Lurie's reform push is running straight into the math. And the math is not helping him. This is San Francisco Politics and Urbanism Daily — I'm Cassidy, Mark's here. Today: RESET's first real utilization numbers, the shelter-bed promise after the net count, and who ends up paying for City College's budget gap. It's been one of those weeks where Lurie keeps making big announcements, and then the receipts show up. Funny how basic arithmetic keeps ruining the vibe. We’ve got Gazetteer SF on RESET, Mission Local on the shelter-bed total, and the SF Standard on the City College cut — let's dig in. Gazetteer SF writes:
According to the sheriff’s office, RESET has admitted 300 people since it opened May 4. That’s an average of 18 people per day, which is below its total capacity of 25, especially if visitors aren’t staying at the center for a maximum 23-hour visit. It’s also a slowdown from the 140 people in its first week that a May 12 announcement by the mayor’s office touted.
Gazetteer SF has been walking through the RESET numbers, and the city’s own math is the problem. The mayor said “dozens and dozens” in the first 24 hours, the sheriff’s office says 300 total in three weeks, and that works out to 18 people a day against a stated capacity of 25. Meanwhile, the photos show empty chairs for stretches. We asked on May 19 what number would actually tell us RESET was working. Now we've got one, and it's not flattering. The city's own milestone release said 140 people in the first week — about 20 a day — and it’s already slowed after that. If you're under capacity in week three of your flagship program, that's a utilization problem, not a victory lap. Here's the contradiction that blows up the story: “dozens and dozens” in the first 24 hours means at least 48 people on day one alone, but the three-week average is 18 a day. So either day one was a huge outlier, or the TikTok version was doing more than a little bit of the work. And it's not just RESET. Mission Local says the administration promised 1,500 new shelter beds in six months, then hit the six-month mark with a net number that’s much smaller than the headline. Big announcement, blurry delivery — that’s becoming a pattern. So what's the actual standard here? A sobering center sounds humane, but how is it different from just giving people a place to sleep it off, and what would count as success? The basic idea is to break what the city itself calls the “revolving door” — someone gets picked up for public intoxication, ends up in the ER or jail, gets sent back to the street, and does it again. San Francisco's version is the RESET Center, which stands for Rapid Enforcement, Support, Evaluation, and Triage. It's overseen by the Sheriff's Office at 444 Sixth Street in SoMa. Per the Chronicle, the Board of Supervisors approved a 26-month contract with Connections Health Solutions, an Arizona-based for-profit operator, and the contract is performance-based, so payment depends on results instead of just keeping the place open. KQED reported that Mayor Lurie signed the enabling legislation in February, and by late April the center was being readied for a spring opening. The city's stated goals are to connect people to treatment and take pressure off emergency rooms and jails, so the numbers to watch are ER diversion, treatment referrals, and whether people actually connect to longer-term care instead of just cycling back out. The performance-based part sounds good on paper, but do we have any similar sobering centers to look at? And more importantly, do officers actually use them? That's the cautionary example here. NSPR reported that Butte County's sobering center in Chico had more than 1,600 visits in its first nine months, but law-enforcement referrals were pretty low, which undercut the whole point of diverting people from jail and the ER. So in San Francisco, the key number isn't just bed use — it's how many drop-offs come from SFPD and the Sheriff's Office, because a sobering center officers don't route people to is just an expensive waiting room. Eleni Balakrishnan, writing in Mission Local:
As of April, 16 months into his tenure, Lurie had overseen the opening of 863 new shelter beds, for a total of about 7,500 city-wide. But the city has also closed multiple shelters, with some shuttered due to decisions made before he took office — meaning the the net gain is considerably less. Under Lurie, San Francisco has only seen a net increase of 403 beds, under 30 percent of his initial goal.
Mission Local did the net math on Lurie's shelter-bed promise, and it's rough. He opened 863 beds, but once you factor in closures, the actual gain is 403 — under 30 percent of the 1,500 he campaigned on. And if the planned closures through 2028 happen on schedule, that drops to 303. And remember, at the six-month mark Kunal Modi basically retired the 1,500 number and swapped in “the right beds.” That's not a fresh plan; that's a reframe to dodge a promise that was already slipping. To be fair, some of the closures were decided before Lurie took office. But the net number is the one that matters to somebody sleeping outside, and 303 beds by 2028 is a long way from 1,500 in six months. We asked last week whether the Housing Trust Fund expansion was stretching the queue instead of fixing the pipeline — this is the same structural issue, just with cots instead of units. And the new beds are weighted pretty heavily toward locked treatment and sobriety-required settings. I'm not against that in principle, but if you prioritize those over basic shelter beds, you're helping people in acute crisis and leaving everybody else on the street. That's a policy choice, and the administration should say it plainly. From Garrett Leahy at SF Standard:
The program, called Free City, would see its funding reduced from $9.3 million to $6.4 million by eliminating cash grants that students use to pay for transportation, supplies, and other expenses. Qualifying students receive $46 per unit with a $552 cap per semester.
The Free City program still keeps tuition free — that part doesn't change. What Lurie is cutting is the cash-grant side: up to $552 a semester that about 6,000 low-income students used last year for transit, supplies, and the costs tuition doesn't cover. Garrett Leahy at the SF Standard broke it down, and that distinction matters. Free tuition doesn't mean much if you can't afford the bus ride to get there. These students are getting $46 per unit — we're talking a few hundred dollars that can decide whether somebody makes it through a semester. Lurie is taking $2.9 million out of a $643 million deficit by cutting into the poorest students in the city. And this is where the Prop D CEO-pay surcharge comes back into the picture. Supporters said that tax was meant to protect services exactly from cuts like this. Now we've got a named program, a named population, 6,000 students, and a very specific dollar figure. A press release saying Free City stays free will sound fine on paper. The students who can't afford to show up just won't enroll. That's the move. This one's from The AI Journal:
Today Zūm, the leader in modern student mobility, announced the deployment of an all-electric school bus fleet to San Francisco Unified School District (SFUSD). Powered by Zum’s Connected Mobility Experience (CMX™) solution, the deployment will combine electric vehicles, charging infrastructure, routing intelligence, real-time operations, and transparency for families into its single integrated transportation system.
Switching gears — SFUSD and a company called Zūm announced yesterday what they're calling the largest electric school-bus deployment in the country, with bidirectional vehicle-to-grid charging built in. Scott Wiener was at the groundbreaking, and so were Superintendent Maria Su and City Attorney David Chiu. V2G on school buses is actually interesting. Those buses sit idle for hours every day, which means they can send power back to the grid during peak demand. That's not greenwashing — that's using an asset you already have to do two jobs. Worth saying the sourcing here is a press release picked up by The AI Journal, so the Phase One details are still thin. We don't have fleet size, cost, or a full deployment timeline yet. I'll take “largest in the nation” a little more seriously once we see a contract number. If you follow San Francisco politics, you'll want the statewide picture too. Check out California Governor's Race for daily 2026 coverage of candidates, polling, debates, fundraising, and policy beyond the horse race. Find it wherever you listen to podcasts.
You'll find links to every story we covered today in the show notes, so if one caught your ear, you can follow it there and read more. That's San Francisco Politics and Urbanism Daily for today. This is a Lantern Podcast.