How to calculate ARV (and why the appraiser gets the last word)
Two investors tour the same tired 3-bed on the same Tuesday afternoon. One walks out with an after-repair value of $185,000. The other says $215,000.
Both of them will happily show you their math. Both of them used the exact same formula.
That should bother you, and by the end of this post it will, because ARV is the number every other number in a flip or a BRRRR hangs off. If you flip, ARV sets your maximum offer. If you BRRRR, ARV sets the size of your refinance, which decides how much of your cash comes back out and whether there is a next deal at all.
Get it right and the strategy works exactly as advertised. Get it 10% wrong and that error runs downstream through your offer, your loan, your cash left in, your cash-on-cash return, and absolutely nothing in the process stops it.
The formula you'll see everywhere is not a formula
Almost every guide opens with this:
ARV = current value + value added by renovations
It looks like arithmetic. It isn't. Neither term on the right is a number you can go look up.
"Current value" of a distressed house is itself an estimate. And "value added by renovations" is the entire question asked a second time, because a $40,000 rehab does not add $40,000 of value. Sometimes it adds $60,000. In an over-improved house on the wrong street, it adds $15,000.
Cost is NOT value. Write that one down somewhere you'll see it again!
The formula is circular, and circular formulas give you whatever you already believed. That is how our two investors got $30,000 apart with straight faces.
So what is ARV, really? Something much less clever: ARV is what renovated, comparable houses near yours have recently sold for. You don't derive it. You go find it. That's the comp method, it's the same method the appraiser will use on you later, and you may as well run it their way from the start.
How to calculate ARV in five steps
1. Pull sold comps, not listings
Asking prices are opinions. Closed sales are evidence.
Work from properties that actually sold, as recently as your market allows, ideally the last three to six months. This isn't just tidy practice, it's the lender standard: Fannie Mae's appraisal guidelines require at least three closed comparable sales, generally from within the last 12 months. Your ARV is eventually going to be graded against a number built exactly that way, so build yours the same.
2. Match the location tightly
Stay inside the same neighborhood. Ideally within half a mile, and NEVER across a boundary the market prices: a school-district line, a highway, the street where the housing stock changes decade.
A comp 0.4 miles away on the wrong side of the boulevard can be worth 20% less than one two blocks from your subject. Here's the quick test: if a local agent would wince at the comparison, the appraiser is going to wince at it too.
3. Match the finished product, not the current one
This is the step people miss, and it is the expensive one.
You are valuing the house after the rehab. So your comps have to be houses in after-rehab condition: renovated kitchens, updated systems, the finish level you are actually building to. Same property type, similar age and style, bed and bath count as close as you can get, square footage within roughly ±20%.
Comp your future renovated 3/2 against tired originals and you have not calculated your ARV. You have calculated what the house is worth if you do nothing at all.
4. Put them on a grid and let price per square foot do the work
Say your subject is a 3-bed / 2-bath, 1,400 sq ft house, renovated to neighborhood standard. You find three renovated solds nearby:
| Comp | Sold price | Sq ft | $/sq ft |
|---|---|---|---|
| A. 3/2, renovated, 0.3 mi | $209,000 | 1,450 | $144.14 |
| B. 3/2, renovated, 0.4 mi | $199,500 | 1,380 | $144.57 |
| C. 3/2, renovated, 0.5 mi | $192,000 | 1,310 | $146.56 |
Average: $145.09 per square foot. Applied to your subject's 1,400 sq ft: 1,400 × $145.09 ≈ $203,000.
Now adjust for the big, concrete differences. A garage your subject lacks. A third bathroom a comp has. A main road versus a quiet street. You adjust the comp's price toward what it would have sold for as your house's twin, then re-run the average. Keep the adjustments few and defensible, and remember the rule of thumb: if a comp needs five adjustments to fit, it was NEVER a comp.
One warning about price per square foot. It drifts upward as houses get smaller, because the lot, the kitchen, and the roof get spread across fewer feet. Comp a 1,000 sq ft house against 1,900 sq ft sales and the shortcut will flatter it badly. Keep your sizes in a tight band and it stays honest.
5. Take a range, not a point
Those three comps support a band of roughly $198,000 to $205,000. They do not support "$203,126," no matter what your calculator says.
State your ARV as the band. Underwrite the deal at the conservative end, which on this grid means we'd be running our numbers at $198,000 and treating anything above it as upside. Then gut-check it against the world: do renovated houses on that street genuinely change hands at that number?
Because if your ARV sits above every single sold comp you were able to find, you don't have an ARV. You have a hope.
Why you can't just use the Zestimate
Automated valuations estimate a house as it sits, from public records and market data. They cannot see a renovation you have not done yet, and they are weakest on exactly the kind of house a BRRRR investor buys.
Zillow's own published accuracy data puts the Zestimate's median error around 2% for homes actively on the market, but around 7% for off-market homes. A distressed acquisition is an off-market home. And "median" means half the misses are BIGGER than that.
On a $200,000 ARV, 7% is $14,000. That is enough to flip the deal's verdict all by itself, before your rehab budget has said a single word.
So are AVMs useless? Not at all! We use them constantly for finding candidate comps fast and for spotting when an estimate has left the atmosphere. As the number you borrow against? No.
The appraiser gets the last word
Your ARV meets reality twice, and both times somebody else is holding the pen.
If you flip: the classic 70% rule sets your maximum offer at ARV × 70% − repair costs. On a $200,000 ARV with $35,000 of repairs, that's $140,000 − $35,000 = $105,000. Look at the leverage in that. Every dollar of ARV you are wrong about moves your maximum offer by 70 cents, so a 7% ARV miss is a $9,800 offer error on this one ordinary house.
If you BRRRR: the refinance lender orders an appraisal and lends a percentage of that number, typically 70 to 75% of appraised value, as covered in BRRRR basics. Watch what a modest miss does:
| You underwrote | Appraisal came in | |
|---|---|---|
| Value | $200,000 | $185,000 |
| Refi loan at 75% LTV | $150,000 | $138,750 |
That is a 7.5% appraisal miss. Right at the published median error for off-market estimates, and completely ordinary in the real world. It just shrank your cash-out by $11,250.
That money is not lost. It is BURIED. It's cash sitting in a house that was supposed to be funding your next purchase, and the "repeat" in BRRRR runs entirely on ARV being right.
Protecting the deal before you offer
You cannot control the appraisal. You can absolutely decide, in advance, how much appraisal risk this deal survives.
Underwrite at the low end of your comp band. Then ask one more question, and be honest with yourself about the answer: what does this deal look like if the appraisal lands at 90 to 92% of my number?
If the answer is "still fine, less cash back, but the property cash-flows and I can live with what's stuck in it," then congratulations, you have a genuinely resilient deal!
If the answer is "the entire thesis collapses," you don't have a deal. You have a bet on an appraiser's mood.
And keep your rehab scope matched to what the comps actually reward. The market that handed you your ARV was pricing a specific finish level, not the nicest house on the block.
Where BuyBox fits, and where it doesn't
Plainly, and up front: we do not calculate ARV for you. There's no AVM inside BuyBox and no comp-puller. After-repair value comes from the work described above, and software that pretends it can skip that work is how deals end up with a hope where the ARV should be.
What we do is everything downstream of the number. Enter your ARV and a Deep Dive models the whole BRRRR sequence: refinance size at your lender's LTV, cash back at refi, cash left in the deal, post-refi cash flow and cash-on-cash. Every figure comes from one engine, audited against industry-standard real-estate math, and every number shows its formula when you hover it.
Because a re-run takes seconds, the stress test stops being a spreadsheet chore. Run the deal at $205,000. Then at $198,000. Then at the appraisal-misses-by-7% number, and watch which verdicts are still standing. That question, which ARV does this deal actually need?, is worth more than any point estimate you will ever produce.
Frequently asked
Is ARV the same as the appraised value?
No. ARV is your pre-purchase estimate of post-renovation value. The appraised value is a licensed appraiser's opinion, built from closed comps, after the work is finished. The lender uses theirs, not yours. Your job is to estimate it the way they will compute it, then leave yourself margin for the difference.
What if I can't find renovated comps?
Treat that as a finding, not an inconvenience! If nothing renovated has sold nearby in the past year, the market is telling you something important: it has not priced the product you are planning to build, so your renovation premium is speculative. Widen the time window before you widen the map. If the comps still aren't there, underwrite as though the premium is small. Sometimes the honest ARV answer is "this neighborhood does not pay for granite."
How many comps do I need?
Three solid ones beat eight loose ones. Three closed comparable sales is also the minimum a lender's appraiser must report under Fannie Mae's guidelines. If your three need heavy adjustments to fit, keep hunting.