I screened close to 50 multifamily deals this spring. Most got a no in the first ten minutes. A handful made it to a full underwrite. Two are in the LOI stage as I'm writing this.

The point isn't the count. The point is the pattern. When you look at 50 deals back-to-back, you stop seeing 50 properties and you start seeing the market. And the market right now is telling us four pretty specific things.

1. Sellers are finally starting to move

For most of 2024 and into 2025, sellers were anchored to 2022 numbers. Their broker walked in and told them, "Your building was worth $4.2 million two years ago, so let's list it at $3.9 million and call it a deal." And it sat. And it sat. And then it sat some more.

That's breaking. About a third of the deals I looked at this spring were second or third price drops. Some were re-listings under new brokers. A few had the broker actually email me with, "The seller is open to creative structures." Translation: I've been holding this too long, please make me an offer.

I'm not saying every seller is reasonable. Plenty are still living in 2022. But the share of motivated sellers is the highest I've seen since 2010. If you've been waiting for the market to crack, this isn't a crack — but it's the first real bend.

2. The pro forma still lies. The rent roll still tells the truth.

Half the deals I killed in the first ten minutes had a beautiful pro forma. Year-three NOI projected at 14% higher than today. Vacancy modeled at 3%. A bullet point about "value-add upside through unit renovation."

Then I'd open the actual rent roll. Six units on month-to-month. Two paying $200 below the others for the same floor plan because grandma's been there since 2009. One unit "occupied" by the owner's cousin who pays $400. Real vacancy: 11%.

Nothing about this is new. But in a flat-ish rent environment, the gap between the pro forma and the rent roll is the whole deal. The buyers losing money this year are the ones who underwrote the pro forma. The buyers making money are the ones who underwrote the rent roll and treated everything else as a maybe.

Old rule. Still works.

3. Insurance and taxes are eating returns nobody is talking about

This is the one that almost nobody on Twitter wants to deal with, and it's killing more deals than interest rates.

On every single deal I looked at in the Southeast, insurance was up 30 to 60% from the seller's last renewal. Some Florida buildings could not get coverage at all from standard carriers. I saw one 24-unit in the Carolinas where insurance alone wiped out 18% of NOI compared to the seller's trailing 12.

Tax reassessments are the second hit. Sell a property and watch the new assessed value come in based on your purchase price — not the seller's 2018 basis. In a few states this is a brutal one-two punch, and your underwriting better account for it.

My rule for 2026 underwriting: actually quote insurance before you sign the LOI. Actually pull the assessor's reassessment history for the county. If you're modeling these as small line items, you don't have a deal — you have a hope.

4. The deals that work look boring

I'll tell you which deals made my LOI pile. Not the trophy mid-rise in a glamour Sun Belt market. Not the value-add play in a "fast-growing" tertiary. Not the brand-new build that someone overleveraged and now needs out of.

The two that survived: a 19-unit in a Midwest workforce market with 96% historical occupancy, boring tenant base, owner who's selling because he's 71 and tired. And a 28-unit in a Northeast secondary, 1970s build, copper plumbing, ugly carpet, run by an out-of-state owner who hasn't raised rents in four years.

Neither is sexy. Neither will get me invited on a podcast. Both pencil from day one with conservative assumptions. Both have real upside that doesn't depend on the market doing me a favor.

That's the playbook in 2026. Boring buildings, boring markets, real numbers, real cash flow. Let other people chase the headlines.

What I'm doing about it

I'm screening more deals than ever — which is why I have 50 to talk about in a month. Most of that volume is possible because the screening itself is mostly automated now. I built a workflow that pulls a deal in, runs the math, flags the deal-killers, and gives me a yes/no recommendation in under three minutes per property. If the property survives that filter, then I spend real time on it.

I wrote about that workflow here if you want the breakdown. Same idea, scaled up.

The takeaway for anyone buying in 2026: the deals are out there. They are not where the headlines tell you they are. Look in markets nobody is talking about. Underwrite the rent roll, not the dream. Get the insurance quote before you fall in love. And when you find a boring deal that pencils, move on it — because that's the whole game right now.

Go find one.