Your Pro Forma Says 5% Vacancy. The Data Says You've Already Spent It.

Every one of us has a vacancy assumption sitting in a spreadsheet somewhere.

For most investors it's 5%. Maybe 8%, if you were feeling cautious the day you built the model.

But almost nobody goes back and checks that number against what actually happens during a lease-up.

We did. The results weren't pretty. 😬

So let's break down what the leasing data says about that 5% vacancy assumption, and what we think belongs in that cell instead.

What 5% Actually Buys You

5% vacancy works out to roughly 18 days of empty house per year. 8% gives you 29 days.

That's your entire annual budget for turnover, make-ready, and finding the next tenant.

Now here's the reality.

What the data says:

In RentEngine's Q2 2026 leasing report, a national scattered-site dataset covering single-family and small multifamily, properties that leased with no price change sat about 20 days.

Properties that took one price reduction averaged 40.

Why this matters:

A single pricing correction, on the very first lease-up, costs you more vacancy than your whole year's budget allowed for.

No turnover. No eviction. No furnace replacement.

Just getting the rent wrong by a little.

How Often Does This Actually Happen?

More than you'd think.

What the data says:

• 36.9% of properties that rented needed at least 1 price reduction to get there
• The ones that had price cuts averaged 1.75 reductions each
• The median total reduction was $100, and that figure hasn't moved in any quarter since the report launched

Why this matters:

That last one is the part we find genuinely interesting.

Investors and managers across every market and every price band are missing by roughly the same amount, quarter after quarter.

That's not bad luck in one city. It's a consistent bias, and it points in one direction: up.

Here's how it compounds:

• 0 cuts: 20 days on market
• 1 cut: 40 days
• 2 cuts: 57 days
• 3 cuts: 70 days
• 4 cuts: 81 days
• 5 cuts: 91 days

Let's Run It on a Real Property

Say you own a single-family rental that comps at $1,400. You list it at $1,500, chasing an extra $1,200 over the year.

The listing sits. You eventually accept the market's answer and cut, and your vacancy runs 40 days instead of 20.

Those 20 extra days cost you about $920 in rent you'll never collect.

So you spent 77% of the upside you were reaching for before the tenant even signed. And you didn't get the $1,500 anyway, because you discounted it.

What if you land at $1,450? You're $50 a month ahead, and earning back that $920 takes over 18 months.

Meanwhile the carry costs don't pause while you wait. Debt service, taxes, insurance, utility minimums, lawn or snow depending on the season.

So Why Does Everyone Miss in the Same Direction?

Good question. And the answer isn't that we're all optimists.

(Although we are. 😇)

Why this is happening:

Nearly all of us price off listing data. Zillow, the MLS, or whichever aggregator you prefer.

All of it shows what an owner asked for. None of it shows what a tenant actually paid.

When a rental listing disappears, nothing tells you whether it leased at asking, leased $150 under, or got pulled and relisted three weeks later at a different number. All three look identical from the outside.

So the data you can see runs high by default. Price against it and you inherit everyone else's optimism, then add your own on top.

What you can do:

Use at least 3 independent sources before you settle on a number.

Active listings give you the competitive read. Paid institutional rent data gives you the real localized median. And skip Craigslist entirely, because the data there is too erratic to be worth anything.

We wrote up our full pricing process here if you want the details.

The Variable Nobody Puts in the Model

Condition.

Think about your underwriting for a second. It's got fields for beds, baths, square footage and ZIP.

Does it have one for "my kitchen is 10 years behind the 3 houses that just leased on this block"?

If not, it should.

What you can do:

Pull the interior photos of your active comps and put them side by side with yours.

If yours is clearly the older product, price 10% under what those comps are asking and plan around it.

If yours is the freshest thing on the street, that's the one situation where it's worth testing the top of the market.

It takes about 5 minutes and it's free.

How to Read the Market's Answer

Once your listing goes live, the market responds fast, and it tells you something specific.

What to look for: If people are finding your listing and nobody books a showing, the price is wrong for the condition. That signal usually shows up within 10 days.

But before you cut, check your traffic first.

What the data says: The report puts the median at 25 leads to lease a property, up from 22 the previous quarter.

If you're only pulling one or two inquiries a month, you can't read the funnel at all. There isn't enough there to read.

That's an exposure problem, not a pricing problem, and cutting the rent won't fix it.

For a sense of how wide that range gets: in our July 2025 Zillow reporting, the market average across Metro Detroit was running under 1 lead per listing per month. At that rate, 25 leads is a two-year wait.

So What Should Be in the Cell?

Back to where we started.

What the data says:

The report's overall median days on market was 28 in Q2 2026, down from 32 the previous quarter. Single-family ran 25 days. Multifamily ran 31.

28 days is 7.7% vacancy. 31 days is 8.5%.

Why this matters:

If your model says 5% and you're buying scattered-site multifamily, you're not underwriting aggressively.

You're underwriting a year where nothing goes wrong and you nail the price on the first try.

The data says about a third of us don't manage that.

What we'd suggest:

Use the national median for your asset type as the floor, then stack a turnover allowance on top for the years you actually turn a unit.

It's a less exciting number. It's also one you won't have to explain to your lender later.

What's Your Number?

We'd genuinely like to hear how other investors handle this one.

Do you set your vacancy assumption off a percentage rule, off local days-on-market, or off your own turn history?

And has anyone here tracked their own price reduction rate to see how it stacks up against that 36.9%?

We manage in Metro Detroit, where the seasonal swing is wide enough that our honest answer changes depending on which month the lease ends.

Would love to hear how the math looks in markets that don't get a real winter. 🥶