One Bad Loan Eats Ten Good Ones: The Math Behind Private Lending Credit Decisions

September 16, 2026.

The best outcome a lender ever gets is their own money back, plus interest. The worst is losing all of it.

Every lender decision that has ever frustrated you comes from that math. The ARV haircut. The slow no. The draw that took a week too long, the questions about your last three projects. Most originators spend their careers fighting it. The best ones learned to pitch to it.

The Math Nobody Runs

Take a lender writing a 12-month fix and flip loan at 10%. Best case, they earn 10 cents on the dollar. Worst case, they lose the whole dollar. One loan that goes to zero erases the profit from 10 that performed. Even a partial loss burns years of margin on a single file: a half-finished rehab, a foreclosure, a contractor who walked, then legal fees and carrying costs piling on top.

Now look at the borrower’s side of the same deal. Per ATTOM’s Q1 2026 U.S. Home Flipping Report, the typical flip grossed 25.4% on purchase price, and the investor put down a fraction of that price. Levered, their cash-on-cash return has no ceiling. If the project outperforms, they keep every dollar above the debt. The lender still gets 10 cents.

That’s the asymmetry. Investors underwrite what happens if everything goes right. Lenders underwrite what happens if everything goes wrong. Both are rational. Most originators just pitch one side’s game to the other side’s player.

The Market Is Doing the Lender’s Math For Them

Right now, that downside math is on every credit desk’s screen.

Start with those flip margins. The first-quarter figure was the first increase in nearly two years, and it still sits near the lowest level since 2008. And it’s gross, before rehab, financing, and selling costs. Thin investor margins mean a thin cushion between the lender’s collateral and a loss. Meanwhile, ATTOM’s Q1 2026 Foreclosure Market Report showed residential foreclosure filings up 26% year over year. Distress is rising in exactly the asset class these lenders hold.

There’s a second layer most borrowers never see. An Urban Institute study published in April 2026 counted more than $85 billion in residential transition loans originated in 2025, and documented how fast the sector is institutionalizing: securitizations, rating agencies, credit facilities. Your private lender’s money has its own lender. The flexibility ends where their capital partner’s math begins. When two lenders quote the same deal at 85% and 70% of cost, you’re watching two different balance sheets price the same downside.

Understand that, and lender behavior stops looking random. Predictable is something you can work with.

Why the No Comes Late

Every originator has watched a deal sail through the account executive, survive underwriting, and die at the credit desk. It feels like betrayal. It isn’t. The machine is working as designed.

Each layer of a lending shop exists to find the one scenario where the money doesn’t come back. The person you pitched wants the deal. They get paid on volume. The layers above them get paid on not losing. So the deeper your file travels, the more skeptical the reader, and the later the no arrives.

The ARV cut works the same way. The appraisal review found comps you didn’t lead with, and the leverage moved with the math. That doesn’t excuse lenders who retrade as a habit. But it tells you which cuts are signal and which are character, and it tells you the fastest way to prevent one: make sure nothing in your file surfaces late.

Pitch to the Downside

Here’s where most originators lose deals they should win. They pitch the upside: the ARV, the spread, the hot zip code, the comps. The lender cannot be paid on any of it. You’re selling equity returns to a party that only owns downside.

The credit desk is asking one question: how does the money come back when the plan breaks? Answer it before they ask, on page 1 of the package:

• Basis against sold comps, not listings
• Sponsor liquidity after closing, not just at application
• A budget with a real contingency instead of a round number
• A second exit: if the flip doesn’t sell, does it rent and refinance into a DSCR loan?

A deal with two exits is a different deal.

Most originators think they already do this because they know lenders are conservative. Knowing it isn’t the same as pitching to it. Pull your last five deal packages and count the pages spent on what happens if everything works versus what happens if it doesn’t. Every downside question you leave open, a more skeptical reader answers for you, and their version is always worse than yours. Write the story yourself before the credit desk writes it.

The Other Side of the Table

The lender across the table is running rational math: one bad loan eats 10 good ones, in a market that just reminded them the bad loan is real.

You can resent that math or you can use it. The originators who use it write tighter packages, get faster answers, and take fewer cuts, because the credit desk’s worst-case question was answered on page 1.

Learn the lender’s math. Then make it easy for them to say yes.