TL;DR: Quantamental Hospitality -- A Framework for Systematic Hotel Investing
Quantamental hotel investing combines institutional-grade quantitative data infrastructure with the irreducible qualitative judgment that hotel operating complexity demands. The framework is not algorithmic -- no screen replaces a 75-day diligence process or a management team interview -- but it is systematic: every market enters the same four sequential filters, every asset runs the same demand decomposition, and every IC submission carries the same sensitivity matrix and Liquidity Stress Delta calculation. The data stack runs STR STAR for trailing RevPAR benchmarks, Lighthouse for forward booking pace, Real Capital Analytics for cap rate and transaction comps, and AirDNA for short-term rental competitive intensity. Valuation applies income approach as primary (direct cap and DCF) and market approach as secondary (PPK and EV/EBITDA multiples by segment). Performance targets are strategy-differentiated: core-plus targets 8-11% net IRR, value-add 12-16%, opportunistic 18-22%. Every deal is stress-tested on a standard sensitivity grid -- exit cap rate, RevPAR CAGR, hold period -- and flagged with Bay Street's proprietary Liquidity Stress Delta (LSD) if timing-dependent risk exceeds 15%. The framework is market-first, asset-second, always: no compelling hotel story overrides a structurally unfavorable market screen.
Pure quantitative models fail at hotel underwriting for a structurally predictable reason: the asset class combines real estate (valued on cap rates and comparable transactions) with operating business (driven by management quality, brand positioning, and demand segmentation) in a way that resists full data reduction. A hotel's RevPAR index against its competitive set is a measurable, benchmarkable number -- but whether the competitive set is correctly defined, and whether the incumbent management team is positioned to capture RevPAR index gains, is a judgment call that requires human assessment. A quantamental framework acknowledges both realities: the quantitative screens eliminate markets and deals that cannot work on the numbers, while the qualitative overlays assess whether the deals that pass the screens can actually perform.
Pure qualitative approaches fail for the opposite reason. A compelling asset story -- a landmark hotel in an iconic location with a charismatic management team -- can mask market-level structural problems that a quantitative screen would surface immediately. Vietnam's hotel market has a compelling qualitative narrative: record arrivals (21.17 million in 2025), 17.1% RevPAR growth, and government infrastructure investment. But the quantitative supply-spread filter -- which requires RevPAR CAGR to exceed supply CAGR by 150 bps -- fails Vietnam for the 2025-2029 investment horizon because its projected 35% supply CAGR overwhelms even strong demand growth. The quantamental framework prevents narrative capture.
Systematic hotel investing requires a layered data stack. The primary benchmark layer runs STR STAR reports: RevPAR Index, Occupancy Index, and ADR Index for the subject property against a defined competitive set over trailing 12 and 36 months. STR is the hospitality industry's equivalent of Bloomberg for equities -- no institutional underwriting process is credible without it. The forward-looking layer runs Lighthouse (formerly OTA Insight) for booking pace analysis at 30/60/90-day windows and rate positioning versus the competitive set. Booking pace is often the most predictive early diligence input: a hotel with healthy group pace at 90 days and building transient pace at 30 days has a fundamentally different risk profile than one running flat-to-prior-year across both windows.
The transaction layer runs Real Capital Analytics and CoStar for cap rate comps, price-per-key benchmarks, and buyer-seller analytics across recent APAC hotel trades. The renovation cost layer pulls from Horwath HTL proprietary benchmarks and Procore project data, triangulated against CBRE and JLL Property Improvement Plan estimates from recent transactions. The alternative data layer includes AirDNA for short-term rental competitive intensity by submarket (a hotel market where Airbnb supply is growing at 25% per year faces a different demand ceiling than one where STR penetration is already saturated), Sojern for consumer travel demand signals, and credit card transaction data from Second Measure for actual consumer spend at hotel properties in target markets.
Demand forecasting platforms -- Duetto, IDeaS, and Atomize -- provide forward-looking revenue management signals that are most relevant post-acquisition, but useful in diligence for understanding the incumbent operator's yield management sophistication. An operator running Duetto's GameChanger and achieving above-index ADR performance is a materially different management quality indicator than one running a legacy rate management process.
Market selection is the most consequential decision in systematic hotel underwriting. No asset can outperform its market over a 5-7 year hold. The four sequential filters must all pass before asset-level work begins -- and they run in order of decisiveness, with the first filter eliminating the largest number of candidate markets.
Filter 1 is the RevPAR-to-Supply Spread. The target is markets where RevPAR CAGR exceeds supply CAGR by at least 150 basis points over a rolling 3-year forecast horizon. This separates demand-led markets from supply-led markets where RevPAR growth is absorbing new rooms rather than reflecting genuine pricing power. Japan passes: double-digit RevPAR growth against approximately 3% supply CAGR produces a spread well above 150 bps. Vietnam fails: 17% RevPAR growth against a projected 35% supply CAGR produces a negative spread. Singapore passes in gateway submarkets (Marina Bay, Orchard) but fails in suburban submarkets where supply pipeline is elevated. Filter 1 eliminates roughly 60% of candidate markets in any given investment cycle.
Filter 2 is the Cap Rate vs. WACC Spread. The target is a minimum 100 basis point positive spread between the going-in cap rate and the fund's weighted average cost of capital. This ensures immediate yield accretion and provides a buffer for operational improvement thesis execution. In Japan's compressed cap rate environment -- prime Tokyo luxury hotel yields are approaching 3.5-4.5% -- this filter redirects the investment approach from core acquisition to value-add, where a going-in yield of 5.5% against a 4.5% WACC passes the filter even as a stabilized core acquisition would fail it.
Filter 3 is the ADR vs. Occupancy Decomposition. RevPAR growth can be generated two ways: higher occupancy (more guests) or higher ADR (each guest pays more). ADR-led growth reflects pricing power -- the market's willingness to pay more for the same room -- and is fundamentally more durable than occupancy-led growth, which is vulnerable to supply additions. The filter runs a RevPAR decomposition using STR data: ΔRevPAR = OCC_contribution + ADR_contribution + interaction_term. Markets where ADR contribution exceeds 60% of total RevPAR growth pass; markets where RevPAR growth is primarily occupancy-led are downgraded. As of Q3 2026, Japan (+11.9% ADR YoY), Australia (ADR-led across Sydney, Melbourne, Brisbane), and South Korea (+15% arrivals driving ADR recovery) all pass this filter. Markets running occupancy-led RevPAR growth are more vulnerable to the supply additions in their pipeline.
Filter 4 is the Treaty and Structuring Efficiency filter. As a Singapore VCC, the fund requires that the target market falls within Singapore's Double Taxation Agreement network to minimize withholding friction on dividends and interest repatriated from hotel-owning SPVs. Japan, Australia, India, and Thailand all have Singapore DTAs providing reduced withholding rates -- 15% on dividends, 10-15% on interest, compared to standard rates that can reach 20-25%. Markets outside the DTA network face a capital efficiency penalty that typically overwhelms the investment case in gateway-quality markets where yields are already compressed.
Once a market passes all four filters, individual assets are scored across five dimensions that feed into the Bay Score -- Bay Street's composite deal quality metric calculated in the Bay Terminal portfolio analytics platform.
Market Quality score covers RevPAR growth spread (from Filter 1), supply pipeline concentration (number of rooms under construction within 2km as a percentage of existing market supply), and exit liquidity depth (trailing 3-year transaction volume and buyer depth from RCA data). Asset Quality score covers brand tier and positioning, physical condition from an independent Property Condition Assessment, and GRESB pre-assessment score. Financial Quality score covers going-in yield versus WACC spread (from Filter 2), the Liquidity Stress Delta, and leverage ratio against the fund's target LTV range. Operator Quality score covers HMA commercial terms achieved versus the five non-negotiable targets (base fee profitability linkage, incentive fee disallowed items schedule, owner priority return hurdle, performance test enforceability, term flexibility). Structural Quality score covers DTA efficiency (from Filter 4), FIRB or equivalent foreign ownership compliance, and title clarity.
Each dimension is scored 0-100 with weights calibrated annually against post-acquisition performance data from the live portfolio. A Bay Score above 75 indicates above-average quality across all five dimensions. Deals with Bay Scores below 55 on multiple dimensions require explicit IC discussion of why the quantitative flags are wrong before capital is approved -- the Bay Score does not veto deals, but it does require the IC to articulate a counter-thesis when the numbers signal caution.
The income approach is primary for all hotel underwriting. Direct capitalization divides stabilized NOI by the going-in cap rate: NOI / Cap Rate = Value. DCF modeling sums discounted annual NOI over the hold period plus a discounted terminal value, where the terminal value applies an exit cap rate to Year N+1 NOI. Exit cap rate assumption is critical: the base case applies 25-50 bps widening for core APAC gateway assets (reflecting cap rate reversion risk as monetary policy normalizes) and 50-100 bps widening for value-add positions (reflecting the added uncertainty of operational repositioning).
Key DCF inputs run from STR/Tourism Economics RevPAR growth forecasts (3.6% APAC 2026, 2.5% 2027 per the consensus), the fund's expense ratio stabilization timeline (typically 18-24 months post-acquisition before management efficiency gains are fully reflected), a CapEx schedule derived from the PIP and Property Condition Assessment, and the WACC derived from the fund's debt terms and equity return target. Every IC submission includes a full sensitivity grid: exit cap rate in five steps (base, +50, +100, +150 bps, -50 bps), RevPAR CAGR in four steps (base, +2%, -1%, -3%), and hold period in three steps (base, +1 year, +2 years). The sensitivity matrix makes visible which assumptions the IRR is most sensitive to -- typically exit cap rate and hold period for value-add positions, and RevPAR growth for core-plus positions.
The market approach provides secondary validation. Price per key benchmarks by segment and geography confirm that the DCF output is within a reasonable range of comparable transaction pricing. EV/EBITDA multiples provide a cross-check against public market comparables -- Singapore-listed hotel REITs, Ashford Hospitality Trust, and park REITs in Japan. For APAC hotel PE acquisitions, PPK and EV/EBITDA multiples serve primarily as sanity checks rather than primary valuation inputs, because hotel operating performance varies too widely across management quality and physical condition to make raw comparable transaction pricing reliable without adjustment.
Target leverage is 50-65% LTV for value-add strategies and 40-55% for opportunistic strategies, reflecting the higher operating risk profile of opportunistic deals and the need for equity cushion against potential NOI shortfalls during repositioning. Senior debt for core APAC markets currently prices at SOFR plus 200-350 bps; bridge and construction debt for value-add and development positions runs SOFR plus 300-500 bps. Mezzanine is available at 12-20% for assets with strong going-in cash flow but elevated reposition risk.
Capital structure decisions interact directly with the LSD calculation. A deal with high LSD (above 15%) -- where a 12-month exit delay causes meaningful IRR erosion -- requires structural reinforcement: lower leverage reduces debt maturity pressure that could force a sale at an inopportune time, higher equity cushion prevents covenant breach during a RevPAR stress event, and a longer original hold period makes a 12-month delay proportionally smaller as a share of total hold. The LSD alert in the Bay Terminal flags live portfolio assets that are approaching timing-dependent risk thresholds, providing early warning before exit decision-making.
DSCR covenant management requires maintaining a minimum 1.25x coverage ratio, stress-tested at -20% RevPAR to confirm the asset can service debt through a moderate operating stress event without triggering covenant breach. For Japan assets with yen-denominated debt against Singapore dollar LP distributions, an FX sensitivity layer is added: the base case models currency at current rates, with a 15% yen depreciation and a 10% yen appreciation scenario included in the IC submission.
ESG integration in hotel underwriting is a valuation discipline, not a compliance exercise. The GRESB pre-assessment is commissioned at the LOI stage -- before full diligence, and before price is locked. An asset scoring below 40 out of 100 on GRESB is in the bottom quartile of its global peer group. Any institutional buyer at exit who requires GRESB participation -- and this now includes the majority of European institutional LPs and a growing share of APAC sovereign wealth funds -- will either discount the asset significantly or decline to participate in the exit process. The acquisition price must reflect this exit discount: typically 3-5% of asset value, reflecting the CapEx and management time required to bring a below-40 scorer to peer-average over a 2-3 year post-acquisition period.
Energy intensity is benchmarked against IHG and Marriott published standards: approximately 250-400 kWh per square meter per year for upper-upscale APAC hotels. Assets running above this range face energy infrastructure CapEx in year 1-2 that must be modeled explicitly. SFDR Article 8 alignment is required for European-domiciled LP entities investing through the VCC -- this shapes which assets can receive European LP co-investment capital and affects the LP base available for each sub-fund.
Target returns are strategy-differentiated and reflect the risk profile of each approach. Core-plus strategies target 8-11% net IRR with equity multiples of 1.5-1.8x over 5-7 year holds, typically applied to stabilized gateway assets in Japan, Singapore, and Australia with strong institutional exit liquidity. Value-add strategies target 12-16% net IRR with 1.8-2.3x equity multiples over 4-6 year holds, applied to assets with clear operational gap -- RevPAR index below 85, management underperformance, deferred CapEx, or brand repositioning opportunity. Opportunistic strategies target 18-22% net IRR with 2.0-3.0x equity multiples over 3-5 year holds, applied to distressed situations, CMBS maturity wall acquisitions, and development plays in under-supplied gateway submarkets.
Vintage year context matters for benchmarking. The 2021-2022 hotel PE vintage produced median net IRR of 7.4% per Preqin's 2026 analysis -- significantly below target -- reflecting elevated acquisition prices ahead of rate hikes and COVID-era recovery uncertainty. The 2023-2025 vintage is improving as cap rate expansion has created better entry points, particularly in markets with forced sellers. The 2026 vintage environment is shaped by the USD 48 billion CMBS maturity wall in the US and European markets, which is creating opportunistic acquisition opportunities as overleveraged assets seek exits, alongside APAC's continued institutional capital deepening as JLL forecasts USD 13.3 billion in APAC hotel transaction volume for 2026.
Post-acquisition monitoring runs a standardized monthly KPI dashboard through the Bay Terminal. RevPAR Index versus the competitive set is tracked weekly via STR, with alert thresholds set at 85 (yellow flag requiring management discussion) and 75 (red flag requiring operator review). GOP margin is tracked monthly against the proforma budget, with a -300 bps variance triggering a management call and -500 bps triggering formal operator review. DSCR is calculated monthly and covenant-tested against the 1.25x minimum. CapEx spend is tracked against PIP timeline milestones in Procore, with material slippage flagged for LP reporting.
The LSD alert system monitors live portfolio assets quarterly, recalculating the metric as exit market conditions evolve. An asset that had an LSD of 12% at acquisition can cross the 15% alert threshold if cap rates compress (improving exit pricing and reducing timing risk) or widen (increasing it). The Bay Score is updated semi-annually incorporating current performance data, GRESB scoring updates, and revised operator assessments -- providing the IC with an updated deal quality composite that reflects actual performance against underwriting.
Why does the framework run market filters before asset scoring?
Because market quality is the primary driver of long-run hotel performance, and no amount of asset-level excellence can overcome a structurally unfavorable market backdrop over a 5-7 year hold. The four filters eliminate markets where the quantitative conditions for a successful investment -- RevPAR growth exceeding supply, cap rate above cost of capital, ADR-led pricing power, DTA efficiency -- do not exist. Asset-level work only begins when the market case is established, which prevents deal flow pressure from leading the team into compelling assets in structurally unfavorable markets.
What does the LSD metric tell you that standard IRR sensitivity analysis does not?
Standard IRR sensitivity analysis assumes you can choose your exit timing. The LSD metric quantifies what happens when you cannot -- when market illiquidity, debt maturity pressure, or adverse operating conditions force a 12-month exit delay. A deal with a base case 15% IRR and a 19% LSD is more dangerous than a deal with a 13% base case IRR and a 9% LSD, even though the first deal has a higher headline return. The LSD forces the IC to confront timing-dependent risk explicitly rather than treating it as a tail scenario.
How does five-bucket demand segmentation change the underwriting?
Demand mix determines both the revenue ceiling and the operational risk profile. An OTA-heavy hotel (40% of revenue through third-party channels) carries structurally higher distribution cost, lower net ADR, and greater rate opacity than a direct-booking hotel at the same gross RevPAR. This changes the NOI underwriting, the management fee structure negotiation (disallowed items must explicitly exclude OTA commissions to prevent incentive fee calculation distortion), and the value-add thesis (direct booking conversion is a recurring operational improvement opportunity that can be sized and tracked). Group-heavy hotels have different forward booking curve dynamics and GOP risk profiles than transient-led hotels. Segmentation analysis is not a reporting exercise -- it drives specific IC decisions on proforma assumptions and HMA terms.
What makes the ADR-vs-occupancy filter more important than RevPAR growth alone?
Two hotels can both report 8% RevPAR growth. One achieved it by running 90% occupancy versus 83% prior year while holding rates flat. The other achieved it by raising ADR 9% while holding occupancy flat. The first hotel's RevPAR gain is fully exposed to supply additions -- every new competitor distributes occupancy across more rooms. The second hotel's gain reflects pricing power -- the market's willingness to pay more, which a new competitor entering at lower rates does not immediately reverse. Over a 5-year hold that includes a supply addition cycle, the ADR-led hotel produces more durable NOI growth and a more defensible exit multiple.
How does the quantamental framework differ from a standard PE diligence process?
A standard PE diligence process is deal-specific: it assembles data for the asset under consideration and produces an analysis for that asset. The quantamental framework is systematic: the same screens, the same decompositions, the same scoring dimensions, and the same sensitivity structure run for every deal in the pipeline. This systematization produces two benefits. First, it enables true portfolio construction -- because every deal runs the same framework, Bay Scores and LSD metrics are comparable across assets in different markets and strategies, enabling the portfolio to be assembled with conscious risk balancing. Second, it creates institutional memory: when a deal that scored 82 on Bay Score underperforms or outperforms its proforma, the portfolio data develops the empirical basis for refining the weights and thresholds in the next vintage cycle.
Bay Street Hospitality is a Singapore-domiciled hospitality private equity fund operating under the Variable Capital Company (VCC) framework, regulated by the Monetary Authority of Singapore. We invest in upper-upscale and luxury hotel assets across Asia-Pacific, deploying capital through a multi-sub-fund VCC structure designed to maximize treaty efficiency and ring-fence risk across geographies. We have publicly stated a 2032 SGX listing target.
This content is for informational purposes only and does not constitute investment advice, an offer to sell, or a solicitation of an offer to buy any securities or fund interests. Past performance is not indicative of future results. All investment involves risk, including the potential loss of principal. Prospective investors should conduct their own due diligence and consult their own legal, tax and financial advisors before making any investment decision.
...
Bay Street Hospitality is a diversified hotel fund series for allocators seeking quant-driven exposure across top countries, operators, and hotel tiers. The fund targets a diversified portfolio in both high-growth emerging markets and stable developed markets, optimizing risk-adjusted returns. The leadership provides structured equity and pivotal strategic guidance to hotel-related investments. The fund establishes joint ventures with best-in-class partners, focusing on deal flexibility, currency considerations, tax efficiency, and clear exit strategies.
