism from journalists saying, well, how do we report on sexual assault?
00:22:51.119 --> 00:22:56.160
The the victims, we can't use the victims' names and they hardly talk to us.
00:22:57.200 --> 00:23:05.680
And I'm like, well, the and me thinking being a big data guy, it's like it's not the details of an individual case that matter.
00:23:06.240 --> 00:23:10.960
Like if you if there's an assault at DePaul University, it isn't about Jane Smith that gets assaulted.
00:23:11.119 --> 00:23:17.759
It's about whether there's a trend going up or down, and what's the trend in the university responding to those things.
00:23:18.160 --> 00:23:22.799
Yeah, it it and it's hard to do on a daily basis and do it coher, you know.
00:23:23.279 --> 00:23:37.839
We talk a lot on this podcast about I think unfortunately, and it's doing us a great disservice, that crime and violence data is exploited for political purposes, for agenda purposes.
00:23:38.160 --> 00:23:42.000
I think the media exploits it for clicks.
00:23:42.240 --> 00:23:45.920
I think polit politicians exploit it for their own agenda.
00:23:46.160 --> 00:23:56.079
I know even in like the and you've talked talked about this on your uh Substack, crime is going down, has gone down, violent crime has gone down over the last couple years.
00:23:56.319 --> 00:24:10.799
While there's a report of a really bad violent incident in a city, and the politicians who are not in office or not in the powerful office they want to be, will criticize and say, Well, crime can't be going down.
00:24:11.359 --> 00:24:12.160
It's impossible.
00:24:12.319 --> 00:24:13.440
Look what just happened.
00:24:13.680 --> 00:24:17.839
And I try to tell people like bad things still happen as crime goes down.
00:24:18.000 --> 00:24:20.480
It just means less of the bad things are happening.
00:24:20.640 --> 00:24:22.640
It doesn't mean they're completely eliminated.
00:24:22.799 --> 00:24:34.160
So I just wanted to get your take on whether or not you think those things, like crime data, violent crime and crime and violence data is being gets exploited to the detriment of the public.
00:24:34.640 --> 00:24:35.599
I mean, entirely.
00:24:35.759 --> 00:24:55.440
And you see this in especially in politics, where uh I did a uh congressional hearing in in North Carolina last year, and they had, you know, the parents of people that had been murdered and the siblings and a police officer that had been shot and these terrible incidents.
00:24:55.839 --> 00:25:03.119
And then, you know, they had all of these people interviewing, and then I came up as the witness that basically said, like, these are horrible tragedies, it's awful.
00:25:03.279 --> 00:25:06.319
I can't imagine what all of this is going down.
00:25:06.480 --> 00:25:10.240
And if you can't accept that it's getting better, then you can never learn why.
00:25:10.319 --> 00:25:14.079
And if you never learn why, then you can never make things actually better.
00:25:14.240 --> 00:25:28.720
And I mean, I got I got a ton of flack from most of the Congress people out there that were like, Well, you know, you're, you know, you're lying, or you're uh you're you're, you know, you didn't consider X moving the goalposts and things like that.
00:25:28.880 --> 00:25:46.880
And it's just, it's not an easy thing, I think, to talk about, because if we talk about how murders probably gonna have we're probably gonna have the fewest murders ever recorded in 2026, like by far, the lowest murder rate ever recorded, you're still talking about 12,000 murders.
00:25:47.119 --> 00:25:58.240
And so if you want to find 12,000 tragedies or 12,000 individual stories of people who lost their lives when they shouldn't have, like that's terrible.
00:25:58.400 --> 00:26:03.680
And absolutely we should always acknowledge that that that the issues remain unsolved.
00:26:04.160 --> 00:26:07.920
But that doesn't mean that the issue isn't getting better.
00:26:08.079 --> 00:26:16.640
And it doesn't mean that it that it would be a wasted opportunity to not know why it's getting better because not learning why is the worst thing we could possibly do.
00:26:16.960 --> 00:26:23.039
Especially on this issue, people don't care why, they just want it going in whatever direction they want it going.
00:26:23.759 --> 00:26:24.480
That's it.
00:26:24.720 --> 00:26:35.359
And for a lot of people, a lot of politicians, especially the media to some extent, it can't ever be going down when their person is not in power.
00:26:35.759 --> 00:26:38.559
Then everyone's lying if it's going down.
00:26:38.720 --> 00:26:44.400
And then once they get power or their people get power, then it's starting to go down and it's okay.
00:26:44.559 --> 00:26:48.960
Then it's directly related to their policies, which just annoys the hell out of me.
00:26:49.119 --> 00:26:55.599
It's like you look, if you're talking about the fewest murders in America, most likely in 2026, maybe for forever, right?
00:26:56.000 --> 00:27:01.200
And people are like, you have all these local politicians and local police chiefs and editorial pages.
00:27:01.279 --> 00:27:03.119
Look, how good are police departments doing?
00:27:03.279 --> 00:27:06.240
And there's no doubt the police department has some impact on that.
00:27:06.319 --> 00:27:08.960
There's unit's undoubtedly undoubtedly, right?
00:27:09.200 --> 00:27:12.640
But at the same point, we're a country of 350 million people.
00:27:12.799 --> 00:27:28.319
And if crime is going down generally all over the country, all at the same time, we might want to look at something other than just the policies and practices of 18,000 separate police agencies as for why the crime is crime numbers are going down.
00:27:28.400 --> 00:27:34.640
It doesn't mean the police don't have a role, but we may look try to find a solution that is more universal throughout our country.
00:27:34.799 --> 00:27:42.240
There probably are disparate policies that are probably opposite of each other in two different cities across the country that have crime drops that are very similar.
00:27:42.400 --> 00:27:45.759
What troubles you the most about how America deals with crime data?
00:27:46.000 --> 00:28:01.920
I mean, I think it it's very similar to your your previous statement that we have a tendency to that whatever was happening at the moment that the trend changed is what we're going to credit or blame for the change.
00:28:02.160 --> 00:28:10.000
And so, you know, sometimes it makes sense that we did X gun violence reduction policy, and hey, gun violence went down.
00:28:10.079 --> 00:28:10.559
That's great.
00:28:10.640 --> 00:28:12.960
That's that's pretty good evidence that this thing worked.
00:28:13.039 --> 00:28:15.920
And if we study it, we can find that you know there's a correlation there.
00:28:16.079 --> 00:28:21.119
Other times it's like I was hopping on one leg and then crime started falling.
00:28:21.200 --> 00:28:26.240
And so if we keep hopping on one leg, then that's gonna lead to fewer crimes.
00:28:26.319 --> 00:28:40.640
And and so sometimes I'm I'm looking, I have a piece coming out soon on looking just at at San Francisco vehicle burglaries and auto thefts and the degree to which technology influenced those those things just plunging.
00:28:40.960 --> 00:28:55.839
And it it makes sense, but then you you one, you look at it, the data, and you see that hey, they went from 30,000 vehicle burglaries in a year to 5,000 vehicle burglaries a year in the span of like four years.
00:28:56.000 --> 00:29:02.160
And all of that started before they, you know, before their flock cons before their flock um contract.
00:29:02.240 --> 00:29:04.079
It started before they had drones.
00:29:04.319 --> 00:29:08.079
Auto thefts again started falling in in the beginning of 2023.
00:29:08.160 --> 00:29:11.359
And it wasn't until mid-2024 that they had Flock.
00:29:11.599 --> 00:29:21.839
And so if you want to credit the technology, you have to explain why these huge declines started even before that and why they match national standards.
00:29:22.000 --> 00:29:32.799
And then, you know, how drones are gonna actually stop uh vehicle burglaries, or how drones are gonna stop, or how uh license plate readers are gonna stop vehicle burglaries when the car doesn't move.
00:29:32.960 --> 00:29:33.759
So good questions.
00:29:33.839 --> 00:29:41.200
Maybe there's a role there, but I think that it's a lot harder to say than just this is a technology we implemented in the last three years.
00:29:41.359 --> 00:29:43.599
It must be responsible because we're seeing a drop.
00:29:43.839 --> 00:29:59.440
And so I wish that we had much more academic research into why the trends we are seeing are happening, and that level research is happening, but maybe not to the degree that the the potential to learn from this historic time is warranting.
00:29:59.759 --> 00:30:03.200
Yeah, I used to have well, I have a friend who's now retired.
00:30:03.279 --> 00:30:08.319
He used to be a district commander in Chicago over the area where Wrigley Field is, where the Cubs play.
00:30:08.400 --> 00:30:10.960
It's a very popular lots of bars and restaurants.
00:30:11.279 --> 00:30:14.319
Gets crowded on the weekends, super crowded on game days.
00:30:14.480 --> 00:30:20.160
And he, when he got there, they had a lot of problems with bar, the bar fights, and violence around the bars and alcohol.
00:30:20.240 --> 00:30:30.319
So he created a special unit and worked with the bars and had a huge presence on those on the strip after game before games, after games on the weekend nights.
00:30:30.480 --> 00:30:36.960
And lo and you know, they had the Com Stamp meetings, and he got a big applause because his numbers plummeted during the Cub season.
00:30:37.200 --> 00:30:42.960
And then he goes to one and they're like, Well, why did your you know your uh car thefts are up and car burglars are up?
00:30:43.119 --> 00:30:45.519
What the hell are you why aren't you dealing with that?
00:30:45.759 --> 00:30:52.079
He's like, Because the other guy had a special unit working on that, but this this other thing went up, the bar issue went up.
00:30:52.160 --> 00:30:58.799
So I took that special unit, moved to the bar thing, which was more violent than the car break-ins, and that's those numbers have plummeted.
00:30:58.960 --> 00:31:01.200
Well, you got to do something about those car things.
00:31:01.279 --> 00:31:01.920
He goes, I am.
00:31:02.000 --> 00:31:05.599
I'm gonna take the unit and I'm gonna move it over to deal with the car burglaries.
00:31:05.759 --> 00:31:07.920
Okay, well, you know, we'll see you next month.
00:31:08.000 --> 00:31:09.519
I'm like, you will, and you know what's gonna happen?
00:31:09.680 --> 00:31:13.759
That's gonna go down, and my numbers at the bars and the violence is gonna go up.
00:31:13.839 --> 00:31:15.519
And you're gonna tell me to do something about it.
00:31:15.680 --> 00:31:18.640
I'm gonna take the guys that are on the cars and move them back to the bars.
00:31:18.799 --> 00:31:21.680
And then he goes, and he would tell me, he goes, No, this is whack-a-mole.
00:31:21.759 --> 00:31:23.200
It's my turn they go get yelled at.
00:31:23.279 --> 00:31:24.720
And he goes, It's whack-a-mole.
00:31:24.799 --> 00:31:28.240
Wherever we have the resources we put it, the thing gets addressed somehow.
00:31:28.319 --> 00:31:34.400
But the other things in the city that are in my area that aren't getting addressed because I have them concentrating on the other thing, it's happening.
00:31:34.480 --> 00:31:35.440
It's whack-a-mole.
00:31:35.519 --> 00:31:38.960
So I just thought that has just stayed with me from a police and perceptive.
00:31:39.200 --> 00:31:39.440
Okay.
00:31:39.599 --> 00:31:42.000
Jeff Asher, thank you so much for jumping in with us.
00:31:42.079 --> 00:31:43.119
We really appreciate it.
00:31:43.440 --> 00:31:44.079
My pleasure.
00:31:44.240 --> 00:31:47.440
You know, that was Jeff Asher from AH Datalytics.
00:31:49.119 --> 00:31:54.880
When the only facts that can exist are the ones that reinforce our political views, the world is in grave danger.
00:31:55.119 --> 00:31:58.319
We have been at this point for many years on our criminal justice system.
00:31:58.799 --> 00:32:06.240
The rabid belief that crime and violence go up or down solely because of the actions of local police and politicians is crazy.
00:32:07.359 --> 00:32:18.480
Especially given the fact that national patterns across dozens, if not hundreds, of cities cannot be based solely on the policies and practices of individual police agencies across all those jurisdictions.
00:32:18.960 --> 00:32:25.440
Arguing we need better crime and violence data and transparency from justice agencies is an argument CJP was built on.
00:32:25.680 --> 00:32:26.559
We're there with you.
00:32:30.640 --> 00:32:37.519
That was That was Jeff Asher from AH Datalytics.
00:32:39.200 --> 00:32:44.960
When the only facts that can exist are the ones that reinforce our political views, the world is in grave danger.
00:32:45.759 --> 00:32:48.960
We have been at this point for many years on our criminal justice system.
00:32:49.680 --> 00:32:57.279
The rabid belief that crime and violence go up or down solely because of the actions of local police and politicians is crazy.
00:32:58.240 --> 00:33:09.599
Especially given the fact that national patterns across dozens, if not hundreds, of cities cannot be based solely on the policies and practices of individual police agencies across all those jurisdictions.
00:33:11.039 --> 00:33:14.799
Arguing that we need better crime and violence and data because of that.
00:33:43.119 --> 00:33:50.400
But arguing we cannot trust any of the data because it doesn't reinforce our political agenda is dangerous and bad for communities.
00:33:51.039 --> 00:33:54.559
Thanks for listening, and as always, please share with your friends and contacts.
00:33:54.720 --> 00:33:56.240
I'll be back with you next week.
