In San Francisco, there are two ways of assessing crime and safety issues: anecdotal and statistical. One is, naturally, more reliable than the other. And that's the anecdotal reports, of course.
Here's what happening anecdotally, according to police sources. Many of the workaday criminals who BART into San Francisco and once dealt drugs near stations at Powell, Civic Center, 16th Street, and 24th Street have, sadly, abandoned mass transit for cars.
They are, we're told, visiting into the outer San Francisco neighborhoods — the Outer Richmond, the Parkside, the Excelsior — and upgrading to a more lucrative line of work by breaking into cars. The MO, per a veteran cop, is for a thief to smash in a window, then wander away for 10 minutes or so. And, if nobody comes running out of a house, he'll return and help himself to the bounties within. "They're making more money [than dealing] by stealing electronics from cars in the sleepy neighborhoods, especially rental cars," says the officer. "They see it's a rental car and they pop it right open. Car break-ins are off the hook."
That's why neighborhood watch message boards are filled with all-caps warnings about a "kid checking all the driver's side doors on Gladstone!" or frantic retellings of last night's escapade, in which some miscreant briefly made off with the family Pontiac, shearing off a front bumper in an abortive attempt to wake up the sleepy neighborhoods.
Is this backed up statistically? Yeah. But even the most cursory examination of police statistics suggests that statistics aren't as statistical as they used to be. The crime totals printed in the most recent San Francisco Police Department annual report for the year 2012 do not match the 2012 totals printed in last year's. And the crime totals listed in the annual report for 2013 do not match the 2013 totals reported to the FBI.
Our calls to the SFPD on this matter were not returned. But, doing the math, this much is clear: Every change to the hundreds or thousands of cases makes things look better, crime-wise, in the most recent year on record.
Or at least as good as they can.
The year 2011 was a momentous one for the San Francisco Police Department. That was the year department employees finally received email accounts.
The department has, for years, experienced the pains you'd expect of an organization trying to implement the data-driven policing of the CompStat system while wallowing in the kind of institutional torpor that leads to a big press conference in 2011 to announce email addresses for police (who could only be reached prior to that by calling one phone in each station house, writing a letter, or committing a crime and waiting).
In 2012, Chief Greg Suhr asked the city controller to audit the department's record-keeping. That was not a pleasant report to read for fans of neat and functional government: The controller found the cops' CompStat system was hamstrung by too many streams of manually compiled information and misinformation, leading to a multiplicity of numbers where there really ought to be only one.
So, the numbers — the numbers that reveal whether or not crime is going up or down and by how much — are unreliable. And have been for some time. The San Francisco Examiner's Jonah Owen Lamb wrote a damning report on this earlier this year.
But the department's own official reports contradicting themselves? That's new. And that's interesting.
Every year the police, not unlike any company with a billion-dollar budget, produces a lengthy prospectus and year in review. You can chart whether things are getting better or worse by comparing the most recent year with previous ones.
By this measure, the year 2013 was a rough one. But not nearly as rough as it would have been if the 2012 statistics hadn't, somehow, changed a great deal.
In last year's annual report, for example, the department reported 3,484 robberies in 2012. And yet, in the current report, the 2012 total is listed as 3,703. That made the reported total of 4,000 robberies in 2013 not look so bad. This happens again and again: In 2012, the department reported 28,242 larcenies. But in the most recent report, that 2012 total has been altered to 30,200. In 2012, 5,317 burglaries and 5,339 auto thefts were reported. But, in the newer report, those totals have been upgraded to 5,671 and 5,770. Regarding burglaries and auto thefts, these post-facto changes allow the department to claim there's been a slight reduction in these crimes in 2013 rather than a sizable increase.
But, of course, it could be worse. Remember those 4,000 reported robberies in 2013? The total the SFPD forwarded to the FBI for the annual (and ironically titled) Uniform Crime Report was 4,202. The city reported 5,574 burglaries in 2013 on its annual report, but told the feds there were 5,931. And the city reported 904 fewer larcenies on its annual report than it did to the FBI.
Mixing and matching the numbers, you could say the city either experienced a 2 percent drop in burglaries between 2012 and 2013 — or a 12 percent increase.
And, either way, you've got the statistics to back you up.
In a recent interview with the libertarian publication Reason Magazine, Suhr explains how, after he largely dismantled the narcotics division, violent crime has, supposedly, dropped. "The thing we're most proud of here in San Francisco is, violent crime is down," Suhr said in the web interview.
The chief, uncharacteristically, did not return your humble narrator's calls. Which is too bad, as, per his own departmental report, violent crime is not down, but up 11 percent over last year. And, per the numbers his department reported to the FBI, it's up 22 percent. That's a 100 percent discrepancy with the SFPD's other numbers but, either way, up is up.
The SFPD, Suhr tells Reason, is "not trying to just keep a stat game going." Well, good.
Anecdotally, that's a pretty hard thing to do.