BRRRR Method: The Valuation Decides
The BRRRR method is buy, rehabilitate, rent, refinance, repeat: improve a property, let it, then remortgage against the higher valuation to recover the deposit and buy again. It works only if the valuation comes in high enough, and a surveyor decides that figure.
How it works
Buy, fix, let, refinance, repeat. The BRRRR method — buy, rehabilitate, rent, refinance, repeat — runs one property through five steps, then starts again with the same money.
The point is getting the deposit back out. You buy below what the property could be worth, improve it, let it, then remortgage against the higher valuation and take the deposit back. If it works, the same trading capital buys several properties.
Which only works if the work raises the valuation. The strategy is meant to pay for forced appreciation — value you created by fixing something — not for a rising market carrying you along while you decorate. Many published examples quietly had both.
And a surveyor decides that, not your spreadsheet. The refinance depends on a valuation carried out by someone with no interest in your projections, and a figure below your model stops the cycle where it stands.
The debt underneath it
Each cycle adds debt on top of the last one. Nothing is repaid at a refinance; the loan is replaced by a larger one, and the next property arrives with its own mortgage beside it.
So a rate rise hits every property at once. A portfolio built this way is leverage applied several times over, and the borrowings reprice together rather than one at a time. A single unmortgaged property has no such sensitivity.
And refurbishments run over time and over budget. Both cost money. The overspend takes it directly, and the delay takes it through carrying costs on the finance while the property sits unlet.
Each refinance pays another set of fees. Arrangement, valuation and legal costs arrive with every remortgage, so the frictional cost of the strategy is proportional to how often it cycles.
In practice
None of this has a market price until you sell. A house has no order book and no volume, so its worth between valuations is an opinion, and liquidity exists only when a buyer does.
One cycle takes the best part of a year. Purchase, works, letting, then a wait before a lender will refinance against the new figure. Feedback on whether the plan was sound arrives very slowly.
And lending criteria can change mid-project. What a lender would advance when you bought is no promise about what it will advance later, and like an opening gap the change lands between one decision and the next.
The exit is a sale that takes months to arrange. There is no stop loss on a house, so risk management here is the cash reserve that carries an unlet, unrefinanced property.
Every transaction costs far more than 2% of a bar. On the 576-bar shared series a round trip costs 2% of a median bar’s range and 45% of the smallest. A property costs a large multiple of that, and this strategy pays it again every cycle.
Testing the deal at a lower valuation
Model the refinance at a figure below the one you expect. Take your estimated post-works valuation, cut it, and run the whole plan again on that number. The deal is obviously worse; that is not the question.
The question is whether the position is survivable. If the surveyor comes in low the capital stays trapped, so name what you own in that case: a let property with a mortgage on it and no deposit released for the next purchase.
Then ask how long you could hold it in that state. Cover the payments from rent with a vacancy allowance, and see whether the reserve survives an empty month and a repair without forcing a sale.
And decide the response in advance, not in the moment. Carry on, refinance later, or sell. Each is a legitimate answer, and choosing before the valuation arrives removes the pressure to hope.
What the BRRRR method is not
It is not a way to buy property without capital. The deposit goes in first, and only sometimes comes back.
It is not house hacking. Living in the building changes the borrowing and the arithmetic.
It is not a passive holding. Every cycle is a building project with a deadline attached.
And it is not a bet on the market. It should pay for the work, not for the year you bought in.
When it fails
In a flat market the refinance simply does not come. With prices in a trading range the only value available is what the work added, and if that is smaller than the cost of the cycle the deposit stays where it is.
The first failure is a valuation below the model. Nothing is wrong with the property. The money is simply still inside it, and the next purchase does not happen.
A second is the refurbishment that overruns. Carrying costs run on an unlet property while the works drag, and the overspend comes out of the deposit that was meant to be recycled.
A third is a change in lending criteria mid-cycle. The decision belongs to someone else, made on their timetable, under rules you never agreed to.
A fourth is a rate rise across a stack of loans. Every property reprices at once, which is exactly what stacking borrowings was always going to do.
And a fifth is mistaking a rising market for skill. On the shared series 95% of bars sat below a prior peak, the longest run 73 bars — a drawdown a property endures without ever printing a chart.
The original data
The audience is enormous and the counterweight is missing. In research/broker-coverage.json, a
scan of the 24,971 videos in research/search-study-corpus.jsonl, 5 videos carry “brrrr” in the title
at a median of 242,150 views across 4 channels, the largest at 425,597 — among the highest medians on
this site. Zero titles in 24,971 mention risk of ruin. A strategy whose entire mechanism is repeatedly
re-levering the same capital is taught to enormous audiences, and the concept describing what happens
when re-levering goes wrong appears nowhere. For scale, 171 videos cover real estate at a median of
27,987 across 112 channels.
And the leverage arithmetic says plainly what stacking does. In
research/series-measurements.json, built by site/measure_series.py, with exposure reset every bar
before costs, the base series returned 3.61%; at two times exposure the actual return was 6.61%
against a naive 7.22%, with a maximum drawdown of 7.45%; at three times, 8.93% against a naive 10.82%,
with a drawdown of 11.08% against the base series’ 3.76%. The return rose by less than the multiple
while the deepest drawdown roughly tripled. Stacking borrowings across properties does the same thing,
to all of them at once. Run the whole plan at a valuation below your estimate before you buy, and
proceed only if that version still works.
Related
Rental property is what each cycle actually leaves you holding, and its costs decide whether you can afford to wait. A mortgage is the instrument the strategy runs on, and its reset dates are where a stacked portfolio is most exposed. And real estate is the wider asset class, including the parts that never let, never refinance and never cycle.
The part I keep underrating is how much of a plan sits with somebody else. I can do the work, check the sums again, and still be waiting on a figure a stranger writes down in an afternoon. If it comes back under what I assumed, the plan does not bend, it stops. So I write the plan out at the number I am afraid of first, and see whether I can live in that version.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.