This is a hypothetical composite built for education. The properties, numbers, people, and decision are invented. They do not describe a current opportunity or anyone’s actual transaction, and they are not a recommendation or offer.
On Friday afternoon, an investor opens two underwriting folders. Deal A leads with the higher projected internal rate of return. Deal B’s projected return is lower. Both are model outputs in this fictional scenario, not forecasts.
“A,” somebody says. “Why are we still looking?”
Because the larger output is the end of the formula, not the beginning of the decision.
The investor does not stop there.
Both properties are ordinary suburban apartments. Both need work. Both could perform. But projected return is the last line in a long chain of assumptions. The useful question is not, “Which number is bigger?” It is, “How many things must go right to produce that number?”
The headline gets one minute
The investor gives the projected returns one minute, then puts both deals into the same table. No sponsor adjectives. No renovation photographs. Just claims that can either survive a document check or fail one.
| Item | Deal A | Deal B |
|---|---|---|
| Units | 48 | 44 |
| Purchase price | $8.64 million | $8.14 million |
| Renovation budget | $1.20 million | $396,000 |
| Debt | 75% of cost, floating | 65% of price, fixed |
| Starting debt-service coverage | 1.08x | 1.43x |
| Occupancy | 89% | 95% |
| Cash reserve at closing | $150,000 | $300,000 |
| Required refinance | Year 3 | None |
Deal A plans to renovate 36 units at about $25,000 each, raise monthly rent by $275, and fill the remaining vacancy. It also assumes 3.5% annual market rent growth. The debt floats at the benchmark rate plus 3.50%, and the model needs a refinance in year three to return capital and replace the expensive loan.
Deal B plans lighter work on 22 units at about $9,000 each. Its modeled rent increase is $140. The seven-year loan is fixed at 6.15%, and the plan assumes 2.5% annual rent growth. Less upside. Fewer moving parts.
Three documents interrupt Deal A
At first, A’s larger rent gap seems to justify the extra risk. Then three source documents arrive.
The signed lender quote requires an interest-rate cap that costs $95,000. The model contains no line for it.
“Add it.”
The insurance quote is $113,000 a year, not the broker package’s $82,000 estimate. That is another $31,000 of annual expense.
“Add that too.”
The rent-comparable file includes four renovated units advertised at premiums near $275, but the executed leases show an average achieved premium of $178 after concessions.
The analyst updates the last input. Asking rent had been representing the property. The signed leases now take over.
After those corrections, Deal A’s first-year coverage falls from 1.08x to 0.94x in the composite model. In plain English, property income no longer covers scheduled debt service. The $150,000 reserve looks large until the revised downside case burns roughly $39,000 a month during renovation and lease-up. It then represents less than four months of breathing room.
Deal B has problems too. Its plumbing bid is $42,000 above the seller’s estimate, and two leases expire earlier than the rent roll summary suggests. But the revised coverage remains 1.31x, the reserve still covers the planned work, and no refinance is required for the base case to function.
The projected-return ranking flips. Nobody discovered a bad property. The room finally met an honest model.
Count dependencies, not excitement
The fictional investor advances Deal B to legal review and stops spending diligence money on Deal A. The decision is not based on fear of renovation or a belief that fixed debt is always superior.
It comes down to dependency count.
Deal A needs renovation timing, achieved rent premiums, occupancy recovery, floating-rate control, and a receptive refinance market to cooperate. A delay in one row pressures the next row. Lower rent slows lease-up. Slower lease-up weakens coverage. Weak coverage consumes reserves. Thin reserves make the refinance more important precisely when the lender has more reason to be cautious.
Deal B mainly needs ordinary collections, controlled repairs, and modest lease execution. Its lower projected return is payment for accepting less modeled upside. Its stronger reserve and fixed debt buy time when the plan misses.
Portfolio context matters too. If this investor already owns two floating-rate renovation deals, Deal A adds the same risk in another address. Diversification is not owning three properties whose models all need cheap debt in year three.
The claims bring their source files
The broker package starts the conversation. It does not settle it.
- Executed leases and the current rent roll show occupied units, legal rent, concessions, expirations, and delinquency. They test whether the claimed rent gap is real.
- The T-12 general ledger and bank deposits test whether collected income resembles reported income. Scheduled rent is not cash.
- The signed lender quote and rate-cap proposal prove pricing, term, covenants, required reserves, and extension conditions. The debt row should match these documents exactly.
- The carrier’s insurance quote and current tax bill replace estimates with actual known costs. A percentage in a model is not a quote.
- Contractor bids with unit counts and scopes test whether the renovation budget includes labor, materials, permits, and contingency.
If a number cannot be traced, mark it as an assumption. Do not quietly promote it to fact because the spreadsheet formats it as currency.
The lower output survives the meeting
The hard lesson is not “choose the lower return.” That would be another lazy rule.
The lesson is that a return projection is a summary of dependencies. Deal A’s headline output was more fragile because several unsupported assumptions leaned on one another. Deal B’s lower output survived more document corrections and did not need a specific capital-market event on a specific date.
Before comparing two opportunities, normalize five rows: actual basis including required work, debt terms, verified rent evidence, cash reserves, and the downside case without a refinance. Then write one sentence naming what must go right.
The transferable rule is to compare how many unsupported conditions each projected return needs. If the sentence explaining the plan contains renovation and rent premiums and occupancy recovery and floating-rate control and a year-three refinance, the larger result may just be showing you the price of more ways to be wrong.
PR Steinfurth Equity provides educational information only. Nothing on this website is an offer to sell or a solicitation of an offer to buy any security, nor investment, legal, or tax advice. Any securities offering is made only to qualified investors through official offering documents. Real estate investments involve risk, including possible loss of principal. Past performance is not indicative of future results.