The formula
Every risk score on this site is one calculation. It takes what the thing costs each month, adds whatever monthly debt you already carry, divides that by your monthly income, and scales the result by your credit score.
That ratio is the entire model. Everything below is just the detail of how each of the four inputs is worked out. Nothing here comes from a credit bureau, a lender, or any outside service. It is computed in your browser from what you type in.
The four thresholds
The ratio lands in one of four bands. These cutoffs are fixed and they are the same on every calculator on the site.
| Risk | Share of income | What it means |
|---|---|---|
| Low | 10% or less | Fits comfortably. Room to take this on without reshaping anything else. |
| Moderate | 10% to 20% | Workable. Noticeable in a monthly budget, but not a squeeze. |
| Elevated | 20% to 36% | Tight. Doable, but it leaves less margin for anything unexpected. |
| High | Above 36% | A stretch. Worth a second look at the price, the terms, or the timing. |
The percentage shown next to the risk bar is that ratio, rounded and capped at 100%. A high reading is a prompt to look again at the price, the term, or the timing - not a verdict on you.
One-time costs versus monthly ones
A recurring cost - rent, a loan payment, a subscription - is compared directly against one month of income, because that is exactly how it lands. A one-time cost is not a recurring commitment, so comparing a $6,000 vacation against a single month of income would overstate it badly. Instead a one-time cost is spread across three months of budget before the ratio is taken. That $6,000 vacation is weighed as $2,000 a month, not $6,000.
The credit-score adjustment
Your credit score does not gate the result and it never blocks anything. It scales the ratio, because the same payment carries genuinely different risk depending on the terms you can get. Strong credit pulls the number down; weaker credit pushes it up.
| FICO band | Score | Multiplier |
|---|---|---|
| Exceptional | 800 and up | ×0.85 |
| Very good | 740 to 799 | ×0.92 |
| Good | 670 to 739 | ×1.00 |
| Fair | 580 to 669 | ×1.15 |
| Poor | Below 580 | ×1.35 |
Leaving your score blank is treated as neutral - the multiplier is exactly 1.00 and the ratio is untouched. For a household with more than one income, the model uses the lowest score entered, not the average, because that is how a joint application actually qualifies. It is the weaker profile that sets the terms.
Which income figure is used
By default the model uses gross household income: every income entry added together, divided by twelve. That is the figure most lenders quote against, which makes it the most comparable starting point.
You can switch to estimated take-home pay instead. That applies federal brackets and the standard deduction, Social Security and Medicare, and an approximate state rate matched to your ZIP code, along with the Child Tax Credit if you have entered dependents. It is an estimate, not a paystub - it knows nothing about your 401(k) contributions, your health premiums, or anything else withheld before you see the money.
A worked example
Take a $50,000 car with 10% down, financed at 7% over five years, against a $60,000 income with a 700 credit score and no existing debt.
- The loan payment works out to roughly $891 a month.
- Monthly income is $60,000 ÷ 12, or $5,000.
- There is no existing debt to add, so the cost stands alone.
- $891 ÷ $5,000 is 0.178.
- A 700 score is the “good” band, so the multiplier is 1.00 and the ratio stays 0.178.
- 0.178 is above 0.10 and below 0.20, which lands in moderate - shown as 18%.
The same car against the same income at a 620 score becomes 0.178 × 1.15 = 0.205, which tips into elevated. Nothing about the car changed; the terms did.
What this model deliberately does not do
Being clear about the edges matters more than looking thorough. This model:
- Does not know your savings. A large emergency fund makes the same purchase meaningfully safer than this score suggests, and the score will not reflect that.
- Does not model your actual spending. Two people with identical incomes and wildly different fixed costs get the same reading, unless the difference is entered as debt.
- Does not know your job security, whether your income is seasonal, or whether a raise is coming.
- Does not price the asset. It says nothing about whether the thing is worth buying, only about whether the payment fits.
- Does not pull a credit report. The score you type is taken at face value, and typing one has no effect on your real credit.
- Uses thresholds that are a convention, not a law. The 10 / 20 / 36 cutoffs are in the same family as the debt-to-income ratios lenders work with, but they are our choice. A different set of cutoffs would give different labels for the same math.
This is not financial advice, a loan offer, or a credit decision. For a real decision, talk to a licensed advisor or your lender.
Where the numbers come from
The figures above are about your inputs. Separately, every calculator ships default values - interest rates, typical prices, tax rates, cost-of-living indices - so a page is useful before you have typed anything. Each one is a hand-set snapshot read from a named public source, not a live feed, and each is a starting point to adjust rather than a quote.
| Figure | Source | Verified |
|---|---|---|
| Default interest rates by loan kind | Mortgage from Freddie Mac's PMMS, new auto from Bankrate, used auto from Experian, personal and HELOC from Bankrate, payday from the CFPB. | August 15, 2026 |
| Federal tax brackets, standard deduction, FICA | IRS Revenue Procedure 2025-32 for tax year 2026, plus the SSA's October 2025 wage-base announcement. Married Filing Separately is derived from Married Filing Jointly by the standard IRS halving rule, since no primary source publishes it separately. | August 19, 2026 |
| Median income by age | The Bureau of Labor Statistics' usual weekly earnings release, Q2 2026. | August 15, 2026 |
| Credit score by age | Experian's State of Credit, 2025 edition - confirmed still current at the last check. | August 15, 2026 |
| Cost of living by state | The Bureau of Economic Analysis' Regional Price Parities, 2024 data, read off News Release BEA 26-10. | August 15, 2026 |
| State income tax rates | Approximate flat rates for all 51 jurisdictions, checked against the Tax Foundation's 2026 tables. Thirteen states changed at the last review. | August 15, 2026 |
| Anchor prices for houses, apartments, and cars | Redfin's median sale price (Q2 2026), RentCafe and Yardi Matrix's national average rent, and Kelley Blue Book's average transaction price. | August 15, 2026 |
| Vehicle sales tax, doc fees, lease defaults | State revenue and motor vehicle departments cross-checked against the Tax Foundation, Avalara, and Policygenius. Doc fee from CarEdge's 2026 survey; lease money factor and residual from Edmunds and Experian. | August 17, 2026 |
| Property tax and insurance rates | Market averages, reviewed against current auto and homeowners cost data. | August 15, 2026 |
| ZIP code to state lookup | GeoNames. | August 15, 2026 |
“Verified” means someone read the source on that date and confirmed the figure in the code still matched it - not that the source itself was published then. Where a check found no change, the date still moves, because the check happened.
Corrections
If a number here is wrong, or a source has moved on and we have not caught it, say so and it gets fixed. Write to contact@caniafford.it or use the contact page. The full terms are in the Terms of Service.