How the HousingGauge score works

What the score measures

The score summarizes how favorable current conditions are for a disciplined buyer or investor: the cost of financing, the monetary backdrop, negotiating leverage, demand, valuation, the local economy and rental economics. It is a market conditions score — not a price forecast, and not a recommendation to buy or sell any property.

• GREEN (70–100): conditions appear comparatively favorable for disciplined buyers and investors.
• YELLOW (45–69): mixed conditions; market selection and individual deal economics matter greatly.
• RED (0–44): conditions currently appear less attractive for aggressive leverage.

Components and weights

Each component is scored 0–100 from its inputs. Each input is mapped linearly between a value that scores 0 and a value that scores 100, and clamped at the ends. Weights and bounds are stored as data (scoring model v1) so they can be revised without changing the application; every snapshot records the model version that produced it.

Core formulas

• Real policy rate = fed funds rate − expected inflation
• Policy gap = real policy rate − r-star (positive means policy is restrictive)
• Real home price = nominal price × CPI_current ÷ CPI_{observation date}
• Real price per sq ft = nominal $/sq ft × CPI_current ÷ CPI_{observation date}
• Price-to-income = median home price ÷ median household income
• Gross rental yield = annual median rent ÷ median home price
• Payment-to-income = principal and interest on a 30-year fixed loan for 80% of the median price ÷ monthly median household income (excludes taxes and insurance)

Why mortgage rates matter so much

Most buyers finance. The mortgage rate sets the monthly cost of the same house, so it moves affordability for every market at once. That is why mortgage conditions carry the largest single weight and appear prominently on every market page. Mortgage and monetary inputs are national, so they move all markets together; local components explain why markets differ.

What r-star means — and what it doesn't

r-star is the estimated neutral real interest rate: the policy rate that neither stimulates nor restrains the economy. When the real policy rate is above r-star, policy is restrictive. HousingGauge uses the published Federal Reserve Bank of New York estimate. r-star is unobservable and estimated with wide uncertainty.

A negative policy gap does not mean home prices will rise. Easier policy tends to ease credit conditions over time, but home prices depend on local supply, incomes, migration, mortgage spreads and much more. The policy component is one input describing the credit backdrop, not a prediction.

Correlation is not causation

The components describe conditions that have historically mattered to housing returns. A high score means several of those conditions are favorable at the same time — it does not mean any one of them causes prices to move, and past relationships may not hold.

Explanations, scenarios and AI

“Why this score” explanations and “what would turn this market GREEN” scenarios are generated by deterministic code from the market's own numbers. Scenarios solve the model for the input values that would cross a status line, spreading the change across the inputs with the most room to move. Written reports and podcast scripts may be drafted by a language model after all numbers are computed; every number in generated text is automatically checked against the underlying data, and anything that fails the check is held for human review instead of being published. An AI model never decides a score or a status.

Data sources

• Mortgage rates: Freddie Mac Primary Mortgage Market Survey (via FRED)
• Fed funds rate, CPI, inflation expectations, housing starts: Federal Reserve Economic Data (FRED)
• r-star: Federal Reserve Bank of New York (Holston-Laubach-Williams)
• Unemployment and employment: U.S. Bureau of Labor Statistics (LAUS)
• Household income: U.S. Census Bureau American Community Survey
• Sale prices, price per square foot, inventory, months of supply, days on market, sale-to-list, price cuts, closed and pending sales: Redfin Data Center — housing market data courtesy of Redfin, a national real estate brokerage (rolling three-month windows, city level)
• Rents: Zillow Observed Rent Index (ZORI), Zillow Research

Pilot status: all scored inputs now come from the public sources above. Where a source doesn't cover a market (for example, BLS publishes city labor data only for cities of 25,000+ people), that input is left empty and its weight is redistributed, as described under Limitations. Each figure on a market page shows its own source and observation date.

Limitations

• Local data can be thin. Small markets have few sales, so medians are noisy week to week.
• Data arrives on different schedules (weekly rates, monthly housing data, annual income). Each snapshot uses the latest available value and records its date.
• When an input is missing, its weight is redistributed within its component and the snapshot records data coverage.
• Market boundaries matter: a city, a ZIP code and a metro can tell different stories.
• The model is intentionally simple and transparent; it does not capture property-specific factors.

How often scores update

Weekly. Each run fetches national data once, refreshes every active market, recalculates components, saves a dated snapshot, and publishes the market page, report and podcast. Historical snapshots are kept so you can see how a market moved.

Independence

Market Partners and sponsors can present a market page but have no access to, or influence over, its data, score or analysis.

HousingGauge provides market information and analytical tools, not individualized investment, financial, legal or tax advice.

Questions about the methodology? See About HousingGauge.