First five years: paper losses, near-neutral cash
Depreciation and mortgage interest create taxable losses that offset W-2 income. The cash the property needs from you after debt service and tax savings is much smaller.
Tax strategy: can the losses offset your W-2 income?
Only if the rental counts as an active business. The defaults assume all three conditions hold. Untick one to see the cash flow if losses are treated as passive and suspended. How this works
Scenario lab
Set a best guess and a low and high for each key input. The point estimate uses your best guess; the distribution draws each ticked input from a triangular range and reruns the full 30-year model.
What moves NPV most
NPV with each input at its low (red) and high (blue), the others held at best guess.
All assumptions
Show every input
Where each figure will come from
Detailed schedules
30-year pro forma, sale, equity
Sale in year
Equity view
NPV and IRR use project cash flows before financing, discounted at the after-tax WACC. The equity view shows the same deal from your down payment.
Methodology
How every number is produced
The question this tool answers
Would buying this property and running it as a short-term rental create more value than it costs? The tool projects every year's cash flow over the holding period, adds the after-tax proceeds from selling, and discounts everything back to today.
- NPV is the sum of those discounted cash flows minus the upfront cost. Positive means the property earns more than the discount rate.
- IRR is the discount rate at which NPV would be exactly zero. Compare it with the discount rate.
- Cap rate is year-1 net operating income divided by the purchase price, a quick yield measure that ignores financing and taxes.
- Cash flow to you is what actually lands in, or leaves, your bank account each year after the mortgage and your income tax savings.
- Net income is the taxable profit or loss reported on your return. Depreciation makes it much lower than cash flow.
NPV and IRR use project cash flows (before the mortgage), discounted at the after-tax weighted average cost of capital. The financing shows up in the discount rate rather than in the cash flows. The schedules also show an equity view based on your down payment.
Where inputs come from
You enter an address and, if you like, a price. Everything else is filled in from data services or reasoned defaults, and every input carries a tag naming its source. Blue tags are data services, green are public data, amber are estimates you should replace with quotes when you can, purple are your own settings, and grey are model defaults. You can override any input, and your tax profile is remembered on your device.
The table under All assumptions lists each source and how its figure is derived.
Revenue estimate: several sources, one footing
Revenue is the most important input and the least certain. No one has published an independent test showing which estimator is most accurate for a single address. AirDNA's well-known accuracy claim compares its national totals with Airbnb's reported results, not single properties, and reviewers report single-address errors of 15–30% in thin markets. So the tool doesn't rely on one source.
- Get two automatic estimates. BNBCalc and AirROI each look at comparable listings near the address with the same number of bedrooms. Each costs about $0.20 per address.
- Add free ones if you like. You can type in the yearly figure from AirDNA's free calculator, Rabbu, or AirROI's website. They're treated exactly like the automatic ones.
- Put them on the same footing. Services define revenue differently. BNBCalc and AirDNA include the cleaning fees guests pay; AirROI and Rabbu report nightly-rate revenue only. The model counts cleaning fees as revenue and cleaning as an expense, so the tool adds estimated cleaning-fee income where it's missing:
cleaning-fee income = (occupancy × 365 ÷ average stay) × guest cleaning feeThe defaults are a 3.5-night average stay and a $175 fee. You can change both in the site settings.
- Average them. The best guess is the simple average of the comparable figures. You can switch to any single source, or type your own number.
- Set the range from the disagreement. The scenario lab's low and high for revenue span all the estimates, plus a cushion of 10% of the average on each side for single-address error. When sources disagree a lot, the range widens on its own and the histogram shows it.
The average occupancy across sources also sets the number of cleanings per month: booked nights ÷ average stay ÷ 12.
Operating costs
Management fee is a share of all collected revenue. Cleaning, utilities, maintenance, HOA dues and insurance start at their year-1 amounts and grow at the expense growth rate. Property tax is the tax rate times the home's market value each year, with no homeowner exemption because this is a rental. Revenue grows at its own rate. Net operating income (NOI) is revenue minus these costs.
Booking platforms charge the host a fee: about 3% under Airbnb's split-fee setup, more if the host pays all fees, which is common with property managers. The revenue estimates are before this fee, so it's an expense here.
Closing costs (2% by default) and furnishing are paid at purchase. Closing costs are added to your cost basis; furnishing is depreciated over 5 years. A one-time major repair (by default $20,000 for a roof in year 15) is also included.
If you'd use the property for trips you'd otherwise pay lodging for, you can enter that saving. It's added to cash flow but isn't taxable. Many managers also ask for a reserve at the start; it's returned with interest when you sell.
Taxes, and whether losses can offset your W-2 income
The property is held in a single-member LLC, so its profit or loss flows onto your personal return at your marginal federal and state rates. In the early years depreciation and mortgage interest usually create a taxable loss even when cash flow is close to zero.
Rental losses are normally "passive" and can't offset wages. A short-term rental can escape that rule when all three of these are true, which is what the default settings assume:
- the average guest stay is 7 days or less;
- your personal use stays small (the default is 14 nights or less);
- you materially participate. One common test is 100+ hours a year and more hours than anyone else, including the property manager and cleaners. With full-service management this test is often hard to meet; the other route is 500+ hours.
If you untick any of the three, the tool treats losses as passive: they save no tax in the year they occur. To stay conservative it doesn't model the release of suspended losses when you sell.
Depreciation, cost segregation and bonus depreciation
Standard depreciation. Land isn't depreciated. The building is written off evenly over 27.5 years, with a half year in the last year. A major repair, such as a roof, starts its own 27.5-year schedule the year after it's done. Depreciation is a tax deduction, not a cash cost.
Cost segregation is an engineering study that splits the purchase price into parts with shorter tax lives: furniture, appliances, carpet and some finishes (5 or 7 years), and landscaping, driveways and fencing (15 years). For furnished short-term rentals, studies commonly reclassify 15–30% of the price. The tool's default is 25% of the building, with 60% of that in 5-year property and 40% in 15-year property. A residential study typically costs a few thousand dollars; the tool deducts the fee as a year-1 expense.
Bonus depreciation lets you deduct shorter-lived property in the year you buy it instead of spreading it out. Federal law now allows 100% bonus depreciation permanently for property acquired after January 19, 2025. For a $650,000 Island Park cabin, the two together turn a year-1 taxable loss of about $34K into one of about $158K, which saves far more federal tax.
Three things that cut the benefit back:
- Your state may not follow. Idaho doesn't allow bonus depreciation, so the tool takes regular depreciation on the reclassified parts when it figures state tax, unless you tick "State allows bonus".
- Some of it is paid back when you sell. Depreciation on reclassified items is generally "recaptured" at ordinary income rates rather than the 25% rate for buildings. The tool does this by default, which is the cautious choice. A seller can often assign little sale value to 30-year-old furniture, which shrinks this tax; untick the option to see that case.
- It only helps if the losses are active. If the three conditions above aren't met, the bigger loss is suspended like any other.
The net effect is mostly about timing: tax saved now is repaid partly at sale, but money saved now is worth more than money paid in 30 years. That's why NPV rises by less than the year-1 tax savings.
Simplifications: reclassified property is depreciated straight-line after the bonus portion (tax law uses faster schedules, which matters only when bonus is below 100%). Furnishings aren't replaced during the hold. Some advisers argue a rental with average stays of 7 days or less is nonresidential property depreciated over 39 years; enter 39 as the depreciable life to test that. Please check the specifics with a CPA.
Discount rate
The discount rate is the after-tax weighted average cost of capital:
With 80% debt at 7.25%, a 29.8% tax rate and a 7.5% cost of equity, that's 5.57%. A larger down payment raises the WACC because equity costs more than after-tax debt. You can override the rate under All assumptions.
Sale at the end of the hold
Sale price is the purchase price grown at the appreciation rate. Following the original workbook, it uses one fewer year of growth than the holding period (29 years for a 30-year hold), which is slightly conservative. Selling costs come off the top. The gain is taxed in layers: depreciation on reclassified items at ordinary rates (cost segregation only), other depreciation at 25%, and the rest at 15%. The working capital comes back with interest, and for the cash-to-you view any remaining loan balance is repaid.
Scenario lab
For each ticked input you give a low, a best guess and a high. The tool draws each input from a triangular distribution with those three points, reruns the full model 1,200 times and plots the results. "Chance NPV is positive" is the share of those runs above zero. Inputs are drawn independently, so in reality the spread could be wider if, say, low revenue and low appreciation tend to happen together. The bar chart of inputs moves one input at a time to its low and high while holding the others at best guess, which shows which assumptions matter most.
Checking the model
The calculations were checked line by line against a spreadsheet model of a $650,000 Island Park cabin (3422 Oakwood Dr.). With that spreadsheet's inputs, the tool reproduces its results to the dollar: NPV $75,956 and IRR 6.17%. That spreadsheet left out closing costs and platform fees, and it included $2,000 a year of lodging savings.
Where the spreadsheet's write-up and its numbers disagreed, the tool follows the numbers: insurance of $4,088.50 (the text said $4,246), and the roof left out of the cost basis at sale (there's a switch for this).
Limits
This is a decision aid, not tax, legal or investment advice. Estimates for a single address can be well off. Local short-term rental rules, permits, insurance availability and HOA restrictions aren't checked. The tax treatment is simplified to a single marginal rate and doesn't model the alternative minimum tax, the net investment income tax, or limits on excess business losses. Before you buy, check local short-term rental permits and HOA rules, get an insurance quote and a management quote, and review the tax plan with a CPA.