At first glance, 94118 pairs a fast rent reading with a far larger resale-price base. In June 2026, its Zillow ZORI—Zillow’s typical observed asking-rent index blended across rental types—was $4,212 per month, 17.2% above a year earlier. The five-digit label is both a Zillow ZIP market identifier and a Census ZCTA match; a ZCTA is a statistical area and is not identical to a USPS delivery ZIP. Annualized ZORI divided by the direct ZIP median sold price is a 2.02% cross-source screening ratio, not property economics or a measure of individual-home expenses. That wide resale-price backdrop sets the central tension; the separate for-sale evidence is developed below.
Rent history gives this snapshot a strong positive path but not an unqualified stability signal. Through the stated endpoint, same-month ZIP ZORI changes were 17.2% over one year, 8.9% annualized over three years, and 7.1% annualized over five years. Thus the latest direction confirms the longer upward path and accelerates substantially beyond it; this describes past observations, not a forecast. The record has complete coverage. Monthly rent changes annualize to 3.4% variability, which supports more confidence than a sparse or highly erratic series would, yet a historical 14.3% peak-to-trough drawdown demonstrates that the index has materially retreated at some point. The transparent national discovery ranks among history-eligible ZIPs, where lower is higher, were 23 for momentum, 2,138 for stability, and 469 for the balanced score. Those rankings are discovery tools rather than investment assessments. Full coverage improves traceability, while the drawdown means one current reading should not be treated as a permanent level.
Level comparisons must not blend their evidence universes. The matched Census ZCTA’s ACS 2024 five-year survey puts median gross rent at $2,721; it reflects occupied renter homes and includes selected utilities, rather than a current asking-rent index. June ZORI is higher, a gap that describes different populations, timing, rental-type mixes and rent concepts, not an error or a conversion factor. HUD’s two-bedroom standard is $3,604 below ZORI. That HUD FMR/SAFMR value is an administrative, bedroom-specific standard—not an asking rent—and it must not be read as a direct listing comparison. The ACS median and HUD standard can frame separate questions, but neither substitutes for a current ZIP asking-rent observation.
The bedroom view is deliberately modelled rather than measured. Scaling the ZIP ZORI by the local HUD bedroom ladder produces monthly modelled estimates of $2,904 for a studio, $3,479 for one bedroom, $4,212 for two bedrooms, $5,381 for three bedrooms and $5,577 for four bedrooms. The two-bedroom figure aligns with ZORI by construction, so it is not independent confirmation. These estimates translate the relative HUD ladder onto a blended asking-rent index; they are not observed bedroom rents, counts of available homes, or a substitute for bedroom-specific rental comparables. A listing’s actual configuration, condition, utilities and lease terms can depart from this modelled ladder.
The current index requires $168,480 of annual household income under a simple 30% of gross-income screen. The matched ZCTA’s ACS median household income is $163,995, making the same arithmetic rent-to-income calculation 30.8%. This is a budget screen only: it is not advice, an applicant qualification rule, or evidence of what any household can pay. Separately, ACS reports 30.2% of occupied renter households with gross-rent burdens at or above that threshold. Its gross-rent concept includes selected utilities and is not a current listing payment. The burden share therefore signals that a meaningful surveyed segment crosses that survey threshold, but cannot establish burden for a particular household, building, lease, or available unit. The history acceleration and this arithmetic tension should be held together, not collapsed into a prediction.
ACS housing counts add supply context without proving immediate availability. The matched ZCTA has 19,266 housing units and includes renter-occupied homes plus both single-family and large multifamily structures. Its 11.1% vacancy rate is a stock measure, not a vacancy guarantee for a particular unit. Reported vacancies include units for rent and seasonal use, so total vacancy should not be translated into listings. For wider context, Zillow’s City of San Francisco context asking-rent measure and the San Francisco County context asking-rent measure were each $4,401, while the San Francisco-Oakland-Berkeley, CA metro context asking-rent measure was $3,301. The ZIP index was therefore below the named city and county contexts but above the named metro context.
Redfin’s direct rolling-three-month ZIP resale observation is for-sale evidence, not rental transactions. Median sold price was $2,499,435, up 29.0% from a year before; 78 homes sold, median marketing time was 13 days, and inventory stood at 19 homes. With 0.8 months of supply, sales averaged 120.6% of list price and 75.1% sold above list. Those are resale liquidity and sale-to-list signals, not rental demand metrics or individual-home operating results. They confirm a tight for-sale snapshot alongside accelerating past asking-rent history. Yet the resale evidence challenges any shortcut from market speed to household fit: the income screen remained just above median income. This for-sale movement does not resolve whether a particular rental is affordable or correctly priced.
None of these series identifies an individual home or future outcome. ZORI is blended across types and the bedroom figures are modelled; ACS is a sampled survey with sampling uncertainty, HUD is an administrative standard, and city, county and metro context values are not ZIP substitutes. History is backward-looking, while resale observations concern completed for-sale activity rather than rentals. For a concrete listing-level review, confirm the bedroom count used against the modelled ladder, current asking payment, utilities included, lease duration, move-in charges, occupancy status, and whether the home is actually available. For a resale record, verify property type, sale date, condition, list-price history, and whether it belongs in the direct ZIP observation. The decision-useful question is not whether these aggregates promise an outcome, but which source matches the exact rent, household budget, or resale question being tested.