Estimating vacancy, repairs, and capex
Most Popular Reply
Member since 2026 · 31 posts · 13 votes
1dI'd stop thinking in percentages for capex and build it from the systems instead: list the roof, HVAC, water heater, windows and so on, note each one's age, and divide its replacement cost by the years it has left. That gives you a monthly number that reflects the actual house, which is why a 90-year-old property can need far more than 5% while a fresh rehab needs less. For repairs, ask the seller for the last couple of years of repair invoices or their Schedule E if they'll share it, since that's real history instead of a guess. For vacancy, look at how long comparable rentals sit on the local listing sites and add the turn cost (cleaning, paint, utilities while it's empty). Then run the deal at a pessimistic case too, and only buy if it still works there.
11 Replies
- Realtor · Memphis, Shelby County · Member since 2026 · 150 posts · 48 votes1d
You are right to question a flat 5% or 3% for every property, because age, condition, and property type change the risk in ways a single percentage cannot capture. I find it clearer to separate the three buckets instead of blending them into one allowance.
Vacancy is about time between tenants and turnover costs, so I look at how long similar homes take to lease and what a turn usually costs. Repairs are the smaller, recurring items that keep a home rentable. Capital expenses are different: they are the big, infrequent items such as a roof, HVAC system, or water heater, and those are better estimated from age and remaining life than from rent alone.
That is why a single-family home and a duplex of the same size can need different reserves. A roof may be one shared cost, while other systems can multiply. For an older home, I would itemize the major systems, test the deal with a higher reserve, and only move forward if it still works. If a deal only works at the most optimistic allowance, I treat that as a warning, not a rounding choice.
- Real Estate Agent · Chicago, IL · Member since 2017 · 2k+ posts · 2k+ votes1d
Completely disregard anything the list agent sends you for numbers lol. I run through each deal with my clients deal by deal. It won't be a set percent but rather fixed numbers per property type, as the figures will be the same for a $1 million 4-unit or a $500k 4-unit. For example, in my market for a 4 unit in decent shape estimate $500 a month for capex reserves per month. For a 2 unit $300 or so a month, if the building runs on an old boiler heat I would avoid the deal or bump this up higher and also bump up water bill expectations higher. Vacancy depends on area, on the northside of chicago it runs 2 to 3% with close to 0 defaults with proper screening while in some lower income or working class areas of city it runs 8 to 10% and you should expect the occasional default. Find a good local investor realtor and they can help make these figures specific to your area.
- Member since 2026 · 31 posts · 13 votes1d
I'd stop thinking in percentages for capex and build it from the systems instead: list the roof, HVAC, water heater, windows and so on, note each one's age, and divide its replacement cost by the years it has left. That gives you a monthly number that reflects the actual house, which is why a 90-year-old property can need far more than 5% while a fresh rehab needs less. For repairs, ask the seller for the last couple of years of repair invoices or their Schedule E if they'll share it, since that's real history instead of a guess. For vacancy, look at how long comparable rentals sit on the local listing sites and add the turn cost (cleaning, paint, utilities while it's empty). Then run the deal at a pessimistic case too, and only buy if it still works there.
- Property Manager · Denver, CO · Member since 2026 · 9 posts · 6 votes1d
+1 to building capex from the systems. On vacancy I'd do the math off how often it turns, not just how long it sits. Vacancy % is roughly days empty per turnover divided by how long the average tenant stays.
So if a turn takes 30 days and tenants stay about 2.5 years (~900 days) thats around 3%. Same 30 days with tenants who leave every year is closer to 8%. Single family tends to hold tenants longer than small multifamily in a lot of markets, which is another reason one % doesn't fit both.
For the days empty part, count the clean/paint time before it's even listed plus the days on market. I'd also ask the seller for the current tenants' move-in dates, it tells you a lot about turnover in that building.
- Staten Island, NY · Member since 2013 · 116 posts · 52 votes1d
I'm a realtor on Staten Island and I run numbers on small multifamily a lot, so here's what I actually use. For a 2 to 4 unit in decent shape I'd put capex around 250 to 500 a month total depending on the age of the roof, boiler and plumbing, instead of a percent of rent. Vacancy I set by area, about 3 to 5% where tenants stay put and 8% or more where turnover is high. Repairs I put at roughly 5 to 8% of rent, and I bump it up if I see old galvanized pipes or knob and tube wiring when I walk it. I'd also stop trusting whatever the listing agent puts on the flyer and rebuild it with your own numbers. If it only works at 3% across the board, it's not really a deal.
- 1dDrew SygitBusiness MemberProperty Manager · Royal Oak, MI · Member since 2012 · 13k+ posts · 9k+ votes
Copy & paste below:
------------------------------------------------------------------
Key metrics for each Property Class:
Class A Properties:
Tenant Pool: Majority of FICO scores 680+, no convictions/evictions in last 7 years.
Tenant Default: 0-5% probability of eviction or early lease termination.
Section 8: Class A rents are too high and won’t be approved.
Vacancies: 5-10%, depending on market conditions.
Cashflow vs Appreciation: Typically, 3-5 years for positive cashflow, but you get highest relative rent & value appreciation.Class B Properties:
Tenant Pool: Majority of FICO scores 620-680, some blemishes, no convictions/evictions in last 5 years.
Tenant Default: 5-10% probability of eviction or early lease termination.
Vacancies: 10-15%, depending on market conditions.
Cashflow vs Appreciation: Typically, 1-3 years for positive cashflow, balanced amounts of relative rent & value appreciation.
Section 8: Class B rents are usually too high for the Section 8 program.Class C Properties:
Tenant Pool: Majority of FICO scores 560-620, many blemishes, but should have no convictions/evictions in last 3 years. Verifying recent 2-years of rental history very important! Same for 2-years of job/income stability.
Tenant Default: 10-20% probability of eviction or early lease termination.
Section 8: Class C rents usually meet program requirements, proper screening still recommended.
Vacancies: 10-20%, depending on market conditions and tenant screening.
Cashflow vs Appreciation: Should cashflow immediately, at the lower end of relative rent & value appreciation.
Class D Properties:
Tenant Pool: Majority of FICO scores under 560, little to no good tradelines, lots of collections & chargeoffs, but should have no convictions/evictions in last 12 months. Verifying last 2-years of rental history and income/employment extremely important to find the “best of the worst”.
Tenant Default: 20-30% probability of eviction or early lease termination.
Section 8: Class D rents meet program requirements, often challenges to pass Section 8 inspection.
Vacancies: 20%+, depending on market conditions and tenant screening.
Cashflow vs Appreciation: Typically, all cashflow with little, maybe even negative, relative rent & value appreciation.Where did we get our FICO credit score information from?
Check out this chart:
FICO Score
Pct of Population
Default Probability
800 or more
13.00%
1.00%
750-799
27.00%
1.00%
700-749
18.00%
4.40%
650-699
15.00%
8.90%
600-649
12.00%
15.80%
550-599
8.00%
22.50%
500-549
5.00%
28.40%
Less than 499
2.00%
41.00%
Make sure you understand the Class of properties you are looking at and the corresponding results to expect.
For example, Metro Detroit has 132 cities and the City of Detroit 183 Neighborhoods, which we’re analyzing and classifying to make better investing decisions.
Logical Property Management4.9465 Reviews - 19hMatthew MorrowBusiness MemberInvestor · PA - NY - NJ · Member since 2019 · 823 posts · 284 votes
@Elliot W. a lot of good answers here already. Building capex from the actual systems like Clever and Jaimee said is close to how I think about it. The only thing I'd add is on older buildings. A lot of what we own is 100+ years old, and I learned the hard way to assume the last owner didn't take care of it the way I would. So I wouldn't trust 3% on a 75-100 year old house, and I'd want reserves for the big items on top of the monthly number. On vacancy, one empty month on a small rental can take most of a year's cash flow, so I'd rather be a little pessimistic there. Hope some of that helps.
CORE Team | Realty ONE Group Supreme548 Reviews - 18hNicholas L.Pro MemberFlipper/Rehabber · Pittsburgh · Member since 2018 · 6k+ posts · 5k+ votes
@Elliot W. i only BRRRR so i estimate the capex by estimating the actual capex and then fixing it. working on a property now that needs a roof for example. so i am getting: bids for a roof.
then, ideally, capex is low for 5-10 years.
percentages are worthless.
- Chicago, Cook County · Member since 2026 · 8 posts · 2 votes15h
Elliot, you're right to distrust 3% on a 90-year-old house. The replies above on building capex from the systems are the right method, so I'll add what I look for on older buildings that the system list tends to miss:
- Water first. Check the basement for stains, white crusty residue on the foundation, or fresh paint on only the bottom of the walls, then the gutters, downspouts, and grading. Water is behind a lot of the surprise costs in old houses.
- What the plumbing actually is. Galvanized supply lines and cast iron drains can look fine and still be near the end. A sewer scope on an old house is cheap compared with a broken main.
- What's behind the panel. A newer panel can still feed old wiring. Look for ungrounded outlets, cloth wire in the basement, or a fuse box left in place.
- Prior work. Old houses have usually had several rounds of owner fixes. Uneven floors, sistered joists with no plan, or patched plaster all deserve a second look.
Then run the deal at a reserve you'd actually be comfortable with on that house. If it only works at 3%, it doesn't really work.
Kyle
- Investor · miami · Member since 2026 · 5 posts · 2 votes11h
You're right to distrust the flat 5% — it's a lazy shortcut. I'd treat them as three separate questions tied to the actual house. Vacancy: use real local rent-up times, not a national number. Repairs: scale to the age and condition — a 1920s house and a 2015 build are not both 5%. Big replacements (roof, HVAC, water heater): take what each one costs to replace, divide by the years of life it has left, and set that aside every month. And yes — those costs land differently on a single-family vs a duplex, so I'd build them up item by item instead of trusting one blended percentage to magically "average out." The percentages are fine for a quick 30-second screen. For a house you're actually buying, price the real pieces.
Join the conversationCreate a free account to reply, vote on answers and follow this thread.