Nine LP positions over seven years. The losers taught me more.

The 93 percent occupancy with a loss is the line I'd put in front of anyone who thinks LP diligence is about the property. You underwrote a building and got handed a loan outcome.

On the construction side I've watched the same thing from inside. A value-add plan that assumed 18 months of renovation and got 26 months because of a switchgear lead time doesn't just cost the extra carry. It pushes the stabilized refinance past the loan maturity, and then the whole capital structure is the problem. When you review a business plan now, do you ask about the renovation schedule's critical path, or just the total budget and contingency?

What did the sponsor say when they decided to sell at a loss rather than extend the cap? Curious whether you got a real explanation or a letter.

The thing I'd push on is your same-check-size rule. It's disciplined but it means your position sizing never reflects conviction, and after seven years you presumably know which of your six sponsors is best. If sponsor A has delivered twice and communicated well through a bad quarter and sponsor F is on their third deal, is a flat $75k really the right answer for both?

I'd also question whether two open positions per sponsor is a real constraint or a comfortable one. Correlation between two deals from the same GP is high, sure, but correlation between nine multifamily value-add deals in the sunbelt is also high no matter how many sponsors are on the cap tables. Have you got anything that isn't multifamily?

Question I would have been embarrassed to ask: when a deal returns 70 percent of capital, what actually happens to the 30 percent? Is it just gone at exit, or do you get some kind of loss you can use? I know the tax part needs a CPA, I'm asking about the mechanics of what you receive.

@compass you get a final distribution and a final K-1, and the shortfall is gone as an economic matter. What you can do with it on your return is entirely a CPA conversation, because passive loss rules depend on your whole situation.

@orchard, on your deployment timing question, I've been stuck in analysis for a year so take this with that in mind, but I don't think quarterly spacing is the right frame. You can't manufacture good deals on a calendar. What you can do is set a ceiling on how much you'll deploy in any twelve month window, so a year when everything looks great can't absorb three years of allocation. That's what would have saved you in 2021, and it's a rule about pace rather than a rule about dates.

The other thing in your list I'd underline is that the same sponsors appear in both your good and bad vintages. Sponsor quality is supposed to be the dominant variable and your own data says entry basis competed with it hard.