My yield column said 10.4 percent for three years. XIRR on the whole sleeve said 3.9.

How did you separate the principal after the fact? My statements say "distribution" for everything too, and I've been assuming I'd sort it out when the tax documents show up. Was there anything in the platform records that actually told you which dollars were return of capital, or did you have to go deal by deal?

The idle cash part is the bit I never see anyone count. Nine months of nothing is a real number and it doesn't appear anywhere on a deal page.

So the money came back and you counted it twice, once as a return and once as capital you still had. That's a clean way to fool yourself.

Question from the outside of this: what is the platform dashboard number actually measuring, then? Mine shows an IRR and I'd assumed it was my IRR. Sounds like it's the return on capital while capital was invested, which is a different question from what happened to my money.

Same error, different shape, on my side. I used to quote profit per flip and never divided by the months the money was tied up or the weeks between closings. Two deals a year at 22k each is not a 40 percent business when the cash sat still for five months of it. @canton the difference is you can at least see your dates.

The drag from cash between positions is worse in a small sleeve because one repayment is a big share of it. At 96k across 19 checks, a single 15k repayment sitting for a quarter moves the blended number a lot. In a bigger book the maturities overlap enough that you're rarely fully out.

Can you write out the actual layout you ended up with? Date, amount, sign, and what did you do with the still-live positions, mark them at the platform's stated value or leave them out?

On the management side we send owner statements that separate operating distributions from refinance proceeds because owners scream if we don't. Retail platform statements often don't bother, and the sponsor's accountant is the one who decides the label. Worth asking a platform up front how they report a return of capital, before you're in.

Refi proceeds aren't a return either way. That's borrowed money handed to you while the risk stays on the property, and you still own the equity that now has more debt in front of it. I've watched that exact move on renovation projects. The check feels great in month 19 and the problem shows up at the maturity.

@plumb that's my understanding too, the platform is computing on invested capital and it has no idea what your cash did before or after. Which makes their number fine for comparing two deals and useless for measuring a sleeve.

Went through this a second time. It never occurred to me that getting money back early could look like a good year.

I'll push back some. The measurement was wrong, granted, but a correct spreadsheet on the same 19 positions still gets you 3.9. The reason the sleeve underperformed is what was in it and when you bought it, and fixing XIRR doesn't change either. @canton what did the ten equity deals actually do? If six of them stopped distributing, that's the story and the accounting is just how long it took you to notice.

Related question I've been stuck on for months. With five to seven year lockups, any mid-life return figure depends heavily on what you carry the live positions at, and the only mark available is the sponsor's own. How much weight do you put on that at year two, when nothing has been sold and nothing has to be proved?

Sponsor marks at year two are a hope with a decimal point on it.

What is XIRR? I know IRR loosely as the annual return accounting for timing, but I don't know what the X adds.

@juniper the X is for irregular dates. Plain IRR assumes even periods, XIRR takes a date next to every amount, which is what you need when wires land whenever they land.

@canton the fix I'd add is to stop leaning on IRR at all for a sleeve this young. IRR is sensitive to early cash and it flatters a refi in month 19 exactly the way your yield column did. Run DPI alongside it, total cash actually distributed to you divided by total cash you put in. Yours is somewhere well under 1.0 by the sound of it, and that single number is hard to argue with. Then carry the live positions separately at whatever the sponsor says and label it as the sponsor's figure, so the estimated portion of your return never blends into the realized portion.

On @keystone's point, I agree with both halves and they compound. Bad selection loses money. Bad measurement means you keep buying the thing that's losing it. The accounting error is the one that turned two disappointing vintages into three years of the same allocation.

The idle cash question is the underrated part of the post. If your plan requires you to be fully deployed to hit your target, and the shelf only has offerings you'd pass on for four months of the year, then your target was never achievable at that check size. Size the target around a realistic deployment rate.