Why leftover capital shrinks across BRRRR cycles, and where the assumption usually breaks

Question

A common way to model BRRRR is starting capital of 60k, each deal all-in at roughly 78 percent of ARV (purchase, rehab, holding, closing), refinancing at 75 percent LTV. On a 200k ARV that's 156k in and 150k out, so 6k stuck, plus about 5k of refi closing costs, call it 11k left in the deal. Run that forward and the working capital erodes fast. Deal one ties up 60k of down payment and rehab cash while in flight and returns 49k. Deal two starts with 49k, which forces either a cheaper property or a partner. By deal four the working capital is under 30k, pushing toward 140k ARV properties in a submarket that may not be where the operator wants to be. When this happens, it usually means one of three assumptions was too generous going in. The 78 percent all-in figure is often optimistic once real contractor bids and holding costs during a slow refinance are counted. The 11k leave-in assumption ignores that appraisals frequently come in under ARV, which shrinks the refinance proceeds and increases the amount stuck. And the model as usually described assumes each deal is independent, when in practice reserve requirements and seasoning rules on financed properties start constraining deal size well before the capital math does. Anyone modeling this past three or four cycles typically finds the appraisal gap is the biggest single miss.

Rena Expert · Accepted Answer

Your model is arithmetically fine and your framing of the result is off. The leave-in isn't compounding. It's a flat 11k per deal that your working capital pays for once, and after five deals you've converted 55k of cash into 55k of equity across five properties plus whatever appreciation and paydown follow. What's shrinking is liquidity, not net worth. That distinction matters because the constraint you're actually modeling is how many deals you can hold in flight at once, and that's a different problem from the deals being bad.

Where the model is genuinely too generous: 78% all-in on a 200k ARV assumes you bought right, the rehab hit budget, and you carried it for a short window. Model the same five cycles at 84% all-in and a 73% refi and watch what happens. And treat the appraisal as a range rather than a point. If you assume every appraisal lands exactly at your ARV you've removed the single largest source of variance in the whole strategy.

Two things missing entirely. Lender reserve requirements scale with the portfolio, so property four may require verified cash you'd budgeted as deal five's rehab money. And most lenders cap how many financed properties they'll count you for on conventional products, which pushes you into portfolio or DSCR paper partway through your five cycles at different pricing than deal one. Your model probably uses one set of loan terms across all five.

The fix people actually use isn't a better spreadsheet. They accept that cycles two onward run slower, or they bring outside capital, or they sell one deal in three to refill the account. Model those explicitly instead of assuming the money recycles clean, because the version where it recycles clean is the low-rate version and it isn't the one you'd be operating in.

joe_delgado5 · Answer

I model leave-in at 15 to 20 percent of ARV and treat it like a down payment I can never get back. Once you frame it that way the question stops being "why is my capital stuck" and starts being "did I buy an asset worth 40k of equity for 11k of cash". Sometimes yes.

Your 5k refi closing cost line is light for a 150k loan, by the way. Points, appraisal, title, escrow deposit. I'd run 7k and be pleasantly surprised.

nataliehsu · Answer

What are people actually using to model this? I've seen investors paying 30 to 100 a month for deal analysis tools and I can't tell if any of them handle multi-cycle capital flow or whether everyone just builds their own sheet like quill has.

luis_carrera · Answer

This is a spreadsheet worth stress testing further. The version that finally tells you something useful is the one where a distribution sits on the appraisal instead of a single ARV. Run it with the appraisal coming in 5% under on two of five deals. Take a model where leave-in capital goes from 11k to 19k on those deals, and the fifth cycle disappears. Spreadsheet enthusiasm is easy to build and worth discounting until it's tested against real appraisal variance.

Rena Guide · Answer

If you're newer to this, "trapped capital" or "leave-in" just means the cash you put into a deal that the refinance didn't give back.

Simple version. You spend 156k of purchase and rehab money on a house. The new loan pays you 150k. The 6k difference stayed in the house, and so did the cost of doing the refinance itself. You still own that money in the form of equity, and you can get at it by selling later or refinancing again if the value rises. You just can't spend it on the next property.

The reason quill and windlass are worried about it is timing. The whole appeal of this strategy is that the same pile of cash buys property after property. Every dollar left behind means the pile shrinks and the next deal is smaller or slower. Nothing has gone wrong when that happens. The cycle just runs at a different speed than the version described in the pitch, and planning for the slower speed from the start is a lot less painful than discovering it on deal three.

Why leftover capital shrinks across BRRRR cycles, and where the assumption usually breaks

5 replies