Investor · Cleveland Ohio · Member since 2023 · 333 posts · 66 votes
After analyzing thousands of properties I have come to realize that having properties and making them fit numbers is better than having numbers and making them fit properties
Anthony, I land on the other side of that — numbers first, property second, every time. My first screen takes about a minute: does the rent clear the 1% rule against the purchase price? In Memphis I buy single-family under $100K and need Section 8 rent of $1,395-$1,950 on a 3BR+ to even keep looking. Second screen is the refinance: will the ARV land in my $180K-$265K range so a DSCR loan at 70-80% LTV actually recycles my capital? If either screen fails, I pass and move on. At 236 properties, discipline on the front end is what keeps the portfolio boring in a good way.
Anthony, I land on the other side of that — numbers first, property second, every time. My first screen takes about a minute: does the rent clear the 1% rule against the purchase price? In Memphis I buy single-family under $100K and need Section 8 rent of $1,395-$1,950 on a 3BR+ to even keep looking. Second screen is the refinance: will the ARV land in my $180K-$265K range so a DSCR loan at 70-80% LTV actually recycles my capital? If either screen fails, I pass and move on. At 236 properties, discipline on the front end is what keeps the portfolio boring in a good way.
I'd push back gently on the idea of finding properties first and making the numbers fit — that's exactly how investors talk themselves into bad deals. The numbers are the discipline that protects you. That said, there are more lenses than the basic calculator: run every deal through multiple exit strategies (what does it need to look like as a flip vs. a BRRRR vs. a straight rental?), do sensitivity analysis (what happens if rents come in 10% lower or rehab runs 20% over?), and factor in your time — a deal that ties up your capital for 9 months isn't the same as one that frees it in 90 days. More analysis is good. Just don't let creativity with the math replace the math.
You're right that there's more than one way. The single-number approaches (1% rule, 70% rule, cap rate) are fast filters, but they only tell you whether to keep looking. Here's the stack I run before a deal goes anywhere near an offer:
1. Price from at least three angles, not one.
Comps: true sold comps within the last 6 months, same bed/bath and condition, not active listings.
Income: what the rent supports. NOI ÷ your target cap rate gives a value you can sanity-check against the comps.
Cost: purchase + rehab + carry + exit costs vs. ARV. If you're flipping, the 70% rule is a starting point, not the answer.
If the three numbers disagree by more than 10–15%, that gap is your analysis. Figure out why before you go further.
2. Underwrite the rent the lender will use, not the rent you hope for. If you're financing with DSCR, the appraiser's rent schedule sets your coverage, not the Zillow estimate. Work backward to the minimum rent the deal needs to cover PITIA at 1.0–1.2x, then ask whether the market supports it.
3. Stress test it. Run it again with rent 10% lower, the rate 1% higher, rehab 20% over and two extra months of carry. If it only works on the best case, it doesn't work.
4. Look for the deal-killers early. Flood zone, insurance quotes (these have wrecked a lot of 2025–26 deals), tax reassessment after sale, HOA rental restrictions, title issues. Most deals die on one of these, not on the math.
5. Get a second set of eyes on the numbers. Disclosure: I run a lender, CR Equity AI. We built an AI valuation tool, AIVAA, because we had the same frustration on our own deal flow. It pulls value, rent and risk signals from multiple data sources in a few minutes, so you can see where your numbers and the data disagree before you pay for an appraisal. It's not an appraisal and doesn't replace one; I use it as a fast cross-check. But whatever you use, the point is to have something challenge your assumptions before the appraiser does.
Happy to run through a specific deal here if you want to post the numbers.