At $2,602 in June 2026, ZIP 92104’s Zillow asking-rent index sits in a measurable affordability tension with the matched ACS median gross rent of $1,993, a 30.6% difference across non-equivalent rent concepts. A 30% required-income screen converts the current asking-rent index to $104,080 of annual income. That arithmetic benchmark exceeds the ZCTA median household income of $94,014, producing a 33.2% asking-rent-to-income screen; it is not advice, an applicant qualification rule, or a statement about any household. The ACS rent-burden measure also reports that 47.9% of renter households pay at least 30% of income toward rent, reinforcing the area-level tension without proving the burden or rent on any particular unit.
The current reading follows a much slower recent path than the longer history. The exact same-month annualized one-year ZORI change was 0.4%, versus 0.9% over three years and 5.0% over five years. Thus, the latest direction breaks from the stronger five-year pace and is also below the three-year trend, although all three retrospective measures remain positive. Annualized monthly-return variability measured 1.9%, indicating relatively restrained movement in the historical index. The maximum drawdown was 1.3%, a limited historical peak-to-trough decline. Coverage is complete at 100% across 138 observations and 137 consecutive monthly returns. Transparent national discovery ranks place momentum at 2,193, stability at 61, and balance at 1,200 among history-eligible ZIPs, with lower ranks stronger; these are descriptive discovery measures, not forecasts or investment ratings.
Definitions matter before comparing those figures. The five-digit label 92104 is both a Zillow ZIP market identifier and a Census ZCTA match; a ZCTA is a Census statistical area and is not identical to a USPS delivery ZIP. Zillow ZORI is a typical observed asking-rent index blended across rental types. ACS median gross rent is a five-year survey measure for occupied renter homes and includes selected utilities. HUD FMR/SAFMR is instead an administrative, bedroom-specific standard, not asking rent. Scaling the ZIP ZORI by the local HUD ladder produces modelled monthly estimates of $1,987 for a studio, $2,130 for one bedroom, $2,602 for two bedrooms, $3,469 for three bedrooms, and $4,202 for four bedrooms. They are modelled estimates, never measured bedroom rents. The corresponding HUD standards are $2,360, $2,530, $3,090, $4,120, and $4,990, respectively.
The matched ACS housing-stock record shows 25,064 housing units and 23,661 occupied units. Renters occupy 16,855 homes while owners occupy 6,806, making the renter share 71.2% of occupied housing. There are 1,403 vacant units, for a 5.6% vacancy rate. Among specified vacant categories, 680 units are reported vacant for rent and 50 vacant for sale. These are area-level survey counts, rather than a real-time inventory of rentable homes, and they do not establish availability, condition, lease terms, or pricing for a specific unit. The stock mix nevertheless provides important context for why an asking-rent index and an occupied-home gross-rent median can describe materially different populations.
Broader benchmarks position this ZIP below its surrounding rent measures, but they remain context rather than substitutes for ZIP evidence: San Diego city context rent is $3,038; San Diego County context rent is $2,991 and its two-bedroom HUD FMR is $3,001; and San Diego-Chula Vista-Carlsbad metro context rent matches the County figure, while metro months of supply are 2.6. Each value belongs to its named city, county, or metro geography rather than 92104 itself. The lower ZIP asking-rent index relative to these wider rent contexts does not make the measures interchangeable, because the city, county, and metro figures summarize broader markets with different housing composition and source coverage. The metro resale supply figure is likewise useful only as a wider comparator to the direct ZIP resale observation.
Redfin provides that direct rolling-three-month ZIP resale observation, and it belongs entirely to the for-sale market rather than rental transactions. Median sold price was $914,793, down 6.7% year over year. There were 76 homes sold, with a median 20 days on market, inventory of 90 homes, and 3.6 months of supply. Sale-to-list signals were mixed: the average sale-to-list ratio was 99.96%, while 35.2% of homes sold above list price. This resale evidence challenges any simple reading of the steadier rent history: asking rents were nearly flat while sold prices declined and supply exceeded the metro context figure. The annualized ZIP ZORI divided by median sold price is a 3.4% cross-source screening ratio only, not a cap rate, net return, expected return, or property yield.
The ACS burden and vacancy measures require especially careful interpretation. Rent burden describes survey-reported renter households in the matched ZCTA, not the affordability of a new lease, and vacancy does not prove that a particular property is available or suitable. The reported ACS renter-total, burden, and gross-rent estimates also carry survey margins of error. Meanwhile, the occupied-home gross-rent median can differ from a current Zillow asking-rent index because of resident tenure, included utilities, rental-type mix, and timing. HUD standards add a useful local bedroom ladder but do not supply market rent comparables. Read together, the sources identify a gap between current asking-rent conditions and older occupied-household outcomes; they cannot identify its cause or translate it into a lease outcome for any individual home.
Several limits remain material for property-level use. ZORI is an index rather than a live listing feed, the bedroom ladder is modelled rather than observed, and Redfin’s ZIP resale statistics are aggregates rather than rental comparables or property economics. A property-level review should verify advertised bedroom count, current asking price, lease start date, utility responsibilities, concessions, furnished status, and actual availability. For any sale-side comparison, it should also verify property type, condition, lot or unit characteristics, list history, closing timing, and whether the relevant sale resembles the subject property. The historical measures are backward-looking and the resale block is a dated observation, so neither establishes a forecast. The unresolved question is whether a specific listing’s terms and physical attributes align with these area-level screens.