.svg)
Most multifamily operators begin market evaluation the same way: pull rent growth trends, check occupancy rates, review cap rate movement. It is a common starting point, and a misleading one.
When deal flow initiates the decision-making process — a site surfaces, underwriting begins, capital follows — market selection becomes reactive rather than intentional. Assumptions vary by deal and by analyst, decisions are made without a common framework, and the organization never gets systematically better at allocating capital.
Stonewater Group completed a national multifamily market selection study in January 2026, screening 85-plus metropolitan statistical areas across the United States. The goal was to rank markets not by how they performed last quarter, but by how they are positioned over the next three to five years. The methodology relied on a seven-pillar z-score framework, four primary institutional data sources (CoStar, RealPage, Oxford Economics, and the US Census), and 1,000 Monte Carlo simulations per market to stress-test each ranking under a wide range of economic scenarios.
The results challenge several assumptions that currently shape how sub-institutional operators and GPs allocate capital.
The seven pillars are not weighted equally. Demographics and household growth carry 27.5 percent of the total score — the single largest weight in the model. Supply and pipeline risk carries 22.5 percent. Economic growth and labor market dynamics account for 20 percent. Rent and occupancy performance, the data most operators prioritize, accounts for only 12.5 percent.
That distribution is deliberate. Rent data is a lagging indicator — by the time strong rent growth appears in CoStar, it has already been priced into acquisition comps, construction pipelines are already responding, and the entry window is often closing. Household formation trends and demographic migration patterns move more slowly and telegraph structural demand years in advance.
The practical implication: a market with strong demographic tailwinds but soft current rents may represent a significantly better long-term investment than a market with record rent growth but decelerating household formation. The framework weights accordingly.
Supply risk sits at 22.5 percent for a related reason. Most operators treat pipeline data as a secondary check — something reviewed after a market passes initial screening — but this study treats it as the second-most important variable in the model. A market with strong demographics and a pipeline set to deliver 8,000 units into a 60,000-unit stock is a fundamentally different bet than the same demographic profile paired with a constrained pipeline, and that distinction changes the deal.
1. Boise City, ID — Z-score: 1.52
Boise holds the widest margin over any other market in the dataset. It ranks third nationally on the demographic and household growth pillar, first nationally in homeownership pressure, and seventh in development feasibility. Near-term supply digestion creates a pause in rent growth, but the structural story remains intact. For operators with a three-to-five year horizon, the fundamentals are as clean as any market in the country.
2. Austin-Round Rock, TX — Z-score: 1.18
Austin is the most discussed market in the dataset, and among the most misread. It ranks first nationally in economic growth and labor market performance, and also ranks 75th in supply risk — yet still placed second overall. The appropriate framing is cyclical entry, not avoidance. Today's pricing dislocation reflects a supply cycle already nearing its delivery peak, and operators projecting a 2027 or 2028 delivery are not buying into peak-cycle conditions; they are buying into a structurally advantaged tech market at a discount to replacement cost.
3. Jacksonville, FL — Z-score: 1.03
Jacksonville's ranking is driven by breadth rather than any single outlier metric. Population growth, employment diversification, in-migration, and a mid-pack supply pipeline (29th nationally) combine into a market with no obvious weak spots, and development feasibility ranks sixth nationally. For operators seeking near-term activity without the concentration risk of larger Florida metros, Jacksonville fits the profile.
4. Lakeland-Winter Haven, FL — Z-score: 0.99
Lakeland ranks first nationally on household growth — the most consequential single-pillar result in the model, given that demographics carry 27.5 percent of the total score. The driver is geographic position: Lakeland sits at the intersection of the Tampa and Orlando metro areas, absorbing population spillover from both. The primary risk is feasibility, ranking 50th nationally, which means developers need to underwrite to tighter margin assumptions.
5. Charleston, SC — Z-score: 0.98
Charleston combines strong population and employment growth with the second-best development feasibility ranking in the country, and its supply pipeline sits 14th nationally in terms of risk. Charleston's multifamily fundamentals are positioned to remain structurally sound over the next three to five years — a durable profile for either development or core-plus acquisition strategies.
Developers benefit most from Charleston, Jacksonville, and Boise, where manageable hard costs and entitlement environments align with real demand-side growth. Lakeland fits for developers who can underwrite to tighter margins.
Investors have the most actionable opportunity in Austin and Raleigh, where cyclical supply peaks are creating entry points that structural fundamentals will eventually correct. Conservative exit cap assumptions and value through basis — not rent spikes — is the appropriate posture in both markets.
Lenders need to read this market list carefully. A market like Cleveland, which ranks in the 97th percentile for hard costs, requires sponsor quality, pre-capitalized interest reserves, and a 12-to-18 month liquidity cushion before it fits a reasonable credit box.
The Austin result is the clearest example of what the framework is designed to surface: a market that ranks 75th in supply risk still ranks second overall. That will be counterintuitive to any operator who uses supply pipeline as a disqualifying filter, but the study's position is that supply cycles are temporary and structural demand is durable, and that the relative weighting of those two signals should reflect that asymmetry.
Most informal market screening processes do not make that distinction — they either exclude supply-heavy markets outright or treat pipeline data as one of many qualitative inputs. The result is that some of the most structurally sound markets in the country get filtered out during early-stage evaluation, and capital concentrates in markets that look clean on the surface but lack the underlying fundamentals to support long-term performance. This is the core problem with deal-driven capital allocation: without a standing thesis, each market is evaluated in isolation and the process never gets systematically better.
This national market screen is Phase I of Stonewater Group's Investment Operating System — a five-phase decision framework that spans the full development lifecycle. Once a market clears national screening, the process moves through submarket intelligence, site sourcing, site-specific market analysis, and institutional underwriting, with each phase narrowing the focus and adding specificity. Over time, these engagements compound into a documented body of market intelligence and underwriting history that becomes a durable competitive advantage.
Sponsors who operate this way enter capital conversations differently. When a lender or equity partner questions a market assumption, the answer is a formal analysis that predates the deal — not a defense constructed to support it.
Stonewater Group completed this national screen for a regional multifamily development firm evaluating new market entry. The underlying study — covering all 85-plus MSAs, full pillar-by-pillar scores, and the Monte Carlo output — is available for download below.
85+ MSAs. Seven-pillar z-score framework. 1,000 Monte Carlo simulations per market.